Business Research Methods

Introduction: Human nature is always curious to learn, understand or investigate about the phenomenon by raising the questions like why, how, what, when etc. Human knowledge possesses of beliefs or judgement about a particular phenomenon. Some beliefs may be supported by evidences based on perception and experiences which are called justified beliefs. But every belief can not, however, be equated with knowledge because some of our beliefs, even the true ones, may turn out to be false on verification. Justified beliefs consist of two types of judgement viz. direct and intuitive judgements, derived judgement. Direct judgements are based on our sense of perceptions. The derived judgements are formed out of the existing body of knowledge and it is gathered through inference. A single judgement may lead to a number of derived judgements. Knowledge may be acquired from different sources like experience, human beings, nature, books etc. It includes what we know about matters of facts as well as principles and processes of inference. It includes three elements:

·   A system of ideas
·   The ideas correspond to things actually existing, and
·   Belief in such correspondence.

Knowledge covers an area where explanation is either not explicitly relevant or just not possible. However, self-evidence is an essential condition for knowledge, but it is not a sufficient condition, because self-evidence is a matter of degree. Knowing depends more on ability rather than on possession. The origin of is empiricism and experience. The most important factor in the growth of knowledge is the inherent human curiosity itself. Knowing entails explanation which is an essential feature of a science. Explanation is implicit in the universe of formal relations and matters of facts. Explanation is the relationship between different parts or facts of the branches concerned. The relationships between the facts, which are used for systematic explanation, are called theories or models. The objective of every model is the simplification of reality, which is essentially complex, is perplexing. To generate new principles and theories or find the solution over existing problems is a part of research.

Meaning and definition of research

 The word research is composed by ‘re’ and ‘search’ where re means repeatedly or again and again, and search means to investigate or find. Thus to search again and again is research. Generally, research is an effort to search new fact, knowledge, and principle in scientific ways.

Research is essentially a systematic inquiry seeking facts through objectives verifiable methods in order to discover the relationship among them and to deduce from them broad principles or laws. It is really a method of critical thinking by defining and redefining problems, formulating hypothesis or suggested solution, collecting, organizing, and evaluating data, making deductions and making conclusions to determine whether they fit the formulated hypothesis. Most of the researchers defined research in their own way at different time and to sum up it is defined as ‘ the process of a systematic and in-depth study or search of any particular topic, subject or area of investigation backed by the collection, compilation, presentation of relevant details or data. In strict sense a research
·    Is directed towards the solution of the problem.
·    Is based upon observable experience or empirical evidence.
·     Demands accurate observation and description.
·     Involves the quest for answers to unsolved problems.

Objectives of research

The main objectives of research are:
·  To achieve new insights into a phenomenon.
·  To highlight the characteristics of a particular individual, situation or group.
·  To test a hypothesis of a causal relationship between independent and dependent variables.
·  To determine how often something occurs or its association with something else.

Types of research:

Generally, a research can be undertaken to solve the currently existing problem and the other is to generate new knowledge in particular area. These are respectively termed applied and theoretical research. Basically a research is either theoretical or applied in nature. However, it may further categorized under various types on the basis of nature, scope and type of research.

Applied vs. Fundamental:   Applied research also called action research, is associated with a specific project or problem. It is concerned to current activity or immediate practical situations. It aims at finding out a solution for an immediate problem faced by a society. It discovered what, how and why questions about policies, programs and procedures of actual life. This type of research can be conducted at any time whenever required to solve over the existing problems.

Fundamental research also called basic or theoretical research is a search for broad principles and laws in selected area of research. It is not concerned with solving any practical problems and policies but with designing and fashioning tools of analysis and with discovering universal laws and theories. It is followed by researcher for the systematic development over the area of their specialization and cannot be conducted immediately whenever required. This type of research takes the shape of an improvement in the existing theory by relaxing some of its assumptions or by re-interpreting a new theory with the existing one as its basis.

 Descriptive vs. Analytical: Descriptive research includes surveys and fact-finding inquiries of different kinds. The main purpose of descriptive research is the description of the states of affairs as it exists at present. It is also termed as ex-post facto research in social science and business research studies. This method assumes that the researcher has no control over the variables and the researcher just report what has happened or what is happening. This method is widely used for frequency of shopping, preferences of people or similar data of survey method of all kinds including comparative and correlation methods. On the other hand, in analytical research, the researcher has to use the facts or information already available and analyze these to make a critical evaluation of the material.

Quantitative vs. Qualitative: Quantitative research is applicable to the phenomena that can be expressed in terms of quantity or amount. In other words, quantitative research is based on the measurement of quality like income, expenditure, import-export prediction, population by age etc. in quantity. The results of quantitative research are purely numeric. On the other hand, qualitative research is concerned with qualitative phenomena relating to quality or kind like attitude or opinion of people to particular subject, law, decision, policy etc. this type of research aims at discovering the underlying motives and desires, using in depth interviews for the purpose. Also it is used in word association tests, sentence or story completion tests and similar other projective techniques.

Conceptual vs. Empirical: A conceptual research is related to some abstract idea or theory. It is generally used by philosophers and thinkers to develop new concepts or to reinterpret existing ones. Generally, this type of research is adopted by the researchers in their area of specialization to express the views. Whereas empirical research relies on experiences or observation alone, often due regard for system and theory. It is data based research, coming up with conclusions that can be verified by observation or experiment. It is also called experimental type research. 

In addition to above mentioned methods, there are various types of research based on either the purpose of research or the time required to accomplish research or the environment in which the research is done or on the basis of other similar factors.

Scientific Research

Scientific Research is a systematic and objective attempt to provide answers to certain questions. The purpose of scientific research is to discover and develop an organized body of knowledge. Therefore, scientific research may be defined as the systematic and empirical analysis and recording of controlled observations, which may lead to the development of theories, concepts, generalizations and principles, resulting in prediction and control of these activities that may have some cause – effect relationship. In the words of F N Kerlinger "Scientific research is systematic, controlled, empirical and critical investigation of hypothetical propositions about the presumed relations among nature phenomena."

The main features of scientific research are:

1.  Purposiveness : Each scientific research or investigation contains a focus or a specific purpose. The purpose of research influence and directs all the activities of the researcher. It forms the basis of research procedure by reducing the major errors to enhance better result.

2.  Testability: The major aspect of scientific research is to test the validity of findings by developing hypothesis. Research studies without using hypothesis testing process do not meet the criteria of scientific research. Use of different statistical tools, the proper test statistics is chosen to verify the findings of research process. Thus, every scientific research is attached with the application of hypothesis testing.

3.  Replicability: Since no two research results even conducted by different researcher in same title by selecting same procedure at same time are similar in business and social research. But the scientific research assumes to some extent of consistency or similarity by repeating each stages of research work (especially in data collection) and selecting common methodology.

4. Objectivity: Objectivity is also an important characteristic of scientific research. The more objective the analysis and interpretation of data, the more scientific the research investigation becomes. To maintain objectivity in the research work, the researcher need to start with some initial subjective values and beliefs by avoiding the biasedness in the sample selection procedure, and statistical tools for analysis.

5. Rigour: Scientific research is also considered a rigorous process which involves a carefully thought-out methodology and good theory. These factors enable the researcher to collect the right kind of information from an appropriate sample with minimum amount of bias and facilitate appropriate method of data collection and analysis. Lack of rigour in research may lead to selection of faculty research design, inappropriate or biased collection and interpretation of data and wrong conclusions.

6.  Generalizability: Every research result is not likely to be generalizable but the result of applying scientific research, the results are generalizable to the whole universe provided the facts or information are collected by selecting appropriate sample design. The more generalizable the research the greater is its utility and value for future information.

 Scientific Research Process

Most of the researchers suggested that the research task must be presented in sequential process involving several clearly defined steps. But no one claims that research requires completion of each step before going to the next. Recycling, circumventing and skipping some steps begin out of sequence, some are carried out simultaneously and some may be omitted. Despite these variations, the idea of a sequence is useful for developing a project and for keeping it orderly as it unfolds. Generally, the research process involves management question and its origin, selection, statement, exploration and refinement in the study. Albert Einstein explained it “the formulation of a problem is far more often essential than its solution, which may be merely a matter of mathematical or experimental skill. To raise new questions, new possibilities, to regard old problems from a new angle require creative imagination and marks real advance in science”.

In basic or applied research, a thorough understanding of the management question is fundamental to success in the research enterprise. A useful way to approach the research process is to state the basic problems that prompt the research and then try to develop other questions bybreaking down the original questions into more specific form. Identification of managemnt problems is not much difficult (unless the organization fails to track its performance factors like sales, profits, employees turn over, manufacturing outputs and defects, on time deliveries, customer satisfaction etc). Choosing the problem incorrectly will direct valuable resources (time, manpower, budget and equipment) on a path that may not provide critical decision making information. The management questions can be categorized under various forms and are chosen on the basis of:

- Choice of purposes or objectives

- Generation and evaluation of solutions

- Trouble shooting or control situations

The first form may arise the questions in terms of “what do we want to achieve?” Similarly the second type rely the questions like “how to achieve?” and the third one stress upon remedial or controlling measures.

Once the researcher has a clear statement of the management questions then s/he must convert it into research question. A research question is the hypothesis of choice that best states the objective. It is fact oriented and answerable to gather information. After a preliminary review of literature, a brief exploratory study or both the project begins to examine the concepts and constructs to be used in the study. Also the investigative questions are included in the research proposal to guide the development of research design which is the foundation of data collection instrument. An exploration begins with a search of published data. Applying the exploration, when the facts are collected, it is analyzed and verified by using suitable hypothesis testing procedure. These all stages are common in business management research activities. However, a scientific research (it does not necessarily mean ‘science’ but refers a manner of doing research) assumes the following steps which are mostly accepted in all sphere of human being.

Sensing or realizing: This is the first and foremost step of scientific research process in which the researcher senses about the problem and development emerging in the environment. Actually the researcher is not familiar about the problems but senses them in terms of what, why, and how is happening i.e. the researcher form an idea about the situation or field in which the research work is to be performed.

Problem identification: After well awareness about the environment, the researcher focuses on the problem and associated factors through further search of information. Generally, there may exist various problems under the selected environment and the researcher try to select the vital problems upon which the research work can be completed.

Theoretical framework: A theoretical framework attempts to integrate the information logically so that the reason for the problem can be conceptualized. In this step, the critical variables are examined and the association among them is identified. Theoretical framework also focuses about the depth of subject matter under study.

Hypothesis formulation: Hypothesis is a conjecture or proposition tested for the validity. Hypothesis formulation for the study is determined on the basis of theoretical framework. Sometimes prior assumptions about certain factor or statement are necessary and this type of statement is tested by using suitable statistical tests.

Research design: The research design is the strategy for conducting research work, which describes the general framework for collection, analysis and evaluation of identified data. The design provides the basis which the researcher wants to know and what has to be dealt with in order to obtain required information i.e. methodology employed for the collection, analysis and evaluation of facts/data.

Collection of data: Data collection also called fieldwork is the implementation of research design. In this method, the researcher adopts the method and instrument like observation, questionnaire, schedule, interview etc. to gather data.

Data analysis: Data analysis is the statistical and analysis of collected, edited, coded and tabulated data. It is especially important in cases where the researcher has amassed large amount of information from many respondents. The validity of collected data is also tested by using statistical tests.

Interpretation and Generalization: This is the final step of every research process in which the findings are interpreted .For the generalization of applied research, specific implementation strategy is proposed to solve the problems identified by the study.

Business Resource

The advances in business management in recent years are the results of adopting research findings in the behavioral sciences to business situations. Research being fact finding process, profoundly influences business decisions. The study of research methods provides the knowledge skill needed to solve the problems and meet the challenges of a fast-paced decision making environment. Research in business administration is also a systematic inquiry whose objective is to provide information to solve managerial problems. The maker wants to simplest model that predicts outcomes reasonably well and is consistent with effective decision-making. Increased need for more and better information, improvement of tools and techniques to meet the challenges at current period and up-to-date information about the discipline are the three factors applied in business administration.

Any managerial analysis should take into account both the internal and external systems. Knowledge of internal system requires an understanding of the physical, human, financial and technological make up of organizational system. It includes all the functional areas that operate as sub-system of the organizational system. An analysis of the external system enables the decision maker or managers to evaluate consumer behavior. The researcher should analyze the decision centers as well as the profit and information centers in the firm to identify the operating conditions of the functional areas. The functional areas are the separate areas of the operational activity of an organizational system. These are the finance, production, general administration (or personnel) and marketing functions. There is very close relationship between these four areas so as to check and control each other. Another field for a research relevant to business administration is the social and political environment in which the business activity operates. This includes trade practices, labour relations, security, explosive technology growth and continuing innovations in global communications. These changes have created need of new knowledge for the manager and new policies to consider when evaluating these decisions.

Research in financial management includes operations of specialized financial institutions as well as the financial side of regular business firm like: capital and investment analysis, insurance, international finance, cash and receivable management. The research in financial management is usually concerned with:

· Determination of financial needs of the enterprise

· Raising funds at minimum cost

· Making optimum allocation of funds in specific aspects

· Development of tools of control for evaluating the financial performance of the enterprise and

· Generation of financial data for decision-making and their use in future.

Production is a process of converting material and human inputs in terms of goods and services. The information from the production system may include details of raw material, inventory, finished goods inventories, work force skill, and turn over, accidents, absenteeism, shifts, strike of employee’s etc. The production or manufacturing research focuses on technical aspects and is conducted to:

· Identify new and better ways of producing goods

· Invent modern technologies

· Reduce cost and improve product quality.

The administrative management has wider scope for research than any other functional areas because human behavior is prone to constant change. The aim of administrative research is the development of policies and programs to motivate the employees to work willingly. It includes:

· Organizational planning and development

· Age and salary administration

· Training and development programs for the employees

· Staffing, motivation and recording of employees.

Marketing is the principal revenue generating activity while other is revenue consuming. Marketing research deals with production and distribution of products, marketing policies and practices. It covers the issues like production planning, development, pricing promotion, distribution, selection of market, advertising strategies, new products, channels etc.

There are many other fields like accounting, taxation etc. upon which the research is essential to develop new tools techniques.

Applying scientific thinking to management problems

Researcher or managers apply sound reasoning to find correct premises, testing the connection between their facts and assumptions and claims based on adequate evidence. Like all specialties, research has its own terminology. To communicate the meanings with precision within a community of research users and research practitioners, who demand consistency of definitions for shared meaning, it is necessary to learn the specialized terminology of research. Every research work is based on the question of perspective. Different fields of business research make different assumptions about reality and the nature of data/facts. Due to selective perception and conditioning, researcher analyzes the problems in different manner. Source of knowledge ranges from untested opinion to highly systematic styles of thinking.

Untested opinion: It is concerned to the facts or events like myth, superstitions, hunches etc. about which the feelings of certainty persists. Managers will find little to improve their understanding of reality from untested opinion, even though human nature indicates that they should be prepared to cope with its use by contemporaries when searching for solutions over management problems.

Self-evident truth: Self-evidence truth can be gained or deduced from known laws of nature. For example, the fire burns; death is inevitable etc., which are well known facts. But, every evidence may not be universal. Some propositions that appear reasonable to one person or even to many at a time may not necessarily true. For example, “everyone drives on the left side of the road.” This truth is self-evident only to some of the world’s drivers but not all. This type of thinking is developed through experience and naturally.

Method of authority: Authority is an important source of knowledge and is judged by integrity, the quality of the evidence they present and their willingness to present an open and balanced case. Generally, authority depends upon status or position rather than on true expertise, is often wrong. Great caution is required to accept the views of such sources. Even authorities that do meet the standards of integrity, quality evidence and balance may find their knowledge misapplied. Authority helps to improve the evidence in our knowledge.

Literary style: The literary style of thinking is used in many classic case studies in social sciences. The literary perspective states that a person, a movement or a whole culture is interpreted in terms of the specific purpose and perspective of actors, rather than in terms of the abstracts and general categories of the scientist’s own explanatory scheme. But literary style of thought restricts our ability to generalize from individual case studies to drive the applicable knowledge or truths.

Scientific method: Current scientific methods wed the best aspects of the logic of the rational approach with the observational aspects of the empirical orientation into a cohesive systematic perspective. Application of previously tested and verifiable procedures is essential tenets of science and this method adopts the following steps:

. Direct observation of phenomena

· Clearly defined variables, methods and procedures

· Empirically testable hypothesis

· Statistical rather than linguistic justification of evidences

· Self-correcting process.

Postulation style: Competing styles of thinking influence research directions in business just as they do through out the social and behavioral sciences. The goal of this perspective is to reduce the object of study in mathematical, formal terms. These terms are called postulates, are used to devise theorems that represent logical proofs. The objective is to deduce a mathematical model that may account for any phenomenon having similar form. For example, many firms run computer simulations of their market before a product rollout. There is no single best perspective to view reality like science.

Nature of business research

The executives or decision maker needs specific studies to find the solution over existing problems and opportunity areas. They may need a market survey, an employee survey, a product- preference test, a sales forecast by region, an analysis of financial statements, or an environmental analysis. The economic, political, social, legal, technological, and competitive factors affect the dynamics of business research. These studies, however, require the talents of skilled researcher who can apply principles of research design, sample size, questionnaire construction, statistical analysis, and interpretation of findings. In today’s competitive environment, the research capability has become a critical managerial skill.

The purpose of conducting research in management is to identify business problems or issues and to find out solutions to such problems or issues. Research is also undertaken for many other purposes like model building, strategy formulation, business process reengineering, and so on. In some functional areas of management, research has received special attention.

Research has become indispensable in all sphere of human being. In the field of business management most of the strategies are chosen on the basis of information. Therefore, it is most essential for the decision maker to find out prior information for the implementation of suitable action. This research includes the study in wide variety of subjects like strategy formulation, environmental scanning, organizational structure, staff policies and rules, system of job analysis, employee participation, record management, and marketing strategy and so on. In finance and accounting area, issues such as operation of banks and financial institutions, portfolio management, sources and uses of funds, depreciation, budgetary system, inventory costing, transfer pricing and the like are studied in order to improve the efficiency of the systems. Similarly, operations research has occupied critical significance for business. This research seeks the determination of the optimum (best) course of action of a decision problem under the restriction of limited resources. This research is often associated almost exclusively with the use of mathematical techniques to model and analyze decision problems.

Approaches to Research

Basically there are two approaches to research i.e. quantitative approach and qualitative approach. A quantitative approach involves the generation of data in quantity or number for qualitative analysis in a formal and right form. It is further classified into inferential, experimental and simulation approaches to research. The purpose of inferential approach to research is to form a data base to infer characteristics or relationships of the population. This usually means survey research where a sample of population is studied by asking questions or observation method to determine its characteristics to examine about the whole population.

 Qualitative approach to research is concerned with subjective assessment of attitudes, opinions and behaviour. The research in such a situation is a function of researcher’s insights and judgements. Such an approach in research generates the results either in non-quantitative form or in the form which are not subjected to rigorous quantitative analysis. Generally, the techniques of focus group interviews, projective techniques and depth interviews are used.

Paradigm

Paradigm is a set of assumptions, concepts, values and practices that constitutes ways of viewing reality for the community that shares them, especially in an intellectual discipline.  What this definition portrays that researcher’s assumptions, developed insights through theories constitute ways of viewing knowledge. In other words paradigm help the researcher in identifying ways that the empirical knowledge can be revealed on the  basis of his/her epistemology and the perceive existent truth.Since the 1960s, the concept of a paradigm shift has also been used in numerous non-scientific contexts to describe a profound change in a fundamental model or perception of events, even though Kuhn himself restricted the use of the term to the hard sciences. Creswell describes the following four philosophical assumptions:

• Ontological (The nature of reality): Relates to the nature of reality and its characteristics.  Researchers embrace the idea of multiple realities and report on these multiple realities by exploring multiple forms of evidence from different individuals’ perspectives and experiences.
• Epistemological (How researchers know what they know): Researchers try to get as close as possible to participants being studied.  Subjective evidence is assembled based on individual views from research conducted in the field.
• Axiological (The role of values in research): Researchers make their values known in the study and actively reports their values and biases as well as the value-laden nature of information gathered from the field.
• Methodology (The methods used in the process of research):  inductive, emerging, and shaped by the researcher’s experience in collecting and analyzing the data.

Positivist vs  interpretivist philosophies

This is the land of isms. It is the set of assumptions that define an intellectual understanding of how the world operates and how knowledge is produced. Positivists view that all true knowledge is scientific and can be pursued by scientific method. The interpretivist acknowledge and explores the cultural and historical interpretations of the social world. The philosophical issues for these two are illustrated as follows.

For positivists, the goal of research is to describe what we experience observation and measurement in order to predict and control the forces that surrounds us. Positivists assume that social phenomena can be approached   with scientific method and makes a number of assumptions about the world and nature of research for example, theory of relativity and law of gravity.

 Also for positivists, the social research is purely scientific in nature to set the rules and procedures. It is empirical-involves exploration of those things that can be seen, felt, heard, tasted and smelled as the grounds for all scientific knowledge; and it is often reductionist- involves the study of discrete parts of a system, rather than the system itself or its interconnectivity.

Similarly, the interpretivists focuses more on methods and methodologies, defines sets of procedures that need to be carried out with exacting detail. The methodologies are usually: deductive-researchers test a theory and look for confirmation through observations: hypothesis-driven- researchers propose a tentative statement that they attempt to prove/disprove; reliable- researchers use methods that will give the same results under repeated trials; and reproducible- methodological procedures can be repeated by other scientists who will glean similar findings. They generally want their findings to have broad applicability to the whole of a population. 

Ethics in business research

Ethics are norms or standards of behavior that guide moral choices about human behavior and their relationship with others. The goal of ethics in research is to ensure that no one is harmed or suffers adverse consequences from research activities. Where as unethical activities are pervasive and include violating non disclosure agreements, breaking respondent’s confidentiality, misrepresenting results, deceiving people, invoicing irregularities, avoiding legal liability etc. The appropriate laws and codes, resources for ethical awareness are discussed below.

1.  Ethical treatment (issues) of participants: When ethics are discussed in research design, we often think about protecting rights of the participant, respondent or subject. Whatever may be the method of data collection, the respondent has many rights to be safeguarded. In strict sense, the respondent does not suffer physical harm, discomfort, pain, embarrassment or loss of privacy because the respondents are the main sources of information. To safeguard against these, the researcher should follow the following guidelines: 

Benefits: While making direct contact with respondent, the researcher should explain the benefits of the study carefully. In this case s/he should neither overstate nor understate the benefits. An interviewer should begin an introduction with his or her name, the name of research organization, and a brief description of the purpose and benefit of the research. This puts respondents at ease, lets them know to whom they are speaking, and motivates them to answer questions truthfully.

Deception: Deception occurs when the respondents are told only part of the truth or when it is fully compromised. The need for concealing objectives leads directly to the problem of deception. Some believe this should never occur and other suggests two reasons for deception viz., to prevent biasing the respondents before the survey or experiment and to protect the confidentiality of a third party (e.g., the sponsor). Deception should not be used in an attempt to improve response rates. To minimize the risk in deception when possible, an experiment or interview should be redesigned to reduce reliance on deception.

Informed consent: Securing informed consent from respondents is a matter of fully disclosing the procedures of the proposed survey or other research design before requesting permission to proceed with the study. There are exceptions that argue for a signed consent form. When doing research with medical or psychological ramifications, it is also wise to have a consent form. Debriefing participants: Debriefing involves several activities following the collection of data: Explanation of any deception.

Description of the hypothesis, goal, or purpose of the study.

Post study sharing of results.

Post study follow-up medical or psychological attention.

Rights to privacy: A researcher may dig out information from respondents on private or personal matters such as marital life or religious faith or personal opinions by employing techniques of participant observation, in-depth interviews or disguised projective test. Such practices amount to invasion of privacy.

Anonymity: Finally, there is the ethical issue relating to the obligation of maintaining the anonymity of the research respondents and keeping research data in confidence. Anonymity might be violated through reports and publications.

2.  Ethics and the sponsor: There are also ethical considerations to keep in mind when dealing with the research client or sponsor. Whether undertaking product, market, personnel, financial, or other research, a sponsor has the right to receive ethically conducted research.

Confidentiality: Some sponsors wish to undertake research without revealing themselves. They have a right to several types of confidentiality, including sponsor nondisclosure, purpose nondisclosure, and findings nondisclosure. Companies have a right to dissociate themselves from the sponsorship of a research project.

3. Right to quality research: An important ethical consideration for the researcher and the sponsor is the sponsor’s right to quality research. This right entails:

Providing a research design appropriate for the research question.

Maximizing the sponsor’s value for the resources expended.

Providing data handling and reporting techniques appropriate for the data collected.

From the proposal through the design to data analysis and final reporting, the researcher guides the sponsor on the proper techniques and interpretation.

Sponsor’s ethics: Some times research specialists may be asked by sponsors to participate in unethical behavior. Compliance by the researcher would be a breach of ethical standard. Some examples to be avoided are:

Violating respondent confidentiality.

Changing data or creating false data to meet a desired objective.

Changing data presentations or interpretations.

Interpreting data from a biased perspective.

Omitting sections of data analysis and conclusions.

Making recommendations beyond the scope of the data collected.

These situations may appear if the sponsor offers a promotion, future contracts, or a larger payment for the existing research contract; or the sponsor may threaten to fire the researcher or tarnish the researcher’s reputation and so on. To overcome the situation following techniques are recommended:

Educate the sponsor to the purpose of research.

Explain the researcher’s role in fact finding versus the sponsor’s role in decision-making.

Explain how distorting the truth or breaking faith with respondents leads to future problems.

Failing moral suasion, terminate the relationship with the sponsor.

4 . Researchers and team members: Another ethical responsibility of researchers is their team’s safety as well as their own. In addition, the responsibility for ethical behavior rests with the researcher who, along with assistants, is charged with protecting the anonymity of both the sponsor and the respondent.

5. Safety: It is the researcher’s responsibility to design a project so the safety of all interviewers, surveyors, experimenters, or observers is protected. Several factors may be important to consider in ensuring a researcher’s right to safety. Some urban areas and undeveloped rural areas may be unsafe for research assistants. It is unethical to require staff members to enter an environment where they feel physically threatened. Researchers who are insensitive to these concerns face both research and legal risks- the least of which involves having interviewer’s false information.

Ethical behavior of assistants: Researchers should require ethical compliance from team members just as sponsors expect from the researcher. Assistants are expected to carry out the sampling plan, to interview or observe respondents without bias, and to accurately record all necessary data. Unethical behavior, such as filling in an interview sheet without having asked the respondent the question, cannot be tolerated.

Protection of anonymity: Finally, there is the critical issue relating to the obligation of maintaining the anonymity of the research respondents and keeping research data in confidence. Anonymity might be violated through reports and publications.

Qualitative Research

Qualitative research methods are widely used in social science and management research. A good deal of theoretical and empirical advancement has come from these approaches since nineteenth century. At present there is a resurgence of qualitative methods in basic and applied research in management disciplines with quantitative methods. “Qualitative research is an inquiry process of understanding based on distinct methodological traditions of inquiry that explore a social or human problem” In general terms, scientific research consists of an investigation that:

• Seeks answers to a question

• Systematically uses a predefined set of procedures to answer the question

• Collects evidence

• Produces findings that were not determined in advance

• Produces findings that are applicable beyond the immediate boundaries of the study

Assumptions

When researchers undertake a qualitative study, they are in effect agreeing to its underlying philosophical assumptions, while bringing to the study their own world views that end up shaping the direction of their research. In the conventional view by statisticians, qualitative methods produce information only on the particular cases studied (e.g., ethnographies paid for by governmental funds which may involve research teams), and any more general conclusions are considered propositions (informed assertions). Quantitative methods can then be used to seek empirical support for such research hypotheses.

In contrast, a qualitative researcher holds that understanding of a phenomenon or situation or event comes from exploring the totality of the situation often has access to large amounts of "hard data". It may begin as a grounded theory approach with the researcher having no previous understanding of the phenomenon; or the study may commence with propositions and proceed in a scientific and empirical way throughout the research process.

Main Types of Qualitative Research

Case study	Attempts to shed light on a phenomena by studying in-depth a single case example of the phenomena.  The case can be an individual person, an event, a group, or an institution.(see chapter-3)
Grounded theory	Theory is developed inductively from a corpus of data acquired by a participant-observer.
Phenomenology	Describes the structures of experience as they present themselves to consciousness, without recourse to theory, deduction, or assumptions from other disciplines
Ethnography	Focuses on the sociology of meaning through close field observation of socio cultural phenomena. Typically, the ethnographer focuses on a community.
Historical	Systematic collection and objective evaluation of data related to past occurrences in order to test hypotheses concerning causes, effects, or trends of these events that may help to explain present events and anticipate future events.

Ethnography

Conventional social scientists often rely on survey or interview process to build understanding of cultural groups. These researchers are likely to make inquiries of a number of group members in a bid to understand their attitudes, beliefs, opinion, behaviour etc. Ethnography is the process of studying and describing a culture. It is a systematic, organized and detailed description of a particular culture and it also includes the process through which such information is collected. In other words, the term “ethnography” refers to both the research process and the descriptive results, and this encompasses both the means and the end of ethnic groups, community and workplace as well.

The origins of ethnography can be traced to cultural anthropology from which it derives its conceptual and methodological bearings. Basically, ethnography was limited to the studies of ‘primitive’ small scale societies. But a common critique in this study is that it is merely descriptive and the process includes following activities

.Observation - Participant observation refers to the research characterised by a period of intense social interaction between the researcher and the respondents. The aim of participant observation is enable the researcher to engage in activities appropriate to the situation and to observe the activities, people and physical aspects of the situation.

Interviews- ethnographic interviews are generally in-depth, open-ended and unstructured, the topics for which are determined by the context. The advantages of these flexible ethnographic interviews are that they can attain more information and capture the differences between ‘ideal’ and ‘real’ values, beliefs and perceptions that a community holds.

Document analysis- sometimes a good way to understand the reality of the researched is to examine the texts that they themselves produce. Depending on the nature of the cultural group being explored, this might involve in examination of local newspapers, television and radio broadcasts. It may also involve analysis of local art, the poetry and essays of school children, journals diaries etc.

Survey- the surveys are often critiqued by those conducting ethnographies for being too reductionist. While studies based on survey research would not qualify as ethnography, a survey instrument such as questionnaire may be the best way to canvas widely within a particular cultural group..

Grounded Theory

Grounded Theory also called rooted theory is a qualitative research approach that was originally developed by Glaser and Strauss in 1960s. The self-defined purpose of grounded theory is to develop theory about phenomena of interest; but they are not taking about abstract theorizing. As the researcher begins to gather data, core theoretical concepts are identified. Tentative linkages between theoretical core concepts and the data are developed in advance. Later on, the research is more engaged in verification and summary.  Grounded theory includes the following key analytic strategies:

Coding is a process for categorization of qualitative data and describing the implications and details of these categories. Initially open coding method is applied to develop initial categories and moved for selective coding for systematic code with respect to core concept.

Memoing is a process of recording thoughts and ideas as they evolve throughout the study. Memories involve an excessive marginal notes and comments. Again, early in the process these memos tend to be open; whereas later on they tend to increasingly focus in on the core concept.

Integrative diagrams and sessions are used  to pull all the detail together to help make sense of the data with respect to the emerging theory. The diagrams can be any form of graphic that is useful at that point in theory development. They might be concept maps, directed graphs or even simple cartoons that can act as summarizing devices. This integrative work can be best done in group sessions where different members of the research team can interact and share ideas to increase insights.

Phenomenological Study

Phenomenology is a study of phenomena as they present themselves in direct experience.  Phenomenology is sometimes considered a philosophical perspective as well as an approach to qualitative methodology. It has a long history in several social research disciplines including psychology, sociology and social work. Phenomenology is a school of thought that emphasizes a focus on people’s subjective experiences and interpretations of the world.

Literature Searching and Theoretical Framework

Review of literature means reviewing research studies or other relevant propositions in the related field of study so that all the past studies, their conclusions and deficiencies may be known and further research can be conducted. It is an integral and mandatory process in research works. In fact, review of literature begins with a search for a suitable topic and continues throughout the duration of the research work. It deals with a literature survey of the existing volumes of similar or related subjects.

The main reason for a full review of research in the past is to know the outcomes those investigations in areas where similar concepts and methodologies had been used successfully. Further an extensive or even exhaustive process of such review may offer vital links with the various trends and phases in the researches in one’s area of specialization, familiarizing with the characteristic percepts, concepts and interpretation, with the special terminology, with the rationale for understanding one’s proposed investigation. In this connection a review of previous related research projects will help the researcher to formulate a satisfactory structure for the project. It is not necessary to discuss about the past studies in detail i.e. only important results are to be described. Here, the author shares the insights s/he has gained from the review of literature and points out the gaps presently existing in what is known in her/his topic, and thus leads directly to the question they purposes to investigate. The review of literature is needed for:

1. Determination of research problem and relevant variables

2. Avoidance of repetition

3. Synthesis of prior work

4. Determining meaning and relationship among variables

Purposes

The purpose of literature review is to diagnosis all past information related to the development of selected topic or area of study in detail so that they can form an idea for further extension of the proposed research. Within a literature review, the literature is reviewed or explored so that the researchers can:

·         Inform readers about the development in the field-not only should a research study should provide the information about particular research question, it should also provide rich learning about the general topic. Inclusion of a strong literature review should provide readers with contextual learning through an up-to-date account and discussion of relevant contextual theories, methods and research studies that make up a particular topic’s body of literature.

·        Establish their own credibility- because researchers are responsible for the production of new knowledge. It is essential to show that the researcher possesses complete knowledge about the field, aware of relevant new development and conversant with academic and scientific discourse and debate within their research area. The literature review allows researchers to establish such credibility rigorous and critical evaluation of relevant research works; a demonstrated understanding of key issues and the ability to outline the relationship of their own work to the rest of the field.

·         Argue the need for, and relevance of the study- the literature review need to make an argument for a researcher’s own research agenda. It needs to set the current study within the context of past research. The literature review has the potential to identify ‘gaps’ that show the appropriate and significant nature of a study’s research questions. It can also justify methodological approaches by critically evaluating methods generally accepted, highlighting the limitations that might be common to past studies and   uncovering the possibly unwarranted assumptions that can highlight method. In addition to above broad purposes, in general, the literature review should:

·         provide a context for the research

• justify the research
• ensure the research hasn't been done before (or if it is repeated, that it is marked as a "replication study")
• show where the research fits into the existing body of knowledge
• enable the researcher to learn from previous theory on the subject
• illustrate how the subject has been studied previously
•  outline gaps in previous research
• show that the work is adding to the understanding and knowledge of the field
• help refine, refocus or even change the topic
• 
  

Kinds and writing a research survey

The process of literature survey or literature review is a critical informational analysis procedure. But listing of bibliography is not a research. Critical evaluation and analysis of research material for the purpose of interpretation and comparision requires reflective thinking for description. In this sense, classical literature review would demand that the researcher provide a critical analysis and comparision to be called a part of research.

In academic research, literature review holds a key place because any knowledge gained in the research is placed in a total perspective of the existing knowledge and integrated with an outcome of research problem. The role of literature review, related to research problems translated from the problems faced by the manager, is only to check whether a similar problem was reached earlier and solved, whether it is totally new or whether a method in the literature needs a modification to solve the current research problem. Literature review has four broad aspects:

Choice of literature: In simple sense, choice of literature means identification of books, journals, articles, reports and review in selected area of research. In academic research, a researcher gradually gains focus on area of research and choose the research materials upon his/her interest and practical experience.

Search for location/source of literature: The institution/university where the scholar is carrying out research may have a good library but in most cases that would be inadequate for getting all features required for the problem area. However it may help them to locate the source through indexes to periodicals, which list journals, articles in alphabetic order; particularly indexes in special areas like business management, operations research, technology and so on.

Study of literature: In recent times, the literature available in any research area is generally voluminous and the researcher needs precaution to select the relevant materials for review,   The researcher can divide literature into three broad categories. Firstly, general literature in the broad area of specialization has to be borrowed quickly, noting only the portion that may have a hearing on the problem area. Secondly, literatures related to the problem area in which research is being carried out.

Organization of the literature: The literature studied and noted in the above step should be organized and arranged (may be at a later date) in a meaningful and systematic way for presentation in the thesis or report. For organisation of the literature, the researcher needs to remember following points:

·         The sequencing of literature presentation should be from the general and broad to the specific and critical, leading to the problem statement.

·         The general literature should occupy fewer pages in the review than the specific one. In general literature only the prominent ones or milestones need to be dwelt upon and others may be clubbed together in a summarised manner.

·         In the literature, close to the problem the details of each study should be specified, classified and critically evaluated. Shortcomings should be discussed and shown how they are related to the research problem at hand.

Literature search through the internet

The literature survey must replicate the essential content of the source and avoid too much interpretation. The author of the literature survey can select the viewpoint and choose what is essential according to that reference. Own conclusions, opinions and interpretations must be separated clearly from the content in sources. Skillful referencing makes the literature survey much more valuable. The more citations a survey has, the more reliable it becomes. Literature search through the internet is an outcome of revolution in technology. Internet facility provides to view or search the world from sitting chair. This facility to review through internet is available from well equipped libraries. In Nepal the central library of Tribhuvan University is located at Kirtipur, Kathmandu and the researchers can join for internet survey.

The internet is a world Wide network of computers that supports electronic communication and gives user access to information and documents from distant sources. Because it is a combination of a world Wide communication system and the world's largest public library containing a seemingly endless range of information. Computer communication and resource discovery has central function of the internet. Electronic mail or messaging is the most wide spread communication function.

The World Wide Web (www) refers to that portion of the internet make up of servers that support a retrieval system that organizes information into documents called web page. World Wide Web documents include graphic images, video clips and sound clips are formatted in special programming languages such as HTML (Hyper Text Mark up Language) and XML (Extensible Mark up Language), that allow for displaying, linking and sharing of information on the internet.

A researcher who wants to find a particular site or document on the internet or is just looking for a resource list on a particular subject can use one of the many internet search engines. A search engine is a computerized directory that allows anyone to search the World Wide Web for information in a particular way. Some of the most comprehensive and widely used search engines are:   yahoo, Google, Lycos, Go, Alta Vista, Hootbot, Excite etc.

Relation of literature to research

In order to acquire existing level of knowledge of research, it is essential and preliminary tasks for each researcher to follow the existing literature in the area of interest. The literature review is an integral part of the entire research process and makes a valuable contribution to almost every operational step. It has value even before thinking about a research question for which the solution is searched in the research journey. In the initial stages of research, it helps to establish the theoretical roots of the study, clarify idea and to develop methodology, but later on the literature review serves to enhance and consolidate the knowledge base and helps to integrate findings with the existing level of knowledge. Literature review also plays an extremely important role to compare the findings with others. A well formulated review of literature is also helpful to write up report and integrate findings with existing knowledge either to support or contradict earlier research. Thus, the higher the academic level of research, the more important a thorough integration of findings with existing literature becomes. 5

Reviewing literature can be time- consuming, frightening and frustrating, but it is also rewarding. A research without review of literature is meaningless and existence of every research is evaluated on the basis of the depth of its review. A review of literature review has a number of functions to the research:

·         broaden knowledge of researcher in selected area

·         clarify and focus the research problem

·         provides a theoretical background of study

·         helps to improve the methodology

·         helps to establish the links between proposed and existing level of development in the study area

·         improve the ability to examine how the findings  contributed existing knowledge of profession

·         enable to contextualise the findings.

Theoretical Framework

The preliminary survey of literature and information provides a sound basis for theoretical framework. Mainly, it describes the relationships among the variables and elaborates the theory underlying these relations. A good theoretical framework provides the logical base for developing testable hypothesis. ‘The theoretical framework is the foundation on which the entire thesis is based. It is a logically developed, described and elaborated network of associations among variables that have been identified through such process as interviews, observation and literature survey. These variables are deemed relevant to the problem situation.’

She also identifies five basic components that should be incorporated in any theoretical framework.

§  The variables considered relevant to the study should be clearly identified.

§  The discussions should state how two or more variables are related to each other.

§  If the nature and direction of the relationship can be theorized on the basis of the findings from previous research, then there should be an indication in the discussions as to whether the relationships would be positive or negative.

§  There should be a clear explanation of why we would expect these relationships to exist the arguments could be drawn from the previous research findings.

§  A schematic diagram of the theoretical framework should be given so that the reader can visualize the theorized relationships.

Theory and Research

A researcher/investigator in business phenomena want to know what produces inflation. Another manager wants to know if organizational structure influences leadership style. Both of them wants to be able to predict behaviour or to be able to say that if we do such and such, then so and so will happen. According to Kerlinger "a theory is a set of interrelated concepts (constructs), definitions and proposition that present a systematic view of phenomena by specifying relationships among variables, with the purpose of explaining and predicting the phenomena (facts)."

A theory may not only explain or predict phenomena, but also specify causal relationships among variables. Prediction and understanding are the two purposes of theory. Accomplishing the first goal allows the theorist to predict the behaviour or characteristics of one phenomenon from knowledge of another phenomenon's characteristics.

Theory development is essentially a process of describing phenomena at increasingly higher levels of abstraction. Things that we observe can be described as concepts. A concept (or construct) is a generalized idea about a class of objects, attributes, occurrences, or processes that has been given a name. If the researcher as an organizational theorist has to describe phenomena supervisory behaviour, s/he needs to categorize empirical events or real things into concepts. Concepts are basic units of theory development. However, theories required an understanding of the relationship among concepts. To clarify the meaning of theory, consider an example ‘the standard of living of a family is the function of its income, size and life style’. This theory provides a basis for studying consumer behaviour and formulating appropriate marketing strategies.

The major components of theoretical ideas in the social and business management are:

·         Assumptions and ideas: They consists of un testable premises about the nature and aspects of events

·         Frames of reference: They identify the major dimensions of events that will be subjected to empirically scrutiny.

·         Concepts: Concepts are abstractions e.g. groups, motivation, democracy etc.

·         Variables: variables are relational units of analysis that can assume designated sets of values, e. g. age, educational level, income.

·         Propositions: Propositions are statements between or among variables.

·         Theory: The end-product of the above components is a theory.

Concepts are drawn from theory, but research cannot proceed on the basis of their theoretical meanings.  Theory and research may be distinct operations but in reality they are inseparable complementary components of scientific endeavour. In turn, finding of facts test theories and in suggesting new problems invite the formulation of new proposals. Theory is useful in research:

·         to suggest a problem for study

·         to formulate hypothesis for test procedure

·         to provide a conceptual model for study.

·          to help for selection of variables, concepts, data collection and analysis and intelligent findings.

To sum up, a researcher should use theory to plan and direct his/her lines of study, but use empirical observations to test and refine their theoretical propositions.

Deduction and Induction

Induction is the process of reasoning from a part to the whole. In other words it is a process of arriving universal generalizations from particular facts. Inductive method begins with a collection and study of facts and draws conclusion through a posteriori reasoning. In this type of reasoning the conclusion goes beyond the premises. In other words inductive conclusion is an inferential jump beyond the evidence presented. The conclusion is only a hypothesis and requires further confirmation before confidence. For example; if we say that some educated girls have expensive habits, therefore all educated girls have expensive habits, which not true in real sense. Similarly, if a firm spends one hundred thousand rupees on a regional promotional campaign and the sales do not increase. This is a fact that the sales did not increase even after promotional campaign. Under such circumstances the question like ‘why didn’t sales increase? may arise. The answer to this question is a conclusion that the promotional campaign was poorly executed.

Deduction is the method of deducing conclusions from universal to particular. This method presupposes certain conclusions and applies them to facts through a priori reasoning. In this method the conclusion only seeks to unfold what is in premises. It does not go beyond premises i.e. the conclusion is never more general than the premises. For example, if we say that all men are mortal and Peter is a man, therefore he is also mortal, which is certain.    

Induction and deduction are used in research reasoning in a sequential manner of double movement of reflective thought. Induction occurs when we observe a fact and ask, “why is this?” To answer this question, we advance a tentative explanation (hypothesis). The hypothesis is plausible if it explains the event or condition (fact) that prompted the question. Deduction is the process by which we test whether the hypothesis is capable of explaining the fact.

 Research Problems

After topic selection and preliminary information collection, the students have to specify the research problem and clearly identify the research questions to be addressed in the thesis. During preliminary survey of literature, the researcher may gain some idea about the problem of study. A research problem simply refers to some difficulty which researcher experiences in the context of either a theoretical or practical situation and wants to obtain a solution for the same. In other words, a problem is the difference between expected and desired values or outcomes or situation. If it does not meet the standard about any phenomena, it is a problem. A researcher may select a problem for investigation from a given theory of his/her area of specialization or considerable interest. Research problem can also be selected on the basis of daily experiences of the researcher. The research problems may be selected from the following sources.

§  Theory of one’s own interest.

§  Daily problems.

§  Technological changes.

§  Unexplored areas.

§  Discussion with experts and authorities including the supervisor/ research advisor.

There are three distinct research problems that differs in form, content and mode of verifications.

1. Empirical Problems: When social researchers answer questions or problems on the basis of what they come to know, through their sensory organs, these are expressed as empirical problems. Social researchers base their conclusions and findings on what they perceive or observe or sense in order to verify, approve or reject relationship between two or more variables.

2. Analytical Problems: Analytical problems are the questions whose answers depend on the meaning of the words in the sentences expressing them. These problems or statements are merely definitional and not empirical. These are essentially language and conceptual problems and not factual and scientific problems.

3. Normative Problems: Normative problems are questions whose answers depend primarily on value judgements. Value judgements are statements of what is desirable, preferred, moral, imperative or obligatory. These problems may take either an evaluation prescriptive form.

Components of a Research Problem

The formulation of a problem depends to the knowledge of the components of the problem. A researcher should identify the following five components of a problem.

§  There must be an individual or group, which has the problem. In case this individual or group utilises research to solve a problem then it may be termed as research consumer.

§  The research consumer must have some objectives, goals, ends or desire.

§  The research consumers must have alternative means available for achieving the ends and objectives. These means may be described as courses of action involving use of objects as instruments. The instruments may not be simple; artifacts but may be in the form of concepts (ideas) like mathematical formula, scientific definition, language etc.

§  The research consumer must have more than one alternative course of actions so that he/she may be in a position to adopt the comparative better course of action. The research consumer must decide which course of action is better and efficient.

§  There must be one or more environments to which the problems pertain. The formulation of a research problem depends to what extent the various components have been made explicit.

Research Questions

There are many questions raised in the research process. Some questions draw direct attention of the researcher for finding out uniformity of relationship between classes of social variables. Such types of questions are usually derived from a general theoretical orientation rather than a definite theory. The questions differ both in scope as well as degree of specificity. But each question has its own value in so far as augmenting of knowledge is concerned. Generally, first a question arises, then why of the question arises, and followed by possible answers and the questions originate from different sources and can be thus:

§  Questions of descriptive facts.

§  Questions dealing with adequacy of concepts.

§  Questions relating to empirical generalization.

§  Questions dealing with observed patterns of social organization and their consequences.

In addition to above sources, the researcher should concentrate about: ‘what one wants to know?’ ‘Why one wants to know and possible answer to the questions?’ etc. Generally such questions be answered which are considered relevant to other ideas and facts in the discipline on the one hand and help people in achieving practical values on the other. Usually such question helps in improving theoretical system. Theoretically a question may also draw attention on certain inconsistencies in commonly accepted ideas or finding conclusions whether such inconsistencies are real or seemingly real.

Research Hypothesis

Every researcher has to start with certain assumptions and presumption through which subsequent study might prove and disapprove. It is the hypothesis round which entire research process revolves. A hypothesis helps the researcher in proceeding further and finding solution of the problem, which he/she wants to study. Without hypothesis, the effectiveness of the research is not possible, to know the scope of study, nature of data to be collected and the one to be discarded. Again, the hypothesis helps in organizing the collected data in a very systematic way and in fact it stands at the mid-point of research directing towards particular way of finding tentative solution to the question of how and why. In scientific inquire, a well established hypothesis provides the guidelines of investigation. It must however, be clearly remembered and understood that there can be no readymade hypothesis, but in social research there is no well-developed theoretical system in many areas of study.

Definition of Hypothesis

A hypothesis is a conjectural statement of the relationship between two or more variables. Hypothesis is always in declarative sentence form, and they relate either generally or specifically, variables to variables.

In the words of George A. Lundberg ‘a hypothesis is a tentative generalization, the validity of which remains to be tested. In its most elementary state the hypothesis may be very hunch, guess, imaginative data, which becomes the basis for action or investigation.’ According to Webster, ‘A hypothesis is a proposition, condition or principle which is assumed, perhaps without belief, in order to draw out its logical consequences and by this method to test its accord with facts which are known or may be determined.’ The general functions of hypothesis are:

§  Adequate explanation of all the facts connected with the hypothesis.

§  It helps to collect necessary evidence in order to discover the order of nature.

§  It leads to the discovery of laws. It explains facts and laws, and thus seeks to verify knowledge.

§  Hypothesis limits the scopes of inquiry to a manageable area, because, instead of random collection of data, it enables us to search only for relevant facts. Therefore, it leads to economy of time and money as well as significant conclusion for the advancement of knowledge

Types of Hypothesis

There are mainly two types of hypotheses viz. crude and refined. A crude hypothesis is at the low level of abstraction. It indicates the kind of data to be collected. It is mainly concentrated to the descriptive method of research. However, it does not lead to higher theoretical research in the nature of theory or law.

The refined hypothesis is more significant in research and the degree of significance depends on the level of abstraction underlying the hypothesis. It is further categorized in to three types. The simple level hypothesis indicates the existence of certain empirical, uniformity in social research. It does not involve much verification. A complex ideal (relatively higher level of abstraction) hypothesis examines the logically derived relations between the empirical uniformity. This type of hypothesis is useful in developing tools of analysis and construction of further hypothesis. The highest level of abstraction is very complex and concerned with the inter relations of multiple variables, identifying the cause and effect relationship. For example, in order to examine the family planning and fertility in less developed locality, a number of complex factors such as income, culture, religion, tradition, education, health etc have to be considered.

Hypothesis may further be classified into descriptive or relational. Descriptive hypotheses are propositions that state the existence, size, form or distribution of some variable. e.g. the current unemployment rate in Kathmandu valley exceeds 15% of the work force is a proposition about the size of the unemployed group. A proposition about the form of unemployment might be ‘the unemployment in Kathmandu valley is concentrated to male adults under the age 22 years who have less than 10 years education.’ If these propositions are advanced as tentative statements about conditions-Kathmandu valley subject to verification by research, they are hypothesis.

When statement describes the relationship between two variables, it is called relational proposition e.g. families with higher income spends more for recreation. In this proposition, the relationship expresses positive correlation between variables, while the correlation implies some degree of interaction; the linkage strength is unclear and may not imply a cause-effect relation. On the other hand, when relationships between variable can be directly and more accurately observed, the linkage may state that A cause B or B is a function of A e.g. high unemployment among male adults in Kathmandu valley is a function of current industrial recession.

Null hypothesis: For statistical testing of hypothesis both alternative and null hypothesis are involved. Null hypothesis in its simplest form means that there is no difference between two populations in respect of some property and that difference, if any is only accidental and unimportant. In other words, null hypothesis is a principle, which states that a person is innocent unless he is proved guilty. It asserts that results found in research do not differ significantly from the expected results on a probability basis. Generally a null hypothesis is stated negatively and the object is to avoid personal bias of the investigator in the matter of collection of data. It is used to collect additional support for the known hypothesis. The null hypothesis is denoted by Ho.

Alternative hypothesis: An alternative hypothesis is the set of alternatives to the null hypothesis. In other words, the complement of null hypothesis is called alternative hypothesis. It always represents all other possibilities that are not included in null hypothesis. A researcher will determine which of the alternative course of actions or solutions or explanations can be applied to the problem. While finding out efficient alternative the researcher will, of course, have to think in terms of money, manpower, predictability, area to be covered etc. The alternative hypothesis is denoted by H₁ or H_a.

Sources of Hypothesis

  A hypothesis may be developed from various sources.

§  Based simply on hunch.

§  Finding through other study or studies.

§  Generated through a theory.

§  Personal experience.

Characteristics of Hypothesis

§  The hypothesis should be conceptually clear.

§  The hypothesis should be specific and capable of empirical test.

§  Related to available techniques.

§  Related to the existing environmental conditions and to body of theory for the purpose of testing.

§  Should identify the specific variable and their relationship.

Formulation of Hypothesis

Formulation of hypothesis is a complex work. Without prior information about the statement of inquiry the formulation of hypothesis is not possible. Following are some common problems that may arise during the formulation of hypothesis.

§  Absence of a clear theoretical framework.

§  Lack of ability to utilize the theoretical framework.

§  Failure to be acquainted with available research techniques so as is able to phrase the hypothesis properly.

Hypothesis is the pivot of the whole study. Without well-formulated hypothesis the whole study will be out of focus and it will be difficult to draw proper conclusions. It is a necessary link between theory and investigation, which will result in the addition to the existing knowledge. All the data is to be collected with an eye on the hypothesis. It is always in the interest of research that s/he should keep open and flexible mind to prepare, retrace the steps and to revise the hypothesis. In some cases, after collection of data, it might even become necessary to abandon the hypothesis.

Formulation of careful hypothesis becomes necessary because that makes the investigation easy. There can be number of hypothesis and some are to be discarded in the very beginning while others in the course of investigation. Hypothesis should be such that it results in efforts to discover something not already known i.e. why it is also called creative art important considerations for workable hypothesis have already been discussed and care should be taken to see that the hypothesis or problem of study is not only interesting but useful both for the society as well as the research.

Formulate the hypothesis under different situation includes: correlation, difference, difference between frequencies, directional statements, non-directional statements etc.

Role of hypothesis in research
Literary, a hypothesis is an assumption about possible outcomes of research is based on reasoning. The hypothesis plays a significant role in the scientific studies. The role of well formulated hypothesis is:
· The purpose of stating hypothesis, like the purpose of theories that may be involved, is to provide a framework for the research procedure and methodology. It directs the research activities.
· A research project is needed to proceed from a statement of hypotheses. Such hypothess are not ends in themselves but rather aids to the research process.
· A hypothesis takes on some characteristics of a theory which is usually considered as a larger set of generalization about a certain phenomenon.
· The verification of a hypothesis does not prove or disprove it, it merely sustains or refutes the hypothesis.
· Such hypotheses are not ends in themselves but rather aids to the research process.
· The conclusion of the research problem may also be stated in the context of the initial hypothesis.
· The hypothesis orients the research process for its verification rather than finding out the solution of the problem.

Research Design

   Concept

An architect prepares a blueprint before he/she approves a construction. An Army prepares a strategy before launching an attack. An artist makes a design before he/she executes his/her ideas. So also the researcher makes a plan of his/her study before undertaking the research work. This will enable to save time and resources. Such a plan of study or blue print for study is called a research design (strategy).

Thus, a research design is a plan for the collection and analysis of data. It presents a series of guide posts to enable the researcher to progress in the right direction in order to achieve the goal. The design may be a specific presentation of the various steps in the research process. These steps include the selection of a research problem, presentation of the problem, formulation of hypothesis, conceptual clarity, methodology, survey of literature and documentation, bibliography, data collection, testing of the hypothesis, interpretation, presentation and report writing. Generally, a common research design possesses the five basic elements viz. (i) selection of problem (ii) methodology (iii) data gathering (iv) data analysis and (v) report writing.

Purpose of Research Design

Research design is essential for the whole study and helps in finding out deficiency in expectation of the starting of work. Basically there are two purposes of a research design viz.

§  To provide answers to research questions; and

§  To control variance.

According to Kerlinger “design helps the investigator obtain answers to the questions of research and also helps him to control the experimental, extraneous and error variance of the particular research problem under study”. Since each research problem is prepared with a view to solve some problem, therefore the purpose of research design is to control variance and the other purpose of research design is that the researcher should find answer to research questions validly, objectively, accurately as well as economically. Designs are carefully worked out to yield dependable and valid answers to the research questions epitomized by the hypothesis.

By constructing an efficient research design the investigator attempts

§  to maximize the variance of the dependent variable influenced by the independent variables or variables of the substantive research hypothesis.

§  to control the variance of extraneous or unwanted variables that may have an effect on his/her experimental outcomes but in which he/she is not interested.

§  to minimize the error of random variation (fluctuation).

Features of Research Design

It is quite difficult to find a perfect research design free from its own strengths and weakness. However, a good research design should possess the following four characteristics.

1.   Objectivity: An objectivity research design implies to examine the evidence independent of beliefs, bias, emotions, perceptions, attitudes, hope and fear of any individual. More specifically, it is a degree of agreement between independent observers regarding the final scores assigned to different individuals. Hence, in order to ensure objectivity of the collected data a fairly objective measuring instrument must be used.

2.   Reliability: This is an attribute of consistency. A scale should give consistent results (for detail see in chapter 4).

3.   Validity: This refers to the ability of a scale to measure what it is supposed to measure (for detail see in chapter 4).

      To differentiate between reliability and validity, consider an example. Let us consider a clock, which is quite good, and accurate in time but is set ahead by, say five minutes. It’s time readings are reliable (consistent), but are not valid as judged by standard time.

4.   Generalizability: The fourth important aspect of the good research design is extent of generalization of the results/findings. The degree of generalization usually depends on population definition, sample size and statistical tools used in the analysis.

Types of Research Design

There are certain elements of design common to the empirical and the analytical study except their interpretation. Selection of proper design is basically determined on the basis of purpose, level and nature of study. Some common research methods, or approaches or designs are categorized under following five types:

1. Exploratory research design

2. Descriptive research design

·         Historical research

·         Descriptive research

·         Developmental research

·         Survey research

·         Case-study research

3. Comparative research design

·         Correlation research

·         Causal-comparative research

4. Interventional research design

·         Lab-based experimental research  

·         Field-based experimental research

5. Qualitative research        design

Exploratory Research Design

An exploratory research is conducted to clarify ambiguous problems. Management may have discovered general problems, but research is needed to gain better understanding of the dimensions of the problems. Exploratory studies provide information to use in analyzing a situation, but uncovering conclusive evidence to determine a particular course of action is not the purpose of exploratory research. Usually, it is conducted with the expectation that subsequent research will be required to provide conclusive evidence. For clarification, suppose an organization considering a program to help employees with child care needs, exploratory research with a small number of employees who have children might determine that many of them have spouses who also work and that these employees have positive reactions to the possibility of an on-site child-care program. In such a case exploratory research helps to crystallize a problem and identify information needed for future research.

An exploratory research has the purpose of formulating a problem for more precise investigation or for developing hypothesis. Besides, this study may, however, have other functions as follows:

·         Clarify concepts

·         Investigate in a subsequent, more highly structured study for research plan.

·         Establish priorities for further research

·         Provide a census of problems regarded as urgent by people working in a given field of social relations.

Thus, the major emphasis of this design is on discovery of ideas and insights. Therefore, the research design must be flexible enough to permit the consideration of many different aspects of a phenomenon.

Descriptive Research Designs

Descriptive research is essentially a fact finding approach relative largely to the present, and abstracting generalizations by the cross-sectional study of the current situations. Such studies involve a systematic collection and presentation of data to give a clear and accurate picture of a particular situation. These studies can be further classified into historical, descriptive, developmental, survey and case studies. The detail of each of these designs is as follows:

Descriptive Research

A descriptive research identifies problem or justify current conditions and practices to make comparisons and evaluations in making future plans and decisions. In other words, it is a process of accumulating facts.   Also its use in social science is more common as in socio-economic surveys and job activity analysis. One may undertake a descriptive study about the workers in a factory, their age distribution, community-wise distribution, educational level, state of physical health etc. A descriptive research is a process of accumulating facts, views or opinion of the people towards any right, capital punishment, college autonomy, an industrial establishment etc. It does not necessarily seek to explain relationships, test hypothesis, make predictions or get at meanings and implications of the study. Also, descriptive research cannot predict and control the events and conditions. It contributes to the facts by formulating exploratory hypothesis and checking the validity of existing theories. The general steps of descriptive study are:

§  Formulation of objectives of the study

§  Defining the population and selecting a sample.

§  Designing the method of data collection.

§  Analysis and interpretation of the data/facts.

Limitation

The descriptive method has certain limitations, because all problems cannot satisfy each of the required criteria. Some of them are:

§  Research may make description an end itself.

§  Research is essentially creative and demands the discovery of facts in order to lead to a solution of the problem.

§  It dominates statistical tools to present the facts in terms of average, correlation coefficient; dispersion may not always be either necessary or welcome.

Developmental Research

The developmental research is concerned to predict the future trends of the characteristics. It is focused on the study of variables, their rates of change, directions, sequences and other-interrelated factors over a period of time. Developmental research/study may be conducted in the following three ways:

Longitudinal study

Longitudinal study measures the nature and rate of change of the characteristic at different stages of development.  In this method, data are collected at two or more points (over a period) of time from the same group of individuals at regular interval. The intervals are not fixed so their length may vary from study to study. Intervals might be as short as a week or longer than a year. Irrespective of the size of the intervals, the information gathered each time is identical. A longitudinal study can be seen as a series of repetitive cross-sectional studies. For example when we wish to study the proportion of people adopting a program in relation to time, Longitudinal studies are also useful when we need to collect factual information on a continuous basis.  The general steps of developmental research are:

§  Defining the problem/statement of objectives.

§  Collection of data

§  Evaluation and analysis of data.

§  Interpretation.

Cross-sectional study

Cross-sectional studies, also known as one-shot or status studies are the most commonly used design in social sciences. This design is best suited to studies aimed at finding out the prevalence of a phenomenon, situation, problem, attitude or issue by taking a cross-section of the population. They are useful in obtaining an overall picture as it stands at the time of the study i.e. drawing samples from a cross-section of society at a particular point in time. It usually includes more subjects, but describe fewer growth factors than longitudinal studies. It is less expensive and faster because the actual passage of time is eliminated by different subjects across the range. This type of study is very suitable to examine the attitude of the people towards of an issue, socio-economic, demographic characteristics, incidence of certain disease, consumer satisfaction with a product etc.

Trend study

When the data are collected at intervals spread over a period of time it is called a trend study. Trend study is designed to establish pattern of changes in the past in order to predict future patterns. In other words, trend analysis enables the researcher to find out what has happened in the past, what is happening now and what is likely to happen in the future. This study is also useful to forecast trends by extrapolating from present and past trends. Time series and regression analysis are also useful tools for trend study.

The design involves selecting a number of data observation points in the past, together with a picture of the present or immediate past with respect to the phenomenon under study and then making certain assumptions as to future trends. For example, to examine the changes in political preference of a study population in relation to age, gender, income or ethnicity.

Cohort Studies

Cohort studies are based upon the existence of a common characteristic such as year of birth, graduation or marriage, within a subgroup of a population. Suppose we want to study employment pattern of a batch of bank managers who graduated from a university in 1990, or study the fertility behaviour of women who were married in 2005. To study the managers career paths we need to contact all the managers who graduated from that university in 1990 to find out their employment history. The same rule is applied to examine the fertility behaviour. Both of these studies could be carried out either as cross-sectional or longitudinal designs.

Panel study

Trend, cohort and panel studies are similar except that panel studies are longitudinal and prospective in nature and collect information from same respondents. In trend and cohort studies the information can be collected in a cross-sectional manner and the observation points can be retrospectively constructed.  In a panel study the observations are made over a period of time and are prospective in nature. Suppose we want to study the changes in the pattern of expenditure on household items in a community. For this purpose we need to select few families to find out the amount they spend every month on household items. We must keep collecting the same information from the same families over a period of time to ascertain in the expenditure pattern. Such a study is called a panel study.

Case study Research

A popular method of qualitative research is the case study which examines in-depth "purposive samples" to better understand a phenomenon hence, smaller but focused samples are more often used than large samples which may also be conducted by the same or related researchers or research centres.5A case study is a fairly intensive examination of a single unit viz. a person, a family, an institution, a commodity, a district or any single event. It is an approach, which views any social unit as a whole. It is a way of organizing social and business data so as to preserve the unitary character of the social and business object. A case study enables us to explore, unravel and understand problems, issues and relationship in a particular situation. However, it cannot allow us to generalize our research i.e. to argue the findings, results or theories from one case study to other case studies. The sources of data are through interview, exploratory observation, personal documents, life history etc. A case study includes following steps:

§  Determination of present status of the problem under consideration.

§  Enumeration of cases of the problem in hand.

§  Analysis of the background information to formulate adequate hypothesis.

§  Testing of hypothesis.

§  Validity of the conclusion drawn.

Assumptions

§  The unit is representative of the total.

§  The total phenomena is much complex and the study of which is neither advisable nor practicable

§  The unit is studied as a whole.

§  Time factor has its impact on social phenomena

Limitations

§  False sense of confidence

§  Difficulties in collection of historical data

§  False generalization

§  Lack of quantitative study

§  Expensive

§  Possibility of error

 Experimental Research

The experiment is the basic tool of physical sciences for tracing cause and effect relationship and for verifying inferences. In other words, experiment refers to the part of research in which some variable are controlled while others are manipulated and their effects on controlled variables are observed. The usual approach is to hold all variables constant except one. By varying this one and monitoring changes in the output, the relationship between variables can be carefully studied and documented. Experimental studies have their purposes to test a hypothesis of a casual relationship between variables. According to Isaac “the purpose of experimental research is to investigate possible cause and effect relationships by exposing one or more experimental groups to one or more treatment conditions and comparing the results to one or more control groups not receiving the treatment”. Mathematically, the approach is described as: y = f(x₁, x₂ .  .  . x_n)

Where y is dependent variable called output, f is the function of x_i’s (i = 1, 2... n) are independent variables called inputs. Then making all x_j’s, j¹i constants and varying x_i the effect of the change in x_i on y is observed. The experimental research technique consists three features namely identification of factors, replication of the experiment and quantitative measurement of the result. Total position of the sample or population that is exposed to a manipulation of the independent variable is known as treatment group and a group in which the independent variable remains unchanged is called control group. This control is achieved by one of the following two methods.

1.   Randomization: It involves rand only assigning subjects to treatment and control groups.

2.   Matching: It is the deliberate assignment of subjects to treatment and control groups to achieve a balance on key dimensions.

Although experimental research is a basic approach in physical sciences, its application in the social sciences is still in its infancy.

Types of Experiments

1.   Natural or uncontrolled: Natural or uncontrolled research is an experimental research based on natural, real setting and mostly on observations. This type of research is applicable in astronomy
.

2.   Laboratory experiment: Laboratory researches are experiments conducted under completely controlled situations (homogeneous conditions) that are created for special purposes. The laboratory experiment is applicable in physics, chemistry, and psychology in which the researcher reproduces and manipulates the desired situation.

3.   Field experiment: These experiments are conducted in a realistic environment thereby minimizing the effects of unwanted variables. However, development of statistical control methods facilitates the researcher to conduct many field experiments with little interference from normal activity and with several variables being manipulated at once. This statistical tool helps in estimating the effect of individual treatment, their interactions and significant differences in their effects. This type of research is best suited to social sciences.

The experimental research is guided by the following logical steps:

§  Identification of the problem

§  Formulation of hypothesis

§  Collection of data

§  Verification of data

§  Interpretation

Limitations

Even though, the method facilitates comparison to other methods, provides accurate results with certainty, logically, there are some limitations too

§  Practical problem of organization

§  Theoretical problem relating to the replication

§  The degree of accuracy is not confidential because of adequate information and model used for verification.

Casual- Comparative Research

Casual- comparative research is also called regression research method that investigates the possible causes affecting a particular situation by observing existing consequences and searching for the possible factors leading to these results. This method also predicts the dependent variable on the basis of independent variables. Also the cause is manipulated to measure the effect in various ways. In casual comparative research, the investigator takes one or more dependent variables and examines the data through backward movement to examine causes, relationship etc. In other words, the researcher searches the likeliness and differences among the selected topic to obtain clues about what might cause or contribute to the occurrence of a particular phenomenon. This research is ‘ex-post-facto’ in nature and the procedures are similar to examine the results.7

The research involves following steps:

§  Formulation of the problem

§  Statement of hypothesis

§  Data collection

§  Analysis and generalization

§  Interpretation

Measurement, Scaling and Sampling

Types of variable

A variable is a representation of a characteristic at different point or a quantitative factor. It is called variable because, numerical response varies in magnitude from one element to other element. Weight, height, income etc of the individuals are variable. In other words, variables are the objects, events, things and beings, which can be measured, manipulated, controlled or observed by the experimenter. P.V. Young defined it as “variable is any quantity or characteristic which may posses different numerical values or categories”. Qualitative phenomena or attributes are also quantified on the basis of the presence or absence of the concerning attributes. Variables can be classified in several ways. Some of the commonly accepted classifications are presented below.

Dependent and independent variables: If one variable depends upon or in a consequence of the other variable, it is called dependent variable and the variable that is antecedent to the dependent variable is called an independent variable. In other words, a variable, which is manipulated or predicted or estimated, explained by other variable, is called dependent and the variable, which manipulates or explains or predicts other variable, is called independent variable. For example, if we say that height depends upon age, then height is a dependent variable and age is an independent. Similarly production of wheat depends upon amount of fertilizer used. Therefore, production of wheat is dependent variable and fertilizer is an independent variable.

Qualitative and quantitative variables: The qualitative variables refer to those variables which consist of categories that cannot be ordered in magnitude. We cannot make such a statement regarding the qualitative variable, e.g., ‘category x possesses higher or lower magnitude of the variable then category y’. Thus the qualitative variable comprises the categories, which don’t have a quantitative relationship among themselves. Sex, race and religion are examples of qualitative variable because they cannot be ordered in magnitude. The quantitative variables refer to those variables which are composed of categories that can be ordered in magnitude.

The quantitative variables are further divided into continuous and discrete form. A continuous variable is one which is capable of being measured in any arbitrary degree of exactness. Age, height, intelligence etc. are examples of continuous variable. The discrete variable also called categorical variables is those variables which are not capable of being measured in any arbitrary degree of exactness because the variables contain a clear gap. For example, the numbers of members in a family, number of newspapers sold daily etc. constitutes the example of a discrete variable.

Intervening variables: So far there is a cause and effect relationship between dependent and independent variables; however, there are many problems where one major variable of interest may depend upon the independent variable provided the third variable does not come into picture. The presence of a third variable modifies the originally expected relationship between the independent and dependent variables. This variable is called intervening or extraneous variable. Thus a variable which influences the nature and degree of relationship between dependent and independent variable is called intervening variable. Whatever effect is noticed on dependent variable as a result of intervening variable is technically described as an experimental error. Such variables are not named in the proposed research and hence, little attention is given to them. For example, a manager’s interest is to study and explain the relationship between training and productivity. The more the training given to the workers, the greater the productivity is expected. Thus, productivity is the dependent variable and training is independent variable. However, this positive relationship holds true only with young workers. The training given to workers who are over 50 years of age may not lead to increase in productivity. Thus, age is the intervening variable.

Measurement and scaling

It is really difficult to measure the quantitative as well as subjective problems in the field of social sciences. For example, it will be difficult to measure attitudes, skills, values, coordination, indebtedness etc. unless certain tools for measuring these with significant certainty are developed. A measurement procedure consists of the techniques for collecting valid, reliable and precise data through different sources. Thus, measurement is the assignment of numerals to objects or events according to rules. In other words measurement is a value, which reflects the presence or magnitude of some characteristics. A numeral is a symbol of the form; 1, 2, 3, .. or I, II, III etc. It has no quantitative meaning unless we give it such a meaning. It is simply a symbol of a special kind. It can be used to label objects such as baseball players, billiard ball or individuals drawn in a sample from a universe. The rules used to assign numerals to object depend upon the following four types of measurement.

Nominal measurement: This is the lowest form or level of measurement, which consists of non-overlapping categories through which observations may be classified. In this scale, the categories are distinguished by the presence or absence of certain properties i.e. ranking of characteristic is not possible. The numerals or symbols used have no numerical meaning, and cannot be added or ordered. Nominal measurement, the statistical operations like frequency, percentage, proportion, mode and coefficient of counting of contingency. Addition, subtraction, multiplication and division are not applied. This is also called attributes or qualitative characteristics such as age, sex, eye colour, religion, nationality etc.

Ordinal measurement: This is the second level of measurement in which, number denotes the rank order of the objects or the individuals. The numbers are arranged from highest to lowest, or from lowest to highest order, and reflects the comparison in terms of higher or lower, heavier or lighter, harder or softer etc. Students may be ranked 1st, 2nd, 3rd etc. in terms of their academic achievements to constitute the example of ordinal measurement. The permissible statistical operations in ordinal measurement are median, percentile, rank correlation coefficients and all those which are permissible for nominal measurement.

Interval measurement: This is the third level of measurement and includes all the characteristics of the nominal and ordinal scale of measurement. Interval assigns each measurement to one of an unlimited number of categories that are equally spaced i.e. interval scales have equal steps between successive intervals so that arithmetic becomes possible. An interval scale has no true zero point to indicate the absence of particular characteristics i.e. zero point on the scale does not represent the true or theoretical absence of the variable being measured. It also determines exactly how much more or how much less variables being measured are represented by each category. For example, there are five employees who are all one year apart in age. Employees A, B, C and D are respectively 4, 3, 2 and 1 year senior than employee E. In this case we can not say that A is twice as old as C and four times older than D. The reason is that, the exact age of any of them is unknown.

The common statistics used in such measurement are arithmetic mean, standard deviation, Pearson’s correlation coefficient and other statistics based upon them. Also t and F tests are also used to test the significance of the characteristics.

Ratio measurement: Ratio measurement is the highest form of data and has the same properties as interval data except that it has an absolute true zero point as its origin. In this scale, the ratio of two points is independent of the units of measurement, the operations and relations who give rise to the numerical values in the ratio scale are such that the scale is isomorphic to the structure of arithmetic. The common examples of ratio scales are the measures of weight, length, loudness etc. Ratio data; has all the properties of all three levels below it and interval data has all the properties of the two levels of data below it. This means that all data can be considered nominal. Also interval and ratio data can be reduced into ordinal data. 

Scale construction technique

While measuring attitudes of people we generally follow the technique of preparing the attitude scales in such a way that the score of the individual responses assigns a scale. Under this approach, the respondents express their agreement or disagreement with a number of statements relevant to the issue. While developing such statements, the researcher must note the following two points. People may conceal their attitudes and express socially acceptable opinions. They may not really know how they feel about a social issue. They may be unaware of their attitude about an abstract situation, until confronted with a real situation so as to predict their reaction. Even behavior itself is at times not a true indication of attitude. For instance, when politicians kiss babies, their behavior may not be a true expression of affection towards infants. Thus, there is no sure method of measuring attitude; we only try to measure the expressed opinion and then draw inferences from it about people’s real feelings or attitudes. With all these limitations in mind, psychologists and sociologists have developed several scale construction techniques for the purpose as follows:

Arbitrary scales: Arbitrary scales are developed on ad hoc basis and are designed largely through the researcher’s own subjective selection of items. The researcher first collects a few statements or items which he believes are unambiguous and appropriate to a given topic. Some of these are selected for inclusion in the measuring instrument and then people are asked to check in a list, the statements with which they agree. The values are summed up to secure the attribute scale. For example, a company’s image among its employees may be measured through the following items:

How do you regard your company’s reputation?

(i). As a place to work Bad…………good

(ii). As a pay master Bad…………good

(iii). as a provider of promotional opportunities Bad…………good

(iv). A s a concern for employees welfare Bad…………good

Each of these may be scored from 0 to 4 according to the degree of favorableness reported. The total of scores of all respondents for each item may be made and studied. The overall image can also be studied by combining the scores of each respondent for the five items.

Summated scales (or Likert-type scales): Summated scales consist of a number of statements which express a favorable or unfavorable attitude towards the given object to which respondent is asked to react. The respondent indicates his agreement or disagreement with each statement in the instrument. Each response is given a numerical score, indicating its favorableness or unfavorable ness, and the scores are totaled to measure the respondent’s position on the continuum of favorable unfavorable ness towards an issue. In a Likert scale, the respondent is asked to respond to each of the statements in terms of several degrees, usually five degrees (but 3 or 7 degrees may also be used) of agreement or disagreement. For example, when asked to express opinion whether one considers his job quite pleasant, the respondent may respond in any of the following ways: (i) strongly agree, (ii) agree, (iii) undecided, (iv) disagree and (v) strongly disagree.

We find that these five points constitute that scale. At one extreme of the scale there is strong agreement with the given statement and at the other, strong disagreement, and between them lie intermediate points. We may illustrate this as under.

(1) Strongly agree  (2) Agree (3) Undecided  (4)Disagree  (5) Strongly disagree

Each point on the scale carries a score. Response indicating the least favorable degree of job satisfaction is given the least score (say 1) and the most favorable is given the highest score (say 5). These score values are normally not printed on the instrument but are shown here just to indicate the scoring pattern. The Likert scaling technique, thus, assigns a scale value to statement in the instrument. This way the instrument yields a total score for each respondent, which would then measure the respondent’s favorable ness towards the given point of view. If the instrument consists of, say, 30 statements, the following score;

30 x 5 = 150 most favorable response possible.

30 x 3 = 90 a neutral attitude.

30 x 1 = 30 most unfavorable attitude.

The scores for any individual would fall between 30 and 150. If the score happens to be above 90, it shows favorable opinion to the given point of view, a score of below 90 would mean unfavorable opinion and a score of exactly 90 would be suggestive of a neutral attitude.

Equal appearing interval scales: Thurstone equal-appearing interval scales also consists of a number of items indicating agreement or disagreement response. Thurstone scale assumes that all the items on the objective test gives equal importance to the attitude items. But actually some items are given higher values than other items. Each item is assigned a scale value that indicates the strength of attitude of an agreement response to the item. Also the universe item is considered different scale values in an equal interval. The following equal-appearing interval items, with the scale values of the items are from Thurstone and Chave’s scale, attitude towards the church are presented as follows:

‘.I believe the church is the greatest institution in America today (scale value 2)’

‘I believe in religion, but I seldom go to church (scale value 5.4)

‘I think the church is a hindrance to religion for it still depends upon magic, superstition and myth (scale value 9.6)

This implies ‘the lower the scale value, the more positive the attitude towards the church’.

Cumulative scale: Cumulative or Guttman scale, consists of a relatively small set of homogeneous items that are unidimensional (one-variable). The scale gets its name from the cumulative relation between items and the scores of individuals. If a scale is unidimensional, then a person who has a more favorable attitude than another should respond each statement with equal or greater favorableness than the other. Under this technique, the respondents are asked to indicate in respect of each item whether they agree or disagree with it, and if these items form a unidimensional scale, the response pattern will be as under. For example, we ask three arithmetical questions to four children as;

a. 28/7 = ? b. 8x 4 = ? c. 12 + 9 = ?

A child who gets (a) correct is very likely to get the other two correct. The child who misses (a) but gets (b) correct is likely to get (c) correctly. A child who misses (c) on the other hand is not likely to get (a) and (b) correct.

Scales commonly used in business research

Generally, comparative and non-comparative scales like summated rating scales, equal-appearing scales and cumulative scales are widely applicable in social and business research. However, due to its nature, following types of scaling are also useful in business research.

Likert scales

This is a most commonly used scale based on judgement methods based on agree or disagree format. Each category is assigned a numerical score for representation. The rule of assigning number is either in ascending or descending order of responses. It is necessary to inform the respondents for clearly to respond their preference. The scores on the scale are summed (summated scaling) up to get the total score for an individual related to management issues.

Example: The objectives of the R & D department of your organization are clearly set.

SD D U A SA

1 2 3 4 5

Where, SD= strongly disagree, D= disagree, U= undecided

A= agree, SA= strongly agree

Semantic differential

This is a quantitative type of judgment method that results in assumed interval scales. This scale is obtained on factor analysis of these assumed scale values and can be used rather easily and useful in decisional survey research employing multivariate statistics. It is based on the proposition that an object can have several dimensions of connotative meanings, which can be located in multidimensional property space. The opinion of respondents is expressed in equal intervals on a set of bipolar adjectives like; extremely clear-extremely ambiguous, extremely strong- extremely weak, extremely agree- extremely disagree and so on. Then arranging integer values to these intervals and the average of these scores for two groups can be compared to get a semantic differential profile. This type of scale is used to compare company ‘images’ and brands, determine attitudinal characteristics of consumers and analyze the effectiveness of promotional activities.

Example: The opinion of assistant about the head of department can be rated in terms of multidimensional bipolar points as;

     Cooperative            1          2          3          4          5          Non-cooperative

Stapel scale

The staple scale is a modified version of semantic differential. It is an even numbered non-verbal rating scale using single adjectives instead of bipolar opposites in which both intensities are measured concurrently. The standard staple is uni-polar in which 10-interval rating scale with values ranging from +5 to -5. However, any number of value can be used and they can be all positive i.e. 1 to 10 as well. In this scale, a single adjective in the centre of an even number of numerical value is placed for analysis. The staple scale produces similar results like semantic differential and the process of analysis is also same. This method, neither the equality nor the additive of ratings of a respondent is assumed.

Example: Respond your opinion about this product;

                        Reliable                       |..................|.....|......|..................|

                                                            +5             +1         -1               -5

Graphic and itemized rating scales

This is a most common method of measuring attitude, in which the rater is required to select one of the limited numbers of categories that are ordered in terms of their scale positions. These scales are widely useful in business research activities and are considered as basic building blocks for the complex attitudes.    This method assumes either a numerical or a graphic form. But rating scale must be adjusted to the nature of information required and the characteristics of the respondent.

Example: How do you rate my service level

i. Graphic

        Good                                   Satisfactory                        Poor                 

             |.......|.......|......|.......|........||.....|......|.......|........|.......|

Rank order rating scales

This also common and widely accepted method of attitude measurement in which the preferences of respondents are assigned ordinal numbers called rank. The rule of expressing preference is from most preferred to the least one. This method is purely comparative in nature. The rank order method requires the respondent to rank a set of objects according to some criterion. This method forces respondents to discriminate among the relevant objects. It consumes less time but is based on ordinal data only.

Numerical scales

Numerical        is also very similar to semantic differential scale except that it uses numbers instead of verbal description for response of respondents. The numerical scale utilizes bipolar adjectives in the same manner as is used in semantic differential. The number of response position can be expressed as the point scale. For illustration; if the scale items have five response positions, the scale is called a 5-point numerical scale, for seven response positions, it is called 7-point numerical scale and so on.

Validity and reliability

Generally, the researcher is always aware about the possibility of inconsistency in the outcomes of research. There are various reasons behind it such as faulty planning, selection of problem, collection of desired facts and tools for analysis. It is necessary to evaluate the accuracy and dependability of the measuring instrument. The criteria for such evaluation are validity, reliability and practicality.  Validity refers to the extent to which a test instrument measure what we actually wish to measure, Reliability reflects the accuracy and precision of measurement procedure. Practicability is concerned with a wide range of factors of economy, convenience and interpretability. Practicability requires expertise and sound resources (detail is beyond the scope of the book). In this chapter, we will discuss about reliability and validity only.

Reliability

Reliability means consistency, dependability, predictability and stability of response or measuring instrument. In other words, a reliable instrument will give trust worthy and stable results if it is applied to the same individuals or the object from time to time, provided the trait being measured has not itself changed in the mean time. For example, a weighing scale is reliable, if it gives the same reading when the same object is weighted several times.

Reliability has two aspects i.e. stability and non-variability or equivalence. The degree of stability is determined by comparing the results of repeated measurements with the same instrument. Along with repeated observations, statistical measures of dispersion and correlation tests are used. While stability is concerned with personal and situational fluctuations from one time to another, non-stability or equivalence is concerned with variations at one point in time among investigators and samples of items.

Validity

Validity is the degree of agreement between actual measurement and proposed measurement. If we measure what we are intended to measure then the measurement is said to be valid. In other words, validity of a test is concerned with the accuracy of the test scores when compared with the performances of an independent standard criterion. Validity of a test is determined experimentally by obtaining the coefficient of correlation between the scores of individuals on the given test and some independent standard test called criterion. [for methods refer book chapter-4]

Introduction of Sampling

Sampling is the process by which inference is made to the whole by examining only a part. It is woven into the fabric of our personal and public lives. In some cultures, a couple enters into marriage partnership on the basis of a short courtship. With a single grain of rice, a village housewife tests if all the rice in the pot has boiled, from a cup of tea a tea taster determines the quality of the brand of tea.

When some of the elements are selected with the intention of finding out something about the population from which they are taken, that group of elements is referred as a sample and the process of selection is called sampling. Simply, speaking the method of selecting a portion of the universe with a view to draw conclusions about the universe under study is known as sampling.

As a part of the information collection and decision making process, sample surveys are conducted on different aspects of life, culture and science. The purpose of sampling is to provide various types of statistical information of a quantitative or qualitative nature about the whole by examining a few selected units. Sampling method is the scientific procedure of selecting those sampling units which would provide the required estimates with associated margins of uncertainty, arising from examining only a part and not the whole.

The enumeration of population (universe) by sampling methods, proposed by Laplace in 1783, came into widespread use only by the mid-thirties of this century. From the outset, some basic questions arose:

§  How should the observations be made?

§  How many observations should be made?

§  How should the total sample be made?

§  How should the data, thus obtained be analysed?

The answers to these questions were sought and in the process a number of different techniques and methods were developed. These methods were tested to determine whether the above mentioned questions were adequately answered or not.

When and Where Sampling/Census is Appropriate

A sampling technique is appropriate

§  When the universe is very large (vast data)

§  When the universe possesses homogeneous characteristics

§  When utmost accuracy is not required.

§  Where census is impossible i.e. in destructive/explosive nature of testing.

A census is appropriate when

§  The universe is small

§  The population is heterogeneous.

§  Hundred percent accuracy is required

§  The population frame is incomplete.

Demerits of Sampling Technique

1.   Less accuracy: In comparison to census technique the conclusions derived from sample are more liable to error. Therefore, sampling technique is less accurate than the census technique.

2.   Changeability of units: If the units in the field of survey are liable to change or if they are not harmonious, the sampling technique will be very hazardous. It is not scientific to extend the conclusions derived from one set of sample to other sets, which are unlike or are changeable.

3.   Misleading conclusions: If due care is not taken in the selection of samples or if they are arbitrarily selected, the conclusions derived from them will become misleading if extended to all units. For example, in assessing the monthly expenditure of consumers if one selects only rich consumers in the sample and extend the results obtained by these sample to the whole consumers. It may be highly erroneous.

4.   Need for specialized knowledge: The sample technique can be successful only if a competent and able person makes the selection otherwise the selection is liable to be wrong.

5.   When sampling is not possible: Under the conditions when 100% accuracy is required, the population is heterogeneous, when population is very small; the sampling method is not useful.

Disadvantages of Census

Every coin has two faces, therefore, whereas census yields highly reliable and detailed information, it is not free from drawbacks. Following are the drawbacks of census methods.

1.   Expensiveness: As the method examines each and every unit of the universe, census method must entail much expenses. Because of the exorbitant cost of such surveys, the census surveys are as a rule made by government only. The government needs accurate records of population; productivity etc. because of excessive expense, even government undertakes such surveys at very long interval.

2.   Excessive time and energy: Besides cost factor, census survey takes too long time and consumes too much energy to be worthwhile normally. This type of survey is undertaken only in very special cases.

3.   Unsuitability in certain cases: Some problems range over such a wide area that to study all units under it is almost impossible. For example, if the scope of the problem is infinite, one cannot succeed in making total survey. For example, if one wishes to study the sexual habit of mankind, it is not quite possible to approach each individual for account of his/her sexual activity.

From above discussions, the following conclusion can be derived, under normal circumstances and in the vast majority of cases there can be no total survey of the field, and therefore census survey technique is simply unsuitable even though it provides detailed and reliable information.

Types of Sampling

The method of selecting a sample is of fundamental importance in sampling theory and usually depends upon the nature of the investigation. The sampling procedures, which are commonly used, may be broadly classified under the following types:

§  Probability sampling (random sampling)

§  Non-Probability sampling (or non-random sampling)

Probability Sampling

Probability sampling is the scientific method of selecting samples according to some laws of chance in which each unit in the population (universe) has some definite pre-assigned probability of being selected in the sample.

§  Each unit has an equal chance of being selected.

§  Sampling units have different probabilities of being selected.

§  Probability of selection of a unit is proportional to the sample size.

A probability sampling provides reliable and valid estimates by indicating the extent of error due to sampling a part of the whole being surveyed, and can also set limits with in which the unknown value that is estimated from the sample data is expected to lie with a given probability.

If the samples are selected partly according to some laws of chance and partly according to a fixed sampling rule (no assignment of probabilities), they are termed as mixed samples and the technique of selection of such samples is known as mixed sampling. The probability sampling is categorized under.

§  Simple random sampling   

§  Stratified sampling

§  Systematic sampling          

§  Cluster sampling

§  Multi-stage sampling

Non-probability Sampling

This is the method of selecting samples, in which the choice of selection of sampling units depends entirely on the discretion or judgement of the sampler. This method is mainly used for opinion surveys, but cannot be recommended for general use as it is a subject to the drawbacks of prejudice and bias of the investigator. However, if the researcher is experienced and expert, it is possible that judgement sampling may yield useful results. However, this method suffers from a serious defect that it is not possible to compute the degree of precision of the estimate from the sample values. The types of non-probability sampling are;

1.      Judgement of purposive sampling

2.      Convenience sampling

3.      Accidental sampling

4.      Quota sampling

5.      Snowball sampling

The most important difference between probability and non-probability sampling is that whereas the pattern of sampling variability can be ascertained in case of random sampling, in non-random sampling there is no way of knowing the pattern of variability in the process.

Selection of Appropriate Method of Sampling

A number of factors such as nature of the problem, size of the universe, size of the sample, availability of resources (time, budget and manpower) etc would influence the selection of a particular method of sampling. This is so because each method has its own speciality. The main consideration in selecting a method of choosing the sample are (i) that it be simple, but above all (ii) that it lines to the assumption of equal probability so that no bias is introduced.

Simple Random Sampling

The simplest and common most method of sampling is the simple random sampling (srs) in which the sample is drawn unit by unit, with equal probability of selection for each unit at each draw. Therefore, simple random sampling is a method of selecting ‘n’ units out of a population of size ‘N’ units by giving equal probability to all units. It is a sampling procedure in which all possible combination of ‘n’ units that may be formed from the population has the same probability of selection. It is also sometimes referred to as unrestricted random sampling. If a unit is selected and noted and returned to the population before the next drawing, this procedure is called simple random sampling with replacement (srswr). If every selected units are not returned it is called simple random sampling without replacement (srswor)

Procedures of Selecting a Random Sample

Since the theory of sampling is based on the assumption of random sampling, the technique of random sampling is of basic significance.

Some of the procedures used for selecting a random sample are as follows:

§  Lottery method

§  Use of random number tables.

Merits of Simple Random Sampling

§  No possibility of personal bias affecting the results because the selection of items in the sample depends entirely on chance.

§  It presents the universe in a better way.

§  Probability theory can be used to measure the precision of sample results because of sampling errors follow the principles of chance.

Demerits

§  Expensive and time consuming especially when the population is large.

§  Omission of any unit fails whole method.

Stratified Sampling

When the population characteristics are heterogeneous, then simple random sampling does not serve as a good design so as to represent the sample units from each characteristic. Then the entire population is divided / sub-divided into homogeneous groups or classes called strata. Then a simple random sampling procedure is used to draw sample from each stratum. The results of the samples from all strata are combined to estimate the universe parameters.

The main objective of stratification is to give a better cross-section of the population so as to gain a higher degree of relative precision. To achieve this, following points are to be examined carefully:

§  Formation of strata

§  Number of strata to be made

§  Allocation of sample size within each stratum

§  Analysis of data from a stratified design

The general principles of stratification are

§  The strata should be non-overlapping and should together comprise the whole population

§  The stratification of population should be done in such a way that strata are homogeneous within themselves, with respect to the characteristic under study and heterogeneous between themselves (strata).

§  In many (practical) situations when it is difficult to stratify with respect to the characteristic under study, administrative convenience may be considered as the basis for stratification.

A stratified sampling may be either proportionate or disproportionate. In a proportionate stratified sampling plan, the number of items drawn from each stratum is proportional to the size of the strata.

Merits

§  The units selected represents whole universe.

§  The estimation of population parameters is more efficient

§  For large and heterogeneous population, stratified sampling is the best design.

Demerits

§  This method requires more time and cost

§  Stratification of the units is quite cumbersome.

Systematic Sampling

A sampling technique in which only the first unit is selected with the help of random numbers and the rest get selected automatically according to pre designed pattern is known as systematic sampling. Suppose N units of the population are numbered from 1 to N in some order. Let N = nk where n is sample size and k is an integer, and a random number called random start is selected between 1 to k. Then every k^th unit will be selected automatically. This type of sampling is called linear systematic sampling. If N ¹ nk, and every k^th unit be included in a circular manner till the whole list is exhausted, it is called circular systematic sampling.

Systematic sampling is a commonly used technique if a complete and up-to-date sampling frame is available. The main advantage of systematic sampling is its simplicity of selection, operational convenience and spread of the sample over the population. It is useful in forest surveys for estimating the volume of timber, in fisheries for estimating the total catch of fish etc.

The drawback of this method is that it requires complete information about the population, which may not be possible and for special character is observed in equal interval of selecting samples, gives misleading results.

Cluster Sampling

In random sampling, it is presumed that the population has been divided into a finite number of distinct and identifiable units called sampling units. The smallest unit into which the population can be divided is called an element of the population. A group of such elements is known as cluster. When sampling unit is a cluster the procedure is called cluster sampling. Clusters are generally made up of neighbouring elements and, therefore the elements within a cluster tend to have similar characteristics. The number of elements in the cluster is not restricted. After dividing the population into specified clusters, the required number of clusters can be selected either by equal or unequal probabilities of selection and all the units in selected clusters are enumerated.

If the entire area containing the population under study is divided into smaller segments and each element in the population belongs to one and only one segment, the procedure is sometimes called area sampling.

For a given number of sampling units, cluster sampling is more convenient and less costly-the advantages of cluster sampling are:

§  collecting data for neighbouring elements in easier, cheaper, faster and operationally more convenient than observing units spread over a region.

§  it is less costly than simple random sampling and stratified sampling.

§  it is useful even when the sampling frame of elements may not be available.

Demerits

A cluster sampling is likely to have the following demerits

§  The efficiency decreases with increase in cluster size.

§  The efficiency per unit cost may be more in cluster sampling.

§  Enumeration of the sampling units within the selected clusters is difficult when the population size is large.

Multi-stage/phase Sampling

A sampling design, which consists in first selecting the clusters and then selecting a specified number of elements from each selected cluster, is known as sub-sampling or two-stage sampling. In such sampling design, clusters, which form the units of sampling at the first stage, are called the first stage units (fsu) or primary sampling units (psu) and the elements within clusters are called second stage units (ssu). This procedure can be generalized to three or more stages and is termed multi-stage sampling. For example, in crop surveys for estimating yield of a crop in a district, a Village Development Committee may be considered a primary sampling unit, the villages the second stage units, the crop fields the third stage units and a plot of fixed size the ultimate unit of sampling.

Multi-stage sampling has been found to be very useful in practice and this procedure is being commonly used in large-scale surveys, but this method is costly. Also a multi-stage sampling is less efficient than a single stage sampling of the same size.

Judgement or Purposive Sampling

A sampling method, in which the researcher selects the sample according to personal judgement, is called judgement purposive sampling. In other words, the investigator uses self-judgement in the choice and includes only those items of the universe in the sample, which are convenience to him/her. While choosing the samples only average items are considered and extreme items are omitted. Selection of the sample is adjusted in accordance with the object of inquiry, so that no significant item may be ignored. This method is suitable only when a universe is small and in solving every day business problem and making public policy decisions i.e. to the urgent problems. For example, the medical representatives’ contacts to the popular and busy Doctor are purposively. This method gives valid results when properly used i.e. if the researcher is skilled and apply the method fairly, otherwise because of personal judgement, biases and prejudices lead improper conclusions.

Convenience Sampling

A convenience sampling is also called a chunk. A chunk refers to that fraction of the population being interviewed/investigated, which is selected either by probability or by judgement. A sample obtained from readily available lists such as automobile registrations, telephone directory etc is a convenience but not a random sample even if the sample is drawn at random from the lists. The results obtained by this method can hardly be representative of the population: they are generally biased and unsatisfactory. However, convenience sampling is often used for making pilot studies. Questions may be tested and preliminary information may be obtained by the chunk before the final sampling design is decided. This method covers a  wide area and represent the units even from the heterogeneous population but the method does not remain idle from the judgement or biases of the selection.

Accidental Sampling

In this method, the researcher simply reaches out and takes the cases that fall to hand counting the process till such time as the sample reaches a designated size. For example, the researcher may take the first hundred persons he/she meets in any one of the pedestrian path of a street who are willing to provide the kind of information that the researcher is seeking. This type of sampling is economical and convenient, can also afford a basis for stimulating insights and hypothesis where too much accuracy is not needed. However, it suffers highly by the biases/prejudices.

Quota Sampling

In quota sampling, the interviewer is instructed to collect information from an assigned number or quota, of individuals in each of several groups. The groups being specified as to age, sex, income or other characteristics like the strata in stratified sampling. Quota sampling usually proceeds in three stages:

§  the population is classified in terms of properties known or assumed to be pertinent to the characteristics being studies.

§  the proportion of the population falling into each group or is determined on the basis of the known, assumed or estimated composition of the population and

§  each interviewer is assigned a quota respondents.

Quota sampling is widely used in market surveys and public opinion pools. It is cheaper and occasionally provides satisfactory results if the interviewer is well trained and unbiased.

Snowball sampling

Snowball sampling is the process of selecting a sample using networks or reference. In this method, first of all one unit or individual is selected from a group or organization to collect required information. Then the next unit is selected on the basis of reference given by the first one. Similarly, third unit is selected with reference of second and so on. This process is continued until the required number or a situation point has been reached, in terms of the information being sought.

This technique is useful to examine little about the group or organisation with few individuals who can direct the other members of the group. This method of selecting a sample is useful for studying communication patterns, medical representatives etc. But the method is totally dependent on the first reference.

Errors in sampling

An ‘error’ refers to the difference between the true value of a population parameter and its estimate provided by an appropriate sample statistic computed by some statistical device. It should be distinguished from mistakes or inaccuracies, which may be committed in the course of making observations, counting, calculations etc. These errors in statistics arise due to a number of factors such as:

§  Approximations in measurement

§  Approximations in rounding of the figures to the nearest integer

§  The biases due to faulty collection, presentation, analysis and interpretation of the results

§  Personal biases of the investigators etc.

In statistical investigation, these discrepancies (errors) between the estimated and the actual values are the net effect of a multiplicity of factors and can be broadly classified into following two groups.

Sampling and Non-sampling Errors

The inaccuracies or errors in any statistical investigation i.e. in collection, processing and analysis of the data may be broadly classified as follows:

§  Sampling errors and

§  Non-sampling errors.

Sampling Errors

In sample surveys, since only a small portion of the population is studied, its results are bound to differ from the census results and thus have a certain amount of error. Thus, the error arises due to estimating population parameters only by selecting few units (sample) is called sampling error or sampling fluctuation. Whatever may be the degree of caution in selecting the sample, there will always be a difference between the population value (parameter) and its corresponding estimate.

This error is inherent and unavoidable in any and every sampling scheme. A sample with the smallest sampling error is considered to be a good representative of the population this error can be reduced by increasing the size of the sample. The sampling errors committed due to

§  Faulty selection of the sample

§  Substitution of convenient unit of the population

§  Faulty demarcation of sampling units

§  Improper choice of the statistics for estimating the population parameter.

Non-sampling Errors

Non-sampling errors are not attributed to chance and are a consequence of certain factors, which are within human control. In other words, they are due to certain causes, which can be traced and may arise at any stage of the inquiry viz. planning and execution of the survey and collection, processing and analysis of the data. Non-sampling errors are thus present both in census surveys as well as sample surveys. Thus, the data obtained in a complete enumeration, although free from sampling errors would still be subject to non-sampling errors where as data obtained in a sample survey would be subject to both sampling and non-sampling errors.

Non-sampling errors can occur at every stage of the planning or execution of census or sample survey. The preparation of an exhaustive list of all the sources of non- sampling errors is a very difficult task. However, a careful examination of the major phases of a survey (enumeration or sample) indicates that some of the more important non- sampling errors arise from the following factors.

·         Faulty Planning or Definitions

·         Response Errors

·         Non–response Bias

·         Coverage Error

·         Compiling Errors

·         Publication Errors

Data Collection and Analysis

Data collection is considered as an integral part of the research activity and is relevant for analysis in a meaningful manner. The sources of information are generally classified into primary and secondary form.

 Primary sources also called field source are original sources from which the researcher directly collects data that have not been previously collected. In other words, data collected by the researcher or through agent for the first time from related field and possessing original character are known as primary source (or data). On the other hand, data collected by some one else, used already and are made available to others in the form of published statistics are known as secondary data. Once primary data have been used, it loses its primary characteristics (originality) and becomes secondary. The difference between primary and secondary data is a matter of relativity.

The sources of secondary data are books, journals, articles, magazines, dissertation, thesis etc. obtained from government, semi-government, non-governmental, private organizations or institutions. It also consists, unpublished records like accounting and financial records, personnel records, register of member, minutes of organizational meetings etc.

Features of secondary (sources) data: Though secondary sources are diverse and consist of all sorts of materials, have certain common characteristics:

·         They are readymade and readily available, and do not require the trouble of constructing tools and administering them.

·         They consist of data over which a researcher has no original control over collection and classification. Both the form and the content of secondary sources are shaped by others. Clearly, this is a feature which can limit the research value of secondary sources.

·         Secondary sources are not limited in time and space i.e. the researcher has no restriction to present when and where they were gathered.

Method of primary data collection: The researcher must decide at the outset about the use of primary or secondary data in an investigation. The choice between the two sources (primary and secondary data) depends on:

· Nature, scope and objective of study/inquiry.

· Availability of financial resources and time

· Degree of accuracy desired and

· The status of the investigator i.e. individual, corporation, government etc.

Primary data are generally used in those cases where the secondary data do not provide an adequate basis for analysis. In certain cases both data may be employed. The reason why secondary data are being increasingly used is that published statistics are now available covering diverse fields so that the researcher gets required data readily available to him/her in many cases. Besides these, the finance and time are also taken into consideration. For the collection of primary data, the following methods can be adopted:

1. Direct personal investigation (or observation)

2. Indirect oral investigation ( personal interview)

3. Telephonic interviews

4. Local correspondence/ channel of agency

5. Schedule / questionnaire through enumerators

6. Mailed questionnaire

Direct personal investigation: In this method, data are collected by the researcher / investigator through personal observations about the behavior of the sources. The researcher would not ask direct questions from the sources about his/her liking but would provide him/her an opportunity of an unmindful selection from various options. It is more suitable for collecting data concerning marketing inquiries. But, the method is quite expensive for detailed and expensive surveys, ensures personal bias with the investigator, observations are affected by the amount of knowledge of the researcher about the customs and psychology of the people.

Indirect oral investigation: Some times due to some specific reasons, the information cannot be obtained directly and on such circumstances an indirect approach is adopted to collect the relevant information. In this method the investigators contact the witnesses who had detailed knowledge, unbiased, capable of expressing themselves clearly and correctly for their replies. Such a procedure is adopted by commission, inquiry committee etc. appointed for carrying out investigation. This approach is more suitable in the inquiry regarding drug addicts, addicts of alcoholic, chronic disease, aids etc. This method does not seek direct contact with the original source; the scope of inquiry is extensive but less expensive and time saving, eliminates prejudices of the investigation unit.

Telephonic interview: A telephonic interview is a kind of oral interview in which the respondents/ informants are interviewed through telephone. This method is carried out for the inquiries through which immediate information is required. This mechanism is widely used in the field of business and commerce. This method saves time, money and labor, respondents are less embarrassed and ensures high rate of response, easy to contact high rank informants, recall is easy, covers wide area and no field staff is required.

Local correspondence: In this method the researcher do not move for the formal collection of the data but the information agents or correspondents are appointed in the areas under survey to send the required information to the head office. The nature of the appointment of these local agents may be regular or adhoc, depending upon the nature of investigation. No doubt, some instructions are issued to these agents but, in practice, they send information according to their own ways and decisions. Also the government offices establish the branch office at local areas to apply the regular information. This method is comparatively cheap and easy, the data are obtained expeditiously and the technique is quite useful when the field of investigation is wide spread. Because of a free hand of the third agency, called correspondents or agents, the data collected are not much reliable and degree of accuracy is quite limited. 

Schedule/questionnaire through enumerators: In this method, enumerators are sent to the respondents along with the schedules of questions or questionnaire to collect necessary information. The enumerators explain the aim and objective of an inquiry to the respondents and emphasize upon them to provide correct and useful replies. The questionnaire should be simple, straight and non-personal queries. Also the enumerator should be polite, courteous and well versed in social dealings. In this method, the enumerators can cross examine for getting the reliable answers, ensures large amount of information, covers wide area and substitution method can be applied for non-response sources. But the method is highly expensive and time consuming, affected by competence and honesty of enumerator.

Mailed questionnaire: In this method, questionnaires containing a list of questions with blank spaces for answers are mailed to the respondents (accompanied by necessary meanings and instructions) with the request that they should return them duly filled in within a stipulated time. The questions are so selected that the informant is expected to possess definite as well as accurate knowledge about them.

Questionnaires

A questionnaire is a form (list of questions) prepared and distributed to secure response to certain questions. It is a device for securing answers to questions by using a form, which the respondent fills by himself/herself. It is that form of inquiry, which contains a systematically compiled and organized series of questions that are to be sent to the selected samples. It is an important instrument in normative survey research, being used to gather information from widely scattered sources.1

Purpose of Questionnaire

The main purpose of questionnaire is:

§   To collect information from the respondents scattered in vast areas

§   To achieve success in collecting reliable and dependable data

§   To find information from personnel matters of respondents

Forms of Questionnaires

Like schedules, the questionnaires are also broadly divided as structured/ standardized, closed, mixed etc. as follows:

Structured questionnaires: "Structured questionnaires are those which pose definite, concrete and preordained questions, i.e. they are prepared in advance and not constructed on the spot during the questioning period",¹Additional questions if any, are put only when replies given to some questions are not clear or these are vague or needed to clarify some of the replies already given or when more details are needed than what these have been supplied by the respondents.

The non-structured questionnaire may consist of particularly completed statements or questions. It is often used in non-directive interview guide. The interviewer possesses only a blue print of the inquiries and is largely free to arrange the form or statements of the questions.

Closed form questionnaires: Closed form questionnaires are used when short cheek responses (categorized data) is required. They provide responses in the form of 'yes' or 'no' and restrict the choice of response for the respondent. In other words the respondents have to reply from one of the alternatives included in the questionnaire. For example, if the question is ‘Do you cook your food by:

§   burning wood

§   kerosene

§   cooking gas

§   use electricity? The informant is required to pick up one of the items suited to his/her conditions and reply.

It is easy to fill out, requires less time, keeps the respondent on the subject, is relatively more objective, more acceptable and convenient to the respondent and is fairly easy to tabulate and analyze.

The open-form, open-end or unrestricted questionnaire calls for a free response in the respondent's own words. The respondents are not given clues about the contents and have to supply response through own mind. This type of item is often difficult to interpret, tabulate, and summarize in the research report. It is useful where primary information is to be developed in qualitative phenomenon.

Mixed questionnaires: The mixed questionnaire consists of both open and close type of questions/items. Each type has its specific merits and limitations and the research worker has to decide about the use of proper type. This method is very useful in social research activities.

Unstructured questionnaires: The unstructured questionnaires also called fact and opinion questionnaires, deal with definite subject matter from the respondent without any reference to his/her opinion or attitude about them. In the words of P.V. young "it is designed to obtain view interconnection between the data, which might escape notice under more mechanical type of interrogation." It is used for the studies of personal experiences, beliefs and attitudes.

Pictorial questionnaires: Pictorial questionnaire is extensively used in studies of social attitudes and prejudices in children or illiterates. In a pictorial questionnaire, the selected alternative answers are given in the form of pictures and the respondent is required to tick the picture concerned. This type of questionnaire is most convenient to collect the data from rural masses that are mostly illiterate and less knowledgeable in a developing country like Nepal. This method is highly expensive and is lengthy in form.

Guidelines on Questionnaire Construction

A questionnaire is the device of collecting information or facts from the respondents. Construction of a good questionnaire is not simple task for the planners or researchers. Even a single characteristics influence the efficiency of facts (data).The structuring of questionnaire is considered a technique and that only few skilful persons can structure good questions of good questionnaires it is expected that:2

1.      The language of the question should be concise; Use words of precise (not-flexible) meaning: useful to insert transitional phrase, at the end use an expression of appreciation for the efforts put forth by the respondents, use cross tabulations and mock tables.

2.      Units of questions should be precisely stated or defined in order to ensure proper orientation of respondent.

3.      When the information sought relates to definite time period, such as fiscal or calendar year, the specific period should be stated in the questionnaire.

4.      subjective words such as 'bad', 'good', 'fair' and the like do not lend themselves either to quantitative or qualitative and as such should be avoided.

5.      The questions themselves should be precise and long questions should be avoided.

6.      No single question should deal with more than one issue and as such the principle of one question, one issue should be followed.

7.      The vocabulary employed in the question should be appropriate to the background of the respondents i.e. technical language should be used when the respondents have technical background.

8.      The questions should be so sequenced that the respondent is motivated and answers all questions.

9.      Non-sensitive and easily answerable questions should be introduced at the beginning and difficult or sensitive questions towards the end of the questionnaire.

10. Questions should be so worded that ego of the respondent is not injured in any way.

11All the questions should be analytical i.e. these should specifically deal with that aspect of the problem about which study is being made.

10.  The amount of writing required on the questionnaire should be kept to the minimum.

11.  It is better to add few questions that will serve as checks on the accuracy and consistency of the answers being given.

12.  As far as possible, the question, which calls for responses towards socially accepted norms and values, should be avoided.

13.  Questions on complex and controversial issues should be broken down into components, so that the tester can determine the respondents feelings about many aspects of the problem, including those to whom the respondent refuses to comment upon a series of specific questions is needed.

14.  When questions aim at seeking degree of intensity or feeling, it is better to find out the extent to which the respondent's attitudes have been crystallized towards the subject.

15.  Only such questions should be asked about which it is believed that the people have sufficient information.

16.  While deciding about the length of questions and statements, respondent's comprehension should be taken into account. Long question should be split up in the best possible manner.

17.  It is usually advisable to assign a serial number to each questionnaire in order to identify and control of interviewing assignments.

18.  When a questionnaire contains more than one page, each page should be numbered serially.

19.  The confidentialness of the data should be made clear in such a way that the respondents will be assured of protection.

20.  A place should be provided for the signature of the respondent unless none is required.

Process of Questionnaire Technique

The entire process of questionnaire is completed into three stages viz. designing, issuing and returning. The designing stage requires three important matters like physical appearance of the questionnaire, contents of questionnaire and subject matter. Special attention should be paid in order to give a good physical appearance to the questionnaire so as to get good response from the respondents. Matters requiring such a special attention include quality and size of paper, printing/type letters, margin, spacing arrangement of items on the questionnaire etc. Issuing and returning stages is the main responsibility of the investigation or enumerators to pay proper attention so as to get back the issued questionnaires completed by the respondents. The investigators are trained well before they go to visit the field.

Techniques of Eliciting Response through the Appeal

A covering letter or introductory letter almost accompanies the questionnaire. It must be brief and contains basic facts like the auspices (sanction authority), why the study, why should the respondent bother to answer, directions as to how to fill it, guarantee of anonymity etc.

Techniques for Returning the Questionnaire

Normally, response rate is low among the less educated, the lower occupational group and uninterested in the subject of survey. It is desirable to reduce the non- response rate to such a low level that it cannot cause a series bias and distort the survey, which is possible if the following four points are widely employed:

§   a stamped (stamp affixed) self-addressed envelope should be sent with the questionnaire,

§   accurate typography, printing if possible,

§   use of good grade, colour and convenient size of paper for preparing the questionnaire, and

§   follow up campaign/sending reminders.

Advantage of Questionnaire Technique

The questionnaire method is popularly used when the field of research is vast and the respondents are scattered over a very large area. By this method a large sample may be drawn and all group of people can easily be covered and contacted.

The advantages over this method may be pointed as follows:

1.   Economical: The questionnaire approach makes it possible to cover a large number of people spread over a large territory. It is more economical in time, money and energy. Hundreds of persons are approached through this method for which the interview may take a very long time.

2.   Suitable in special types of response: The information about certain problems can be best obtained through questionnaire method. For example, the research about sexual habits, marital relations, dreams etc can be easily obtained by keeping the names of respondents anonymous.

3.   Ensures anonymity: As the respondents are not required to indicate their names on the questionnaires, they feel free to express their views and opinions. Anonymity is a meaningful characteristic of questionnaire method, which shows more effectiveness than an interview.

4.   Less pressure on the respondents: The method places less pressure on the subject for immediate response, and gives more time to the respondents for properly answering the questions.

5.   Uniformity: The questionnaire is an impersonal technique of uniformity from one measurement situation to another is provided by virtue of its standardized wording of questions, sequence of questions and fixed instructions for recording responses.

Limitations of the Questionnaire Technique

Although the questionnaire technique is less expensive, covers wide area and requires less manpower, but it is not out of exceptions to be free from following limitations:

Illiterate and less educated: One of the major limitations of the questionnaire method is that it can be administered only on subject matters with a considerable amount of education, complex questionnaires requiring elaborate written replies can be used indeed on a very small percentage of population.

Proportions of returns are usually low: In a mailed questionnaire, the proportion of return is usually low: it may sometimes be as low as 10%.

Unintelligible reply: If a respondent misinterprets a question or writes reply unintelligibly, there is no possibility to correct it. In this approach, there is no facility for repeating questions, explaining them or seeking clarification of a particular response.

Processing of Data

After the collection of research data, an analysis of the data and interpretation of the results are necessary. Analysis of data comes prior to interpretation. The facts and figures collected are to be processed with a view to reducing them to manageable proportions. Data processing comprises of editing, coding, categorization and tabulation is an intermediary stage between the collection, analysis and interpretation of data.

Editing:  Editing is the process of examining the data collected in questionnaire/schedule to detect errors and omissions. When the whole data collection is over, a final and through check up is made in the first step. As far possible the edited facts/data are accurate, consistent, and complete and accepted for tabulation. Editing is a major step in the data processing and the editors should follow following guidelines:

·         Each editor should be familiar with instructions given to interviewers and coders as well as own editing instructions.

·         The editor should not destroy, erase or make illegible the original entry. The interviewer’s original entries should be crossed out with a single line so as to remain legible.

·         Make all entries on a schedule in some distinctive colour and in a standardized form.

·         All answers changed or supplied by the editor should be initiated.

·         Editor’s initials and date of editing should be placed on each schedule completed.

To determine quality, the accepted procedure is to sub sample all the completed data forms to answer the questions as:

·         Are the data forms complete?

·         Is there obvious indication of bias?

·         Are these errors in the recording?

·        How much inconsistency is there between and among the answers pertaining to the same subject?

·         Is there evidence of interviewer’s dishonesty?

Coding: Coding is the process or operation, by which responses or data are organized into classes/categories and numerals or other symbols are given to each item according to the class in which it falls. In other words, coding involves two important operations viz. deciding the categories to be used and allocating individual answer to them. In this method, coding frame is developed by listing all possible answers to each question and assigning code numbers or symbols to each of them, which are the indicators used for coding. While coding, the coder should follow the rules given below:

·   Give code number for each respondent for identification; generally different digits, numbers are used to identify the area, institution, department etc.

·        Give code numbers for each question.

·         Prepare the coding frame and instructions to coders.

·         Carefully scrutinize every coded item in the initial stages of coding.

·    Decide any change in coding due to change in the nature or importance of data after collection.

·        For open questions or verbation, clarify each response according to the basic content and code them accordingly.

·        No opinion/don’t know response, generally indicates neutrality and code them by giving a number common for representation.

Analysis of data means studying the tabulated materials in order to determine inherent facts or meanings. It involves breaking down existing complex factors into simpler parts and putting the parts together in new arrangements for the purpose of arrangement. A plan for analysis should be prepared in advance before the actual collection of material. In case where a plan of analysis has not been made beforehand, following four points are kept in mind.

·         To think in terms of significant tables that the data permit.

·       To examine carefully the statement of the problem and the earlier analysis and to study    the original records of the data.

·         To get away from the data and to think about the problem in layman’s terms.

·         To attack the data by making various simple statistical calculations.

In this method, coding frame is developed by listing all possible answers to each question and assigning code numbers or symbols to each of them, which are the indicators used for coding. While coding, the coder should follow the rules given below:

Then the next step of data analysis/processing is to categorize the data or facts in suitable classes. Analysis of data means studying the tabulated materials in order to determine inherent facts or meanings. It involves breaking down existing complex factors into simpler parts and putting the parts together in new arrangements for the purpose of arrangement. A plan for analysis should be prepared in advance before the actual collection of material. In case where a plan of analysis has not been made beforehand, following four points are kept in mind.

 Statistics deals with the collection, presentation, analysis and interpretation of numerical data. Every research is completed by using suitable statistical analysis. There are two types of statistical data application viz. descriptive statistical analysis and inferential statistical analysis.

Descriptive method of data analysis: Descriptive statistical analysis is concerned with numerical description of a particular group observed. Data collected from tests and experiments often have little meaning or significance until they have been classified or rearranged in a systematic way. This procedure leads to the organization of materials into few heads.

1.      Determination of range of the interval between the largest and smallest scores.

2.      Decisions as to the number and size of the group to be used in classification. Class interval is, therefore helpful for grouping the data in suitable units and the number and size of these class intervals depend upon the range of scores and the kind of measures with which one is dealing. The number of class intervals which a given range will yield, can be determined approximately by dividing the range by the interval tentatively chosen.

Most commonly used statistical tools are:

1.      Calculation of frequency distribution usually in percentages of items under study.

2.      Testing data for normality of distribution-skew ness and kurtosis.

3.      Calculation of percentiles and percentile ranks.

4.      Computation of measures of central tendency- mean, median and mode and establishing norms.

5.      Computation of measures of dispersion- standard deviation, mean deviation, quartile deviation and range.

6.      Measures of relationship- coefficient of correlation, reliability and validity by the rank difference and product moment methods (i.e. rank and Pearson’s correlation coefficient).

7.      Diagram and graphical presentation of data - frequency polygon, histogram, cumulative frequency polygon, bar, pie diagrams etc.

While analyzing the data, investigators usually make use of as many of the above simple statistical devices as necessary for the purpose of their study. There are some other complicated devices of statistical analysis which, the researcher use in particular experimental or complex causal- comparative studies and investigations.

1.      Student’s (t) and analysis of variance (F) for testing significance difference between statistics especially between means.

2.      Chi-square test for testing null hypothesis related to frequency of qualitative characteristic.

3.      Calculation of partial and multiple correlation, multiple regression for finding out causal relationship between various phenomenon involved in a situation.

Inferential method of data analysis: The primary purpose of research is to discover principles that have universal application. But to study a whole population in order to arrive at generalization would be impracticable or almost impossible. A measured value based upon sample data is called statistic. A population value estimated from a statistic is called parameter. A sample is a portion of a population selected for analysis. By observing the sample, certain inferences may be made about the population. Samples are not selected haphazardly, but the chosen in a deliberate way so that the influence of chance or probability can be estimated. Several types of sampling procedures are described, each one is particularly appropriate in a given set of circumstances. Thus, Inferential statistical analysis involves the process of sampling, the selection for studying a small group called sample that is assumed to be related to the large group from which it is drawn. The sample statistic computed from the selected sample observations, is used to estimate the population parameters whose actual value is generally unknown.

There are various statistical techniques for analyzing data. To choose an appropriate technique of an statistical analysis is the challenging task to the researcher.

Interpretation of data

Analysis and interpretation are the central steps in the research process. The object of analysis is to summarize the collected data so as to answer the questions under consideration. Interpretation is the search for the broader meaning of research findings. It makes an effort to establish continuity in work and to establishment of explanatory concepts. According to Cook “Scientific interpretation seeks for relationship between the data of a study and between the study findings and other scientific knowledge”. Interpretation cannot be considered in the abstract form and mechanical process. It calls for a critical examination of the results of one’s analysis in the light of all the limitations of his/her data gathering. It is an important step in the total procedure of research.

Forms of interpretation: It is an accepted creed that statistical data and information may be interpreted in various forms depending on the size and nature of data and the need of its interpretation. Some of the common and widely accepted form or basis of interpretation may be described as:

Relationship: One of the fundamental basis of interpretation is to find out the relationship between the phenomenon under consideration. In general it is seen that unless comparative analysis or study is made, true form of relationship between subject and the object can not be determined.

Proportion: Proportion is another aspect of making a study of interpretation and is generally ascertained to determine the nature and form of absolute changes in the subjects of study. Also when the object of study is too much variable over a period, then proportions are ascertained to interpret the data or information in a true form.

Percentages: When the object of inquiry is to determine the nature and extent of approximations only for a particular objective, then the method of percentage is often used for making interpretation as the basis.

Averages or other measures of comparisons: Method of averages or other methods of comparison is used to interpret statistical data and information. It is a matter of common experience that if a long statistical table is to be analyzed and interpreted, then the measures of central tendencies or other related devices are used to determine the clear cut results. Therefore, use of central tendencies is an essential and integral part of interpretation.

It is always possible to interpret all the collected data and information but if the aim is to interpret the given tables and information in a perfect and desirable manner certain pre-conditions like sufficient and accurate data, proper type of classification and tabulation, consistency of information and possibility of statistical treatment.

Precaution in interpretation: Just like in other step of research, it is important to recognize the error that can creep in interpretation. This step is almost purely subjective and many errors are possible at this point. Following are common errors of interpretation, which need to be avoided.

1. Failure to see the problem in proper perspective.

2. Failure to appreciate the relevance of various elements.

3. Failure to recognize limitations in the research evidence.

4. Misinterpretation due to unstudied factors.

5. Ignoring selective factors.

6. Difficulties of imperative evaluation.

7. In adequate technique of comparison etc.

Research Proposal and Report Writing

Topic Selection

The selection of a topic for research is a commitment of one’s time and efforts in a particular direction. There should not be any haste in deciding on the topic, nor in defining its scope.

The selection of a suitable topic for research is really a difficult task. There are many pitfalls to be avoided and there are many students who have failed to complete their research, not because they were lazy or badly organized but because their topic was not suitable for research. Probably, the most common mistake is to choose a topic that is too large at the level at which the student is studying/working: in their enthusiasm undergraduates/ postgraduates often consider undertaking research that would overawe an M.Phil. or M.Litt. or even Ph.D. Student. A topic, which may be suitable for a Doctoral thesis, may not be so for a dissertation for a graduate or postgraduate degree, which has to be completed in less than two semesters. Obviously, extensive research involving large samples and presenting considerable problems in terms of analysis can not usually be undertaken by a single undergraduate. Some topics are simply not researchable at any level. This is because the sources do not exist or the difficulty may arise because no means has yet been devised for investigating that particular field.

Time is another important factor for gathering information takes many months or even years, and then the topic is obviously not suitable for an undergraduate/post graduate student with only a few months to complete the project. It is advisable to those to confine themselves to research based on local sources.

A thesis paper is an investigation of an unsolved problem. An important consideration for the choice of a topic is that, it comes from an existing unsolved problem. The criterions followed to select a research topic are summarized as:

§  The student should choose the topic of his/her own interest. It can be easily judged through newspaper, article and library references: different case studies completed by the post-graduate students and researcher.

§  It should not be too vast beyond the level i.e. the field of study should not be complex. In general one should not choose more than two or three variables or factors for investigation.

§  The size of the topic should be manageable. Pick up the subject, which is not too broad or too small, so that it can be easily investigated in available resources (time, cost and manpower).

§  Choose the unexplored areas or new areas so as to obtain new concepts and ideas.

§  Discuss with experts and authorities including the supervisor or research advisor.

Preliminary Survey of Literature for Topic Selection

A preliminary survey of the literature on the topic should be carried out to find out the possibility of original contribution to the concerned area of knowledge. A study of the current literature in the chosen field will indicate the problems that are being investigated and will suggest further problems for investigation. The student should be familiar too with recent doctoral studies in the field of interest special assignments, term papers, dissertations and thesis usually conclude with suggestions for further research.

Generally, it is very difficult to know about the current research work in specific area. The possibility of such project topics being underway may be indicated by reading the professional journals and recent papers, discussion in workshops, conferences, and seminars. Encyclopaedia of social science research, dissertation abstracts, international and similar publications are rich sources for problem seekers. The major sources of preliminary information could be

§  Student’s own observation of the situation

§  Interviews and library source

§  Published data.

Once the field of interest is narrowed down and several problems/areas for possible investigation are identified, then the following questions should be asked about the topic as to its feasibility.

            ‘Does the topic really interest you?’

            ‘Is the problem a significant one?’

            ‘Is there adequate supervision/guidance?’

            ‘Can the topic be completed in the required time?’

 ‘Are the necessary equipment’s and adequate library facilities available?’

‘Are subjects available to experimental treatment and testing?’ etc.

 Research Proposal

A research proposal is a detailed plan of the research to be conducted. The research proposal is comparable to the blue print which the architect prepares before the work of building commences. The research proposal provides the basis for evaluation of the submitted project. Many institutions require that proposal must be submitted before it is finally approved. Researches often seek financial support for their research work by submitting grant proposals to government and even private agencies. The major purpose of research proposal is to ensure a workable experimental design which, when implemented, may result into analyzable and interpretable piece of research or finding of significant scientific merit.

Preparation of research report is an important aspect of research process. A researcher must be aware of what he is going to do, why he is doing and how he is going to perform it before attempting research activities. Unless a researcher is clear about what, why and how of the activities, he is going to perform, research work is almost impossible. Therefore, it is essential to throw light upon the nature and need for a research proposal.

Purpose of a Research Proposal

The purpose of a research proposal is to clearly communicate the following to the sponsor:

§  Need of the particular research

§  Benefits of the research

§  Type of data to be collected

§  Type of analysis that will be done

§  Need of help from other organizations

§  Duration, facilities and funds required to carry out the research.

§  Credentials of the proposers.

An important use of research proposal is that it helps the decision maker and the researcher to arrive at agreements on the problem with regard to objectives, information required and methods of analysis.

Types of Research Proposals

The type of research proposal depends on the type of project it deals with. It may range from a simple pilot study to a large complex project. It may originate from corporations research organization, academicians etc. One way of classifying research proposals is to consider them as internal proposals or external proposals.

Internal Proposals: These are proposals generated within an organization or agency and submitted to its management for approval or funding. They are responses to specify management needs of problem solving or product or process development and are funded internally. The emphasis is on solving the immediate problem or developing new product/process or modifying old ones. They do not emphasis in literature review. An executive summary is required in these proposals for quick management appreciation, schedule of funds and the frame for completion should also be included. Generally, project plans like Program Evaluation and Review Technique (PERT), Critical path method (CPM) charts are not required.

External Proposals: A proposal generated within an organization and directed to an outside customer organization or funding agency is an external proposal. This may be against an advertisement or solicitation from the customer. Then the proposal becomes a competing bid. It may also be unsolicited, in which case the proposing scientist or organization makes the proposal based on a perceived general/natural need after a preliminary/feasibility study. Usually, it aims at winking funding for research as in research institution/universities or winning contracts to generate profits as in industry.

In external proposals, objectives, detailed research design, credentials of the research scientist/team, and the budget become vital. In complex and large projects, a detailed project plan like PERT charts may be required but specifications of the funding agency must be met in this regard.

The external proposals can further be guided into solicited and unsolicited proposals. A solicited proposal is developed by an external researcher in response to a request for proposals. It is likely to complete against several others for the contract or grant. As such, it concentrates on a known problem and purpose an appropriate research design to address the problem. The problem statement is brief and to the point. The research objectives are known to the client and are therefore, briefly reviewed. Most of the proposal of this type concentrate on the research design and the distinguishing characteristics of the proposal i.e. qualification of the researchers, time schedule, budget, required resources and house-keeping arrangements.

An unsolicited proposal has the advantage of not competing against others but has the disadvantage of having to speculate on the ratification of a problem facing the firm's management. These proposals lay greater emphasis on the problem statement and the research objectives. The methods section of the proposal elaborated and explained in greater detail. In addition, the qualification of the researcher, time-schedule budget and required resources are mentioned.

Structure of the Research Proposal

The structure of proposal differs in their purpose and nature, tend to follow more or less in a defined pattern. Cooper and Schindler (2000) defined the following model for a research proposal structure.

§  Executive summary

§  Background information

§  Statement of the problem

§  Review of literature and theoretical framework

§  Objectives, research questions, Hypotheses

§  Significance, relevance and practical implication of the study.

§  Research methods and procedures

                        Research strategy

                        Research design

                        Population and sample

                        Method of data collection and analysis

§  Expected results

§  Research plan (House-keeping arrangements)

      Time frame

      Budget and resources

      Project management

      Equipment and instruments

§  Qualification of the researchers

§  References

§  Appendixes

Above mentioned contents are included in a general research proposal. However all these module contents may not be required for all types of research. A comparison of management/social science – oriented proposals and academic research proposal are shown in following table:

Normally, a written research proposal follows the general format of a journal article but headings of various sections are a bit different. The following nine steps are generally followed in preparing a research proposal (many institutions or agencies may suggest some other formats for the research proposals).

1.   Statement of the problem

2.   Definitions, assumptions, limitations or delimitations

3.   Review of related literature

4.   Statement of hypothesis/objectives

5.   Methodology

6.   Time schedule

7.   Expected result

8.   References

9.   Appendix

A brief content of proposal mostly adopted by Nepalese universities is given as follows:

Title of the Proposed Study

Chapter One

Introduction

1.1. Background of the study

1.2. Statement of the problem (Research questions)

1.3. Objectives of the study.

1.4. Conceptual framework.

1.5. Rationale of the study.

Chapter Two

Review of the Literatures

2.1. Theoretical reviews.

2.2. Reviews of the previous studies

(Organization of this chapter depends up on the research topic and its research questions).

Chapter Three

Research Methods

3.1 Rationale of the selection of the study area.

3.2 Research design

3.3 Nature and sources of data.

3.4 Universe and sampling

3.5 Data collection techniques.

3.6 Reliability and validity of the data.

3.7 Operational definition and measurement of selected concepts/ variables

3.8 Data processing and analysis.

3.9 Limitation of the study.

Reference

List of used materials in alphabetic order.

Note: The proposal must be submitted with questionnaires and checklist.

Proposals for Funded Research

The proposal for funded research involves the following additional information i.e. covering costs, personnel, time-frame, expected outputs and other related issues over an academic research proposal.

1.   Budget: A detailed breakdown of the funds required to complete the research work must be listed in the application forms issued by sponsors (or funding bodies).

2.   Equipment and instruments: A detailed list of all the equipment and instruments like computers, vehicles, camera, measuring and experimental equipments should be furnished with the budged required if these have to be hired or procured.

3.   Research personal and their qualifications: CVs of main member of research team should be enclosed with proposal to bit the competition based on status and experience expected by the sponsors.

4.   Time-frame: A detailed time-frame for each activity of research work is to be presented in bar chart and sequence of their occurrences.

Sometimes, funding agencies or institutions expects a specific structure for acceptance or rejection of proposal. A proposal prepared by following above guideline and content must meet the requirement of the sponsors.

Evaluation of Research Proposal

An evaluation of the research proposal by the research team/ organization is very essential before finalization. Sometimes, a pilot study is conducted to rehearse the draft proposal in order to find insights into the loopholes, flaws, inadequacies and possible improvements in the proposal. The proposal is modified particularly the sample, instruments and sometimes the scope of the study before submission to the supervisory authority.

In unsolicited proposals of single bidders the supervisory authority may evaluate the proposal and suggest modifications as per the suggestions of expert. In case of competitive budding, test of criteria for evaluation are developed and the bidders will be rated against them. Weightage factors may be provided for each criterion. Scores are obtained by multiplying the ratings and their weightage factors and summing them up. Whichever organization scores the highest is chosen for funding. An independent expert panel may be used for this purpose.

The sponsor should also make consideration in the following aspects before making final selection:

§  Neatness

§  Organization, in terms of being both logical and easily understood.

§  Completeness in filling up request for proposal (RFP) specification including budget and schedule.

§  Appointment of writing style.

§  Submission within the RFP's timeline. Although a proposal produced on a word processor and bound with an expensive cover will not overcome design or analysis deficiencies, a poorly presented unclear or disorganized proposal will not get serious attention from the reviewing sponsor.

Research Report

Presentation of research finding in the form of report is a necessary part of the research process. The main purpose of report is to convey to the interested individuals, the empiricists or the theoreticians the whole result of the study in sufficient details, so that new findings or new methods of analysis can be incorporated into the general store of knowledge available in the area. Any research inquiry begins with an idea and ends with a report.

Meaning and Definition

The word ‘report’ is derived from the Latin word ‘reportare’ which means to carry back. A report, therefore, is a description of an event carried back to someone who was not present on the scene. It is a formal communication written for a specific purpose, it includes a description of procedures followed for collection and analysis of data, their significance, the conclusions drawn from them, and recommendation if required.^1.

In other words, research report is a written integrated and systematic representation or communication of the investigation, facts or evidences, finding conclusions and suggestions/ recommendations of the research works. The objective of the research report is to communicate the exact detailed result of the research study to the audiences (readers or evaluators). Communicability demands clarify about the audiences for whom the given researches report is intended.

Types of Reports

The type of research and report are not limited however in business situation, there are various types of reports such as formal or informal, written or oral, internal or external, short or long, technical or popular, analytical or informational etc. The length and form of research are largely dictated by the problems at hand. For instance, business firms prefer reports in the letterform, comprising one or two pages in length. Banks, insurance, financial institutions keep short balance sheet type of tabulation for their annual reports to their customers and shareholders. Similarly mathematicians prefer to write the results of investigation in terms of algebraic notations, chemist report their results in symbols and formulae. Literatures usually write long reports presenting the critical analysis of some writer or period. Clinical psychologist and social pathologist frequently find it necessary to make use of the case history form. Also the news items in the daily papers book reviews etc are also forms of report writing.

The reports prepared by governmental bureaus, special commission and similar other organizations are very comprehensive reports on the issues involved.

But for thesis assignment purpose, the reports are categorized into descriptive and analytical form.

United Nations statistics office (1964) has approved three types of reports namely:

Preliminary report: A preliminary report giving the available data of current interest is required to presented as soon as possible. Such results may relate to characteristics of the survey population and variables. Sometimes these results are based on a sub sample of the full sample. The report contains a brief statement about survey method, size of the sample used, method of selecting sample and the limitation of the data.

General report: A general report includes information on:

§  Statement of purposes of the survey.

§  Description of the coverage.

§  Nature of the information collected.

§  The method of collection

§  The extent and causes of non-responses.

§  Whether the survey is an isolated one or is one of similar surveys.

§  Numerical results.

§  The period to which data refers or reference period.

§  Period of data collection i.e. time taken for the fieldwork.

§  Accuracy

§  Cost

§  Assessment of the fulfilment of the purposes of survey.

§  Names of the organization conducting and sponsoring

§  References to available reports or papers relating to the survey.

Technical report: In technical report more emphasis is given to methodological part of the study/survey and the detailed presentation of the findings including their limitation and supporting data.

The report should include information on:

§  Specification of the frame (list of population)

§  Sample design

§  Personnel and equipment used

§  Statistical analysis and computational procedures.

§  Accuracy of the results including sampling non-sampling errors.

§  Comparison with other sources of information.

§  Cost analysis

§  Efficiency of sample design.

§  Observations of technicians.

In addition to U.N. guidelines, the various types listed above can be categorized under descriptive and analytical. A descriptive report includes the descriptions of facts or opinions, their presentation in an organized form and analysis in understanding form. These reports indicate the nature of the problems facing the organization under study and also indicate the reforms required to overcome the problems.

An analytical report presents the facts and statistics, interpret these information in relation to the problem under consideration. This type of report follow the process of scientific investigation and reporting of the problems focused in the selected area of inquire with recommendation of proper course of action for improvement. A report is sometimes an amalgam of both descriptive and analytical form.

Procedure for Writing Reports

Written reports themselves are of different types and their planning, drafting, and documentation vary with the types, purpose, and readers. The report for the intelligent layman is intended very largely to disseminate the broad facts, findings and recommendations. It must be lucid, simple and yet dignified in summary form along with a full technical report.

There is no explicit rule for research report writing. The researcher as the research report writer should plan and organize the report writing in the process or completion of research workings in a detailed outline. The common procedures are illustrated subsequently.

Preparation of outline: On the basis of collected facts/data, an outline of reporting prior to actually starting of the writing report is drawn. This will help the researcher in arranging ideas in a comprehensive way to identify the points related to one other.

Planning of time:  A researcher has to plan about the entire research process in advance. To meet the required time schedule, it is always better for the researcher to plan the available time accordingly. Time planning may save the researcher from these inconveniences and last minute to rush. It is also better to keep margin of time. If any unforeseen problem causes a delay, the researcher can prepare the report within the stipulated time.

Arrangement of data: The collected data are generally in raw form and are processed and tabulated before starting the analysis with usual edit work; the analysis of data is to be made by choosing proper method. Thus, it can be said that the ‘better organization and arranges the data, the better the data will be revealing.

Drafting reports: Three steps are involved in drafting a report irrespective of the nature of audience. In each case the first, second and final drafts writing may have to be done quite a few times but each of these reports are having their own distinguishing features.

First Draft

A first draft may be written a number of times and still it continues to remain only a working draft. There are three stages in writing the first draft, namely, to weave the material together for making clear connections, to assure the investigator him/herself of a satisfactory organization and fullness of the facts, and to avoid blank paper fright that may be present in every first research. While preparing first draft one should adopt the following steps:

§  Shuffle the notes under each sub-topic, and put them in sequence.

§  Expand the notes and outlines into manuscript and read own notes to draw out their full significance.

Second Draft

After a lapse of sometimes from the completion of first draft, make revision for writing second draft concentrating largely on form and language. The researcher should give the first draft, a lucid and clear shape so that it can be readable. Finally, at the second draft stage, critical evaluation will have to be made of all that has been written facts, findings, conclusions and recommendations. At the end of the second draft stage and even all through the writing of the draft, the researcher may do well to ask him/herself and answer a few questions.

‘Does the title cover the scope of the subject?’

‘Is the initial hypothesis tested adequately?’

‘Does the beginning begin and conclusion concludes?’

‘Are the opening sentences, paragraphs, sections etc attractive?’

Finally, ‘Is the researcher, as his/her own critic, satisfied with the report?

Third Draft

The final stage in drafting is the preparation of final report. It concentrates mainly on documentation and polish to make the research weighty, authoritative, convincing and attractive. A good research paper depends not only upon the amount of readings or notes taken or upon the form of presentation but also upon the accurate and thorough investigation.

Make Direct and Positive Sentences

The researcher should not use unessential words and phrases. Use of unnecessary long, technical or unusual words or phrases should be avoided. In writing a report, the researcher should write naturally and directly using familiar words, short sentences with simple constructions.

Presentation

Presentation of each step and section should be in sequential order. Charts, sections, tables etc should be labelled adequately. The system of presentation should be simple and logical. The report must be a coherent whole; it must be tightly woven fabric of facts and ideas of sections, subsections, paragraphs and sentences.

The report is to be written in the third person. The introduction, data analysis and findings are to be in the past tense: conclusion in the present tense and recommendations in the future tense. The pronouns such as I, my, mine, our, we, us and me should be eliminated from the report. If required to refer to the researcher, ‘the writer’ or ‘the investigator’ should be used.

Precautions for Writing Research Report

Research report is a channel of communicating the research findings to the readers of the report. A good report must be prepared efficiently and effectively. The general purpose of a research report is not to convince the readers but to let them know what has been done, why it was done, what results were obtained etc. Thus, the research report aim at telling the readers the problems investigated, the methods adopted, the results found and conclusion reached. The research report should be written in a clear and unambiguous language so that, the reader can also judge the adequacy and the validity of the research. For attaining objectivity person pronouns such as I, you, we, my, out etc. should be avoided and as their substitutes expressions like ‘investigator’, ‘researcher’ etc. should be used. While preparing a report one should keep the following precautions in view:

§  The length of report should be long enough to cover the subject and short enough to maintain interest. In fact, report writing should not be a means to learning more and more about less and less.

§  The report should be written in an objective style in simple language, avoiding expressions such as ‘it seems’, ‘there may be’ and the like.

§  The layout of the report should be well thought out and must be appropriate in accordance with the objective of the research problem.

§  Objective of the study, the nature of the problem, the methods employed, and the analysis techniques adopted must be clearly stated in the beginning of the report in the form of introduction.

§  In addition to the summary of important findings, charts, graphs, and statistical tables may be used for various results.

The style of writing report

While writing the research report, the researcher has to strictly comply with a set of mechanical rules related to simplicity and objectivity, use of quantitative terms, mistake-proof writing, footnote citations, use of abbreviations, layout for headings and pagination.

Simplicity and objectivity: The researcher must present his/her report in a creative, clear and concise form by avoiding slang, flippant phrases and folksy style. Every care must be taken to ensure objectivity without being influenced by any element of exhortation or persuasion. As far possible, the researcher is required to explain rather than convince the readers. The researcher should avoid the use of such personal terms as ‘I’, ‘we’, ‘my’, ‘our’, ‘you’, ‘us’ etc. Instead the term ‘researcher’ or ‘investigator’ may be used. Likewise, they should also avoid the use of terms such as ‘Professor’, ‘Dr.’, ’Mr./Mrs.’ etc. It would be appropriate to use the last name of the author like, Joshi, Sharma, Gupta etc. There should be use of past tense in describing the research procedures which have been completed. The abbreviations can only be used in the footnotes, tables and figures but not in the text.

Use of quantitative terms: While using quantitative terms, such expressions like ‘less in quantity’, ‘few in number’, ‘low response’ etc. are desirable. The numbers appearing at the beginning of the sentence must be spelled out (i.e. expression of first quantity must be in words rather than value).  Likewise, attempt may be made to spell out fractions, except when they are combined ( i.e. one-half rather than ½, 2.5 rather than two and one-half). Also the percentage value must be spelled except in the tables and figures. But in text, they need to express the percent figure in terms 10 percent (say) rather than 10%. Commas are also used to point off thousands or millions (8,848, 1,25,235 etc.) if the numbers involve more than three digits. The researcher should indicate only unusual statistical formula in the report and not the standard ones.

Use of good dictionary for writing without errors: The researcher should make use of standard and latest dictionary to make correct usage. Attempt should be made to avoid spelling mistakes and inconsistencies. The manuscript must be given to a competent proof reader to ensure correct usage before giving it for final typing to a highly proficient typist. Every effort should be made to avoid the mistakes stemming from one’s own carelessness. Indeed, an effective research report goes to several revisions before it is ready for publication. Now a day, use of computer minimizes these difficulties and precautions.

Footnote citation: Footnotes may be used to present exploratory statements which usually interfere with the continuity of the textual material. Usually, the researcher may place footnote citations at the bottom of the page separating them by a two-inch horizontal line drawn from the left margin, with one double space below the last line of the text. They are single-spaced, with double spaces between citations.

Pagination: the page numbers should be indicated at each page of the report. Although a number is allowed for each division (chapter, major sub-division, bibliography or appendix), it not typed on the page. The researcher may indicate page numbers in the upper right hand corner, one inch below the top of the page and aligned with the right margin. Indeed, the page numbers are assigned on the final edited copy before giving the manuscript for binding. Initially, pagination can be done with light pencil.

To sum up, it is a very difficult task to write the report free from error. The researcher has to go on revising it constantly by reorganising and pairing, deleting redundancies and allied verbal facts. the report should be examined repeatedly with clarity and brevity.

Format of the Research Report

A format (layout) of research report includes the contents of research report. There is no uniformity about the format of research report. The structure varies from person to person and place to place. In general the format of research is categorized as:

1.      Preliminaries

2.      The main body or the text

3.      The end section/reference material.

Preliminary Section

The preliminaries consist of the following components.

§  The title page

§  Acceptance page

§  Acknowledgement page (or preface)

§  Table of contents

§  List of tables

§  List of figures (and illustrations)

Main Body/Text

As the name highlights, it is the nucleus of the research work and is usually classified into three sub-sections.

§  Introduction (introductory chapters)

§  Main body of the report

§  Conclusions (summary, recommendations/suggestions)

Reference Material

§  Bibliography

§  Appendix

§  Glossary of terms (if any)

§  Index (if any) and

§  Vita (if required)

REFERENCE MATERIAL

A.        Bibliography: The bibliography must include all published or unpublished sources involved in finding the references. The general entries in bibliography should be made either in APA (American Psychological Association) or MLA (Modern Language Association). The APA style of citation is illustrated in the following order:

For books

1.         Name of author last name first or issuing body.

2.         Title of the publication/book (underlined to indicate Italics)

3.         Place, publisher, and date of publication

4.         Number of volumes (in Roman Letter)

5.         Page number

Example: For single author

Joshi, P. R. (2016). Business Research Methods. Kathmandu : Samjhana Publication

For two authors

Cooper, D & Schindler, P (2003). Business Research Methods, 2^nd edition. New York: Tata Mc Graw-Hill.

For more than  two authors

Joshi, P. R. et.al (2012). Operations Management. Kathmandu: Buddha Academic Enterprises Pvt. Ltd.

For magazines and Newspapers or serial publications.

1.   Name of the author, (Sur name first)

2.   Title of article (in quotation marks)

3.   Name of periodicals (underlined to indicate italics)

4.   The volume or volume and Number

5.   Date of the issue

6.   The pagination

Example:

Robert V. Roosa, ‘Coping with short-term International Money Flows’, The Banker, London, September, 1971 P. 995

For Thesis

1.   Author (Surname first)

2.   Full title of the Thesis (in quotation marks)

3.   The course of degree

4.   The name of the university.

Example:

Singh, M.L.(1979). Population Dynamics of Nepal-An analytic study. T.U., Kathmandu.

Remarks: However, recent practice is to write, name of the author starting from last name, year of publication in bracket, title of the Article or work, name of the journal or publication, place of publication and page.

The reference are to be presented sequentially as they appear on the text, Bibliography should be presented in the following groups,

1.   Books

2.   Journals/Articles

3.   Newspapers

4.   Published documents

5.   Unpublished documents etc.

---