What is validity? How does it differ from reliability and what are its types?

Validity: A measurement scale may be considered to be valid if it effectively measures a specific property or characteristic that it intends to measure.

The question of validity does not arise in the case of measurement of physical characteristics such as length, weight and height. This is because the measurement is direct and can be done through standard measuring devices. 

Types of Validity: Following are the main types of Validity:

1. Content Validity: This type of validity may be of two types – a) Face validity and b) Sampling validity. Face validity is determined through a subjective evaluation of a measuring scale. However, the limitation of this type of validity is that it is determined by opinions, rather than through a statistical method. Sampling validity refers to how representative the content of the measuring instrument is. In other words, the measuring instrument’s content must be representative of the content universe of the characteristic being measured.
2. Predictive Validity: This type of validity refers to the extent to which one behavior can be predicted based on another, based on the association between the results yielded by the measuring instrument and the eventual outcome. One limitation of determining predictive validity using this statistical association is that the eventual outcome, in this case, the grade point average of students during the first semester, may be influenced by other “extraneous” variables or factors. In other words, the grade point average may have been influenced by other factors and may not necessarily be linked to the score on the admission test. Therefore, predicting behavior from one situation to another is not always accurate.
3. Construct Validity: A construct is a conceptual equation that is developed by the researcher based on theoretical reasoning. Various kinds of relationships may be perceived by the researcher between a variable under study and other variables. These relationships must be tested in order to determine the construct validity of a measuring instrument. The instrument may be considered to have construct validity only if the expected relationships are found to be true.

Reliability: Reliability refers to the ability of a measuring scale to provide consistent and accurate results.

For example, a weighing machine may be said to be reliable if the same reading is given every time the same object is weighted. 

There are two dimensions of reliability – Stability and Equivalence or Non-Variability.

Stability refers to consistency of results with repeated measurements of the same object, as in the weighing machine example. Equivalence or Non-Variability refers to consistency at a given point of time among different investigators and samples of items.

Reliability can be improved in three ways–

1. By reducing the external sources of variation. This in turn can be achieved by standardizing the conditions under which measurement is carried out.
2. By making the measuring instrument more consistent internally, through an analysis of the different items.
3. By adding more number of items to the measuring instrument, in order to increase the probability of more accurate measurement.

The significance of research in social and business sciences

According to a famous Hudson Maxim, “All progress is born of inquiry. Doubt is often better than overconfidence, for it leads to inquiry, and inquiry leads to invention”. It brings out the significance of research, increased amounts of which make progress possible. Research encourages scientific and inductive thinking, besides promoting the development of logical habits of thinking and organization. 

The role of research in applied economics in the context of an economic or business is greatly increasing in modern times. The increasingly complex nature of government and business has raised the use of research in solving operational problems. Research assumes a significant role in the formulation of economic policy, for both the government and business. It provides the basis for almost all government policies of an economic system. Government budget formulation, for example, depends particularly on the analysis of the needs and desires of the people, and the availability of revenues, which requires research. Research helps to formulate alternative policies, in addition to examining the consequences of these alternatives. Thus, research also facilitates the decision making of policy-makers, although in itself it is not a part of research. In the process, research also helps in the proper allocation of a country’s scare resource. Research is also necessary for collecting information on the social and economic structure of an economy to understand the process of change occurring in the country. Collection of statistical information though not a routine task, involves various research problems. Therefore, a large staff of research technicians or experts are engaged by the government these days to undertake this work. Thus, research as a tool of government economic policy formulation involves three distinct stages of operation which are as follows:

• Investigation of economic structure through continual compilation of facts;
• Diagnoses of events that are taking place and the analysis of the forces underlying them; and;
• The prognosis, i.e., the prediction of future developments. 

Research also assumes a significant role in solving various operational and planning problems associated with business and industry. In several ways, operations research, market research, and motivational research are vital and their results assist in taking business decisions. Market research refers to the investigation of the structure and development of a market for the formulation of efficient policies relating to purchases, production and sales. Operational research relates to the application of logical, mathematical, and analytical techniques to find solutions to business problems such as cost minimization or profit maximization, or the optimization problems. Motivational research helps to determine why people behave in the manner they do with respect to market characteristics. More specifically, it is concerned with the analyzing the motivations underlying consumer behavior. All these researches are very useful for business and industry, which are responsible for business decision making.

Research is equally important to social scientist for analyzing social relationships and seeking explanations to various social problems. It gives the intellectual satisfaction of knowing things for the sake of knowledge. It also possesses practical utility for the social scientist to gain knowledge so as to be able to do something better or in a more efficient manner. Thus, research in the social sciences is concerned with both knowledge for its own sake, and knowledge of what it can contribute to solve practical problems.

Types of Research

Although any typology of research is inevitably arbitrary, Research may be classified crudely according to its major intent or the methods. According to the intent, research may be classified as: 

Pure Research - It is undertaken for the sake of knowledge without any intention to apply it in practice, e.g., Einstein’s theory of relativity, Newton’s contributions, Galileo’s contribution, etc. It is also known as basic or fundamental research. It is undertaken out of intellectual curiosity or inquisitiveness. It is not necessarily problem-oriented. It aims at extension of knowledge. It may lead to either discovery of a new theory or refinement of an existing theory. It lays the foundation for applied research. It offers solutions to many practical problems. It helps to find the critical factors in a practical problem. It develops many alternative solutions and thus enables us to choose the best solution.

Applied Research - It is carried on to find a solution to a real-life problem requiring an action or policy decision. It is thus problem-oriented and action-directed. It seeks an immediate and practical result, e.g., marketing research carried on for developing a new market or for studying the post-purchase experience of customers. Though the immediate purpose of an applied research is to find solutions to a practical problem, it may incidentally contribute to the development of theoretical knowledge by leading to the discovery of new facts or testing of theory or o conceptual clarity. It can put theory to the test. It may aid in conceptual clarification. It may integrate previously existing theories.

Exploratory Research - It is also known as formuletive research. It is a preliminary study of an unfamiliar problem about which the researcher has little or no knowledge. It is ill-structured and much less focused on pre-determined objectives. It usually takes the form of a pilot study. The purpose of this research may be to generate new ideas, or to increase the researcher’s familiarity with the problem or to make a precise formulation of the problem or to gather information for clarifying concepts or to determine whether it is feasible to attempt the study. Katz conceptualizes two levels of exploratory studies. “At the first level is the discovery of the significant variable in the situations; at the second, the discovery of relationships between variables.”

Descriptive Study - It is a fact-finding investigation with adequate interpretation. It is the simplest type of research. It is more specific than an exploratory research. It aims at identifying the various characteristics of a community or institution or problem under study and also aims at a classification of the range of elements comprising the subject matter of study. It contributes to the development of a young science and useful in verifying focal concepts through empirical observation. It can highlight important methodological aspects of data collection and interpretation. The information obtained may be useful for prediction about areas of social life outside the boundaries of the research. They are valuable in providing the facts needed for planning social action program.

Diagnostic Study - It is similar to descriptive study but with a different focus. It is directed towards discovering what is happening, why it is happening and what can be done about. It aims at identifying the causes of a problem and the possible solutions for it. It may also be concerned with discovering and testing whether certain variables are associated. This type of research requires prior knowledge of the problem, its thorough formulation, clear-cut definition of the given population, adequate methods for collecting accurate information, precise measurement of variables, statistical analysis and test of significance.

Evaluation Studies - It is a type of applied research. It is made for assessing the effectiveness of social or economic programs implemented or for assessing the impact of developmental projects on the development of the project area. It is thus directed to assess or appraise the quality and quantity of an activity and its performance, and to specify its attributes and conditions required for its success. It is concerned with causal relationships and is more actively guided by the hypothesis. It is concerned also with change over time.

Action Research - It is a type of evaluation studies. It is a concurrent evaluation study of an action program launched for solving a problem for improving an existing situation. It includes six major steps: diagnosis, sharing of diagnostic information, planning, developing change program, initiation of organizational change, implementation of participation and communication process, and post experimental evaluation

The Components of a Research Design.

It is important to be familiar with the important concepts relating to research design. They are:

1. Dependent and Independent Variables: A magnitude that varies is known as a variable. The concept may assume different quantitative values, like height, weight, income, etc. Qualitative variables are not quantifiable in the strictest sense of objectivity. However, the qualitative phenomena may also be quantified in terms of the presence or absence of the attribute considered. Phenomena that assume different values quantitatively even in decimal points are known as ‘continuous variables’. But, all variables need not be continuous. Values that can be expressed only in integer values are called ‘non-continuous variables’. In statistical terms, they are also known as ‘discrete variable’. For example, age is a continuous variable; whereas the number of children is a non-continuous variable. When changes in one variable depend upon the changes in one or more other variables, it is known as a dependent or endogenous variable, and the variables that cause the changes in the dependent variable are known as the independent or explanatory or exogenous variables. For example, if demand depends upon price, then demand is a dependent variable, while price is the independent variable. And if, more variables determine demand, like income and prices of substitute commodity, then demand also depends upon them in addition to the own price. Then, demand is a dependent variable which is determined by the independent variables like own price, income and price of a substitute.
2. Extraneous Variable: The independent variables which are not directly related to the purpose of the study but affect the dependent variable are known as extraneous variables. For instance, assume that a researcher wants to test the hypothesis that there is a relationship between children’s school performance and their self-concepts, in which case the latter is an independent variable and the former, the dependent variable. In this context, intelligence may also influence the school performance. However, since it is not directly related to the purpose of the study undertaken by the researcher, it would be known as an extraneous variable. The influence caused by the extraneous variable on the dependent variable is technically called as an ‘experimental error’. Therefore, a research study should always be framed in such a manner that the dependent variable completely influences the change in the independent variable and any other extraneous variable or variables.
3. Control: One of the most important features of a good research design is to minimize the effect of extraneous variable. Technically, the term control is used when a researcher designs the study in such a manner that it minimizes the effects of extraneous independent variables. The term control is used in experimental research to reflect the restrain in experimental conditions.
4. Confounded Relationship: The relationship between dependent and independent variables is said to be confounded by an extraneous variable, when the dependent variable is not free from its effects.
5. Research Hypothesis: When a prediction or a hypothesized relationship is tested by adopting scientific methods, it is known as research hypothesis. The research hypothesis is a predictive statement which relates a dependent variable and an independent variable. Generally, a research hypothesis must consist of at least one dependent variable and one independent variable. Whereas, the relationships that are assumed but not be tested are predictive statements that are not to be objectively verified are not classified as research hypothesis.
6. Experimental and Control Groups: When a group is exposed to usual conditions in an experimental hypothesis-testing research, it is known as ‘control group’. On the other hand, when the group is exposed to certain new or special condition, it is known as an ‘experimental group’. In the aforementioned example, the Group A can be called a control group and the Group B an experimental one. If both the groups A and B are exposed to some special feature, then both the groups may be called as ‘experimental groups’. A research design may include only the experimental group or the both experimental and control groups together.
7. Treatments: Treatments refer to the different conditions to which the experimental and control groups are subject to. In the example considered, the two treatments are the parents with regular earnings and those with no regular earnings. Likewise, if a research study attempts to examine through an experiment regarding the comparative impacts of three different types of fertilizers on the yield of rice crop, then the three types of fertilizers would be treated as the three treatments.
8. Experiment: An experiment refers to the process of verifying the truth of a statistical hypothesis relating to a given research problem. For instance, the experiment may be conducted to examine the yield of a certain new variety of rice crop developed. Further, Experiments may be categorized into two types namely, absolute experiment and comparative experiment. If a researcher wishes to determine the impact of a chemical fertilizer on the yield of a particular variant of the rice crop, then it is known as absolute experiment. Meanwhile, if the researcher wishes to determine the impact of chemical fertilizer as compared to the impact of bio-fertilizer, then the experiment is known as a comparative experiment.
9. Experiment Unit: Experimental units refer to the predetermined plots, characteristics or the blocks, to which the different treatments are applied. It is worth mentioning here that such experimental units must be selected with great caution.

The Sampling Process.

The sampling techniques may be broadly classified into:

• Probability Sampling
• Non-Probability Sampling 

1. Probability Sampling: Probability sampling provides a scientific technique of drawing samples from the population. The technique of drawing samples is according to the law in which each unit has a probability of being included in the sample.

• Simple random sampling: Under this technique, sample units are drawn in such a way each and every unit in the population has an equal and independent chance of being included in the sample. If a sample unit is replaced before drawing the next unit, then it is known as simple Random Sampling with Replacement. If the sample unit is not replaced before drawing the next unit, then it is case, probability of drawing a unit is 1/N, where N is the population size. In the case probability of drawing a unit is 1/Nn.
• Stratified random sampling: This sampling design is most appropriate if the population is heterogeneous with respect to characteristic under study or the population distribution is highly skewed.
• Systematic sampling: This design is recommended if we have a complete list of sampling units arranged in some systematic order such as geographical, chronological or alphabetical order.
• Cluster sampling: The total population is divided into recognizable sub-divisions, known as clusters such that within each cluster they are homogenous. The units are selected from each cluster by suitable sampling techniques.
• Multi-stage sampling: The total population is divided into several stages. The sampling process is carried out through several stages.

    2. Non-Probability Sampling: Depending upon the object of inquiry and other considerations a predetermined number of sampling units is selected purposely so that they represent the true characteristics of the population.

• Judgment sampling: The choice of sampling items depends exclusively on the judgment of the investigator. The investigator’s experience and knowledge about the population will help to select the sample units. It is the most suitable method if the population size is less.
• Convenience sampling: The sampling units are selected according to convenience of the investigator. It is also called “chunk” which refer to the fraction of the population being investigated which is selected neither by probability nor by judgment.
• Quota sampling: It is a type of judgment sampling. Under this design, quotas are set up according to some specified characteristic such as age groups or income groups. From each group a specified number of units are sampled according to the quota allotted to the group. Within the group the selection of sampling units depends on personal judgment. It has a risk of personal prejudice and bias entering the process. This method is often used in public opinion studies.

The criteria of a good research problem.

Hornet and Hunt have given following characteristics of scientific research:

1. Verifiable Evidence: That is factual observations which other observers can see and check.
2. Accuracy: That is describing what really exists. It means truth or correctness of a statement or describe things exactly as they are and avoiding jumping to unwarranted conclusions either by exaggeration or fantasizing.
3. Precision: That is making it as exactly as necessary, or giving exact number or measurement. This avoids colorful literature and vague meanings.
4. Systematization: That is attempting to find all the relevant data, or collecting data in a systematic and organized way so that the conclusions drawn are reliable. Data based on casual recollections are generally incomplete and give unreliable judgments and conclusions.
5. Objectivity: That is free being from all biases and vested interests. It means observations are unaffected by the observers values, beliefs and preferences to the extent possible and he is able to see and accept facts as they are, not as he might wish them to be.
6. Recording: That is jotting down complete details as quickly as possible. Since human memory is fallible, all data collected are recorded.
7. Controlling Conditions: That is controlling all variables except one and then attempting to examine what happens when that variable is varied. This is the basic technique in all scientific experimentations – allowing one variable to vary while holding all other variables constant.
8. Training Investigators: That is imparting necessary knowledge to investigators to them understand what to look for, how to interpret in and avoids inaccurate data collection.

Why literature Survey is important in Research?

Doing a literature survey before you begin your investigation enables you to take advantage of the unique human capacity to pass on detailed written information from one generation to another. Reading all the knowledge that's accumulated so far on the problem you want to study can be time-consuming and even tedious. But careful evaluation of that material helps make your investigation worthwhile by alerting you to knowledge already gained and the problems already encountered in your areas of interest.

A literature survey amounts to reading available material on a given topic, analyzing and organizing findings, and producing a summary. There are many sources for literature reviews, including journals of general interest in each discipline, such as the American Political Science Review. There are also journals for specific topics such as the Leadership and Organization Development Journal. Governments publish great quantities of data on many topics. The United Nations and the United States Government Printing Office are two major sources. In addition, businesses and private organizations gather and publish information you might find useful. For certain problems you may want to search through popular or non-scholarly periodicals as well. While it's customary to include only data from sources that actually research the problem in a precise fashion, articles in more popular sources may provide interesting insight or orientations. Talking to knowledgeable people may also give you information that helps you formulate your problem. 

Thoroughness is the key. Most libraries have staff trained in information retrieval that can help find sources and suggest strategies to review the literature. The Internet, of course, now allows easy access to limitless information on given topics. Thoroughness in your review means not only finding all current publications on a topic but locating earlier writing as well. There's no easy rule for how long ago literature was published on your topic. The time varies from problem to problem. A useful way to locate past as well as current writing is to begin with the most current sources likely to contain relevant material. Then, follow these authors' footnotes and bibliographies. At some point in this search you'll find the material is beginning to be only peripherally related to your current interest or that author’s claim originality in their work. 

Of course, doing a good literature survey is easier when you know a great deal about the subject already. In such a case you'd probably be familiar with publications and even other people who do research in your area of interest. But for the novice, efficient use of library/Internet services and organizing how they check sources are especially important skills. 

Having located literature, keeping a checklist of useful information will help you read each source. You might ask yourself, particularly for research articles:

1. What was the exact problem studied?
2. How were the topics of interest defined?
3. What did the authors expect to find?
4. How were things measured?
5. What research did this author cite? Have you read it?
6. Who were the subjects of study?
7. What do the results show?
8. Do the data presented agree with the written conclusions?
9. What were the limitations of the study? 

A thorough literature survey should demonstrate that you've carefully read and evaluated each article or book. Because research reports can be tedious and difficult to understand for new researchers, many tend to read others' conclusions or summaries and take the author's word that the data actually support the conclusions. Careful reading of both tables and text for a while will convince you they don't always agree. Sometimes data are grossly misinterpreted in the text, but on other occasions authors are more subtle. Consider, for example, the following statements: 

Fully 30 percent of the sample said they did not vote. Only 30 percent of the sample said they did not vote. 

The percentage is the same, but the impression conveyed is decidedly different. Reading the actual data before accepting the author's conclusions will help prevent some of these errors of interpretation from creeping into your own research.

It's important that after you finish your reading, you're able to write your literature survey in a way that's clear, organizing what you know about the content and methods used to study your problem. You may find it helpful to record information about each source on a separate card or piece of paper so that information can later be reshuffled, compared, and otherwise reorganized. Note in most journal articles that what probably began as a long literature survey is usually condensed on the first few pages of the research report, explaining previous research on the problem and how the current study will contribute. You, too, want to add to this growing body of knowledge we call social science by a creative summary of what's been accomplished by others as well as by your own research.

The Different Types of Research Designs.

Historical Research Design - The purpose is to collect, verify, synthesize evidence to establish facts that defend or refute your hypothesis. It uses primary sources, secondary sources, and lots of qualitative data sources such as logs, diaries, official records, reports, etc. The limitation is that the sources must be both authentic and valid.

Case and Field Research Design - Also called ethnographic research, it uses direct observation to give a complete snapshot of a case that is being studied. It is useful when not much is known about a phenomenon. Used in few subjects.

Descriptive or Survey Research Design - It attempts to describe and explain conditions of the present by using many subjects and questionnaires to fully describe a phenomenon. Survey research design /survey methodology is one of the most popular for dissertation research. There are many advantages.

Correlation or Prospective Research Design - It attempts to explore relationships to make predictions. It uses one set of subjects with two or more variables for each.

Causal Comparative or Ex Post Facto Research Design - This research design attempts to explore cause and affect relationships where causes already exist and cannot be manipulated. It uses what already exists and looks backward to explain why.

Developmental or Time Series Research Design - Data are collected at certain points in time going forward. There is an emphasis on time patterns and longitudinal growth or change.

Experimental Research Design - This design is most appropriate in controlled settings such as laboratories. The design assumes random assignment of subjects and random assignment to groups (E and C). It attempts to explore cause and affect relationships where causes can be manipulated to produce different kinds of effects. Because of the requirement of random assignment, this design can be difficult to execute in the real world (non laboratory) setting.

Quasi Experimental Research Design - This research design approximates the experimental design but does not have a control group. There is more error possible in the results.