The distinction between methodology and method is best expressed as:

Methodology is the logic that justifies the choice of techniques; a method is a specific technique of collecting or analysing data

The two terms mean exactly the same thing

A method is the philosophy of research; methodology is the questionnaire

Methodology is the logic that justifies the choice of techniques; a method is a specific technique of collecting or analysing data

Methodology refers to qualitative research and method to quantitative research

Which of the following statements about Weber's position on values in social research are correct? Select all that apply.

Values legitimately influence the choice of research problem

Social science can and should be entirely unconnected with values at every stage

Values should determine what the findings are

Values legitimately influence the choice of research problem

Karl Popper held that what makes a theory scientific is its:

Verification by many confirming instances

Acceptance by the scientific community

Assertion (A): In hypothesis testing, researchers speak of rejecting or failing to reject the null hypothesis rather than of proving the research hypothesis. Reason (R): Statistical evidence can show only that data would be unlikely if the null were true; it cannot establish a hypothesis with certainty. Choose the correct option.

Both A and R are true, and R is the correct explanation of A

Both A and R are true, but R is not the correct explanation of A

Which of the following correctly match a type of research with its description? Select all that apply.

Longitudinal — studying the same phenomenon over a period of time

Exploratory — gaining familiarity with a little-understood problem and formulating hypotheses

Explanatory — testing cause-and-effect relationships

Longitudinal — studying the same phenomenon over a period of time

Exploratory — gaining familiarity with a little-understood problem and formulating hypotheses

Explanatory — testing cause-and-effect relationships

Cross-sectional — following one cohort for twenty years

"Grounded theory", in which theory is built inductively from data rather than tested deductively, was developed by:

Donald Campbell and Julian Stanley

Which of the following is a directional hypothesis?

Districts with longer-serving Collectors achieve higher coverage of the scheme

The tenure of Collectors has no effect on scheme coverage

The tenure of Collectors is related to scheme coverage

What explains variation in scheme coverage across districts?

Research Methodology I: Social Science Research; Methodology and Method; Facts and Values; Research and Theory-Building; the Scientific Method and Objectivity; Types of Research; the Research Problem; Hypotheses, Null Hypotheses and their Validation

Unit X reads: "Research Methodology: Social Science Research- Meaning and Significance; Distinction between Methodology and Method; Facts and Values in Research; Role of Research in Theory-Building; Scientific Method; Objectivity in Social Research; Types of Research; Identification of Research Problem; Hypotheses and Null- Hypotheses; Validation of Hypothesis; Research Design; Methods of Data Collection- Primary and Secondary sources- ( Observation; Questionnaire and Interview, Use of Library and Internet);Sampling and Sampling Techniques; Scales of Measurement; Analysis of Data and Use of Computers in Social Science Research- SPSS; Citation patterns and Ethics of Research ; Bibliography; Report Writing." This chapter takes the logic of inquiry — what research is, how it relates to theory and values, and how a question becomes a testable hypothesis; the next takes design, data, sampling, measurement, analysis and reporting. The examples are drawn from public administration, because the fact–value problem, in particular, is one this discipline has argued about since Simon and Waldo.

1. Social science research: meaning, significance, methodology and method

Research is systematic, controlled, empirical and critical inquiry into a question, carried out so that its findings can be checked by others. Social science research studies human behaviour, relationships and institutions — which makes it harder than natural science because its subjects think, react to being studied, and differ across cultures (the Hawthorne studies are the classic reminder). In public administration, research explains how institutions work, evaluates whether policies achieve their goals, tests theories such as those of motivation or implementation, and informs reform — the "evidence-based policy" movement rests on it.

Method and methodology are often confused. A method is a specific technique for collecting or analysing data — a survey questionnaire, participant observation, a regression, a case study. Methodology is the study of methods: the logic and philosophy that justify choosing particular methods for a particular question — the assumptions about what exists and can be known (ontology and epistemology) that separate, for example, a positivist from an interpretivist. Methodology asks why a method is appropriate; method is how the data are gathered. A researcher studying citizens' trust in the police may use the method of a structured survey, justified by a positivist methodology that treats trust as a measurable attitude — or in-depth interviews, justified by an interpretive methodology that treats trust as a meaning to be understood.

2. Facts and values; objectivity in social research

A fact is an empirically verifiable statement about what is; a value is a judgement about what ought to be. David Hume's "is–ought" gap holds that no amount of factual statement can by itself yield a value conclusion. Max Weber distinguished value-relevance — values inevitably guide the choice of research problem — from value-freedom (Wertfreiheit) in the conduct of research: once a problem is chosen, the researcher should report findings without letting preferences distort them, and should not present personal value judgements as scientific conclusions. Herbert Simon built this distinction into administrative theory: every decision has factual and value premises, and a science of administration can deal only with the factual. Gunnar Myrdal argued that since values cannot be kept out, the honest course is to state one's value premises explicitly at the start. Critics from Waldo onward insist that in public administration values — equity, accountability, efficiency — are part of the subject matter itself.

Objectivity means that findings reflect the object studied rather than the researcher's biases. In social research complete objectivity is difficult, because the researcher is part of the society studied, and because of personal bias, the influence of the observer on the observed, the sponsor's interests, and ideological commitments; Karl Mannheim's sociology of knowledge showed how thought is shaped by social position. Objectivity is therefore best understood as intersubjectivity — findings that other competent researchers, using the same methods, can check and reproduce. It is pursued through standardised procedures, operational definitions, representative sampling, controls, triangulation of methods, peer review, transparency about data and methods, and reflexivity — the researcher's conscious scrutiny of his or her own position and influence.

⚠️ Value-relevance is not value-freedom

Weber did not claim that research can be free of values altogether. He held that values legitimately decide which problems we study (value-relevance) but must not decide what we find (value-freedom in analysis). An option saying Weber thought social science should have no connection with values misreads him.

3. Research and theory-building; the scientific method

A theory is a set of interrelated concepts and propositions that explains and predicts a class of phenomena. Research and theory feed each other in a cycle — Walter Wallace's "wheel of science" (1971): theory generates hypotheses by deduction; hypotheses are tested against observations; observations lead to empirical generalisations by induction; and generalisations refine or build theory. Research initiates theory (grounded theory — Barney Glaser and Anselm Strauss, The Discovery of Grounded Theory, 1967 — builds theory inductively from data), reformulates it (when findings contradict it), refocuses it, and clarifies it. Robert Merton argued for "theories of the middle range", between grand theory and raw empiricism — much of public administration's usable theory (on motivation, implementation, street-level bureaucracy) is of this kind.

The scientific method proceeds by observation, formulation of a problem, hypothesis, testing through systematic data collection, analysis, and conclusion, with results open to replication. Its hallmarks are empiricism, verifiability, systematic procedure, generality, predictability and self-correction. Karl Popper, in The Logic of Scientific Discovery (1934 in German; English 1959), argued that what makes a theory scientific is falsifiability — it must forbid some observable outcomes and so risk being refuted; theories are never finally proved, only corroborated by surviving attempts at refutation. Thomas Kuhn's The Structure of Scientific Revolutions (1962) argued that science advances not only by accumulation but through paradigms — shared frameworks of "normal science" that are overturned in revolutionary paradigm shifts when anomalies accumulate. Henry's "paradigms" of public administration borrow Kuhn's word.

4. Types of research and the research problem

Types of research
Basis	Types
Purpose	Pure (basic, fundamental) — to extend knowledge and build theory; applied — to solve a practical problem; action research (Kurt Lewin, 1946) — researchers and practitioners jointly study and change a situation in cycles of planning, acting, observing and reflecting
Objective	Exploratory — to gain familiarity with a little-understood problem and formulate hypotheses; descriptive — to describe characteristics accurately; explanatory (causal) — to test cause-and-effect relationships; evaluative — to judge a programme's effectiveness
Nature of data	Quantitative — numerical measurement and statistical analysis; qualitative — words, meanings, observation and interpretation; mixed methods — combining both
Time	Cross-sectional — at one point in time; longitudinal — over time (panel, trend and cohort studies)
Setting and approach	Historical, comparative, case study, experimental (laboratory or field), survey, ethnographic

Identifying a research problem is the first and often hardest step. Problems come from theory and the literature (gaps and contradictions), from practice (an administrative puzzle — why do some districts implement a scheme better than others?), from policy debates, and from replication of earlier studies. A good problem is researchable (answerable with data), significant (theoretically or practically), feasible in time, money and access, novel or at least not trivially repetitive, ethical, and clearly delimited. It should be stated as a specific question, with its key concepts defined, after a review of literature that shows what is already known.

5. Hypotheses, null hypotheses and the validation of hypotheses

A hypothesis is a tentative, testable statement of the relationship between two or more variables — a proposed answer to the research question. A good hypothesis is conceptually clear, specific, empirically testable, related to available techniques, related to a body of theory, and parsimonious. Hypotheses may be descriptive or relational, directional ("officers with longer tenure in a district achieve higher scheme coverage") or non-directional ("tenure is related to scheme coverage"). The null hypothesis (H₀) states that there is no relationship or difference ("tenure has no effect on coverage"); the alternative or research hypothesis (H₁) states that there is one. Statistical testing proceeds by trying to reject the null hypothesis: one never "proves" H₁, one only finds the evidence strong enough to reject H₀ at a chosen level of significance.

Errors in testing a hypothesis
Decision	H₀ actually true	H₀ actually false
Reject H₀	Type I error (false positive), probability α — the level of significance, conventionally 0.05	Correct decision; probability 1 − β is the power of the test
Fail to reject H₀	Correct decision (probability 1 − α)	Type II error (false negative), probability β

Validation of a hypothesis combines statistical and logical tests. Statistically, a test (t-test, chi-square, correlation, ANOVA) yields a p-value — the probability of obtaining results at least as extreme as those observed if H₀ were true; if p is below α (say 0.05), H₀ is rejected and the result called "statistically significant". Logically, validation also requires that the relationship is not spurious (produced by a third variable), that the cause precedes the effect, that the measures are valid and reliable, that the sample permits generalisation (external validity) and that the design rules out alternative explanations (internal validity) — and, following Popper, that the finding survives replication. A significant result is not the same as an important one: with large samples trivially small effects become significant, which is why effect size matters.

Key takeaways

A method is a technique of data collection or analysis; methodology is the logic and philosophy (ontology, epistemology) that justifies the choice of methods.
Weber separated value-relevance (choice of problem) from value-freedom (conduct of analysis); Myrdal urged stating value premises; objectivity is best understood as intersubjectivity and reflexivity.
Research and theory cycle through deduction and induction (Wallace's wheel); grounded theory builds theory from data; Popper made falsifiability the test of science and Kuhn explained paradigm shifts.
Research may be pure, applied or action research; exploratory, descriptive, explanatory or evaluative; quantitative, qualitative or mixed; cross-sectional or longitudinal.
The null hypothesis states no relationship; rejecting a true H₀ is a Type I error (α), failing to reject a false one a Type II error (β); significance is not the same as importance.

Practice questions (10)

Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.

Rejecting a null hypothesis that is actually true is called:
1. A Type I error
2. The power of the test
3. A Type II error
4. A sampling frame error
Show answer
Answer: A — A Type I error
A Type I error is a false positive, with probability α (the significance level). A Type II error is failing to reject a false null; power (1 − β) is the probability of correctly rejecting a false null.
The distinction between methodology and method is best expressed as:
1. The two terms mean exactly the same thing
2. A method is the philosophy of research; methodology is the questionnaire
3. Methodology is the logic that justifies the choice of techniques; a method is a specific technique of collecting or analysing data
4. Methodology refers to qualitative research and method to quantitative research
Show answer
Answer: C — Methodology is the logic that justifies the choice of techniques; a method is a specific technique of collecting or analysing data
Methodology asks why a method suits a question, resting on assumptions about knowledge; a method is how data are gathered or analysed. Both qualitative and quantitative research have methodologies and methods.
Which of the following statements about Weber's position on values in social research are correct? Select all that apply.
1. Social science can and should be entirely unconnected with values at every stage
2. Values should determine what the findings are
3. The researcher should not present personal value judgements as scientific findings
4. Values legitimately influence the choice of research problem
Show answer
Answer: C — The researcher should not present personal value judgements as scientific findings; D — Values legitimately influence the choice of research problem
Weber combined value-relevance in choosing problems with value-freedom in analysis and reporting. He did not think research could be unconnected with values altogether, and he rejected letting values decide findings.
Karl Popper held that what makes a theory scientific is its:
1. Verification by many confirming instances
2. Acceptance by the scientific community
3. Falsifiability
4. Mathematical form
Show answer
Answer: C — Falsifiability
For Popper a theory is scientific if it could in principle be refuted by observation; confirming instances can never prove a universal theory. Acceptance within a community is closer to Kuhn's account of paradigms.
Assertion (A): In hypothesis testing, researchers speak of rejecting or failing to reject the null hypothesis rather than of proving the research hypothesis. Reason (R): Statistical evidence can show only that data would be unlikely if the null were true; it cannot establish a hypothesis with certainty. Choose the correct option.
1. Both A and R are true, and R is the correct explanation of A
2. Both A and R are true, but R is not the correct explanation of A
3. A is true, but R is false
4. A is false, but R is true
Show answer
Answer: A — Both A and R are true, and R is the correct explanation of A
Both are true and R explains A: because a test yields only a probability of the data under H₀, the logically sound conclusion is to reject or retain H₀, never to prove H₁ — which is also Popper's point about corroboration.
Action research, in which researchers and practitioners jointly study and change a situation through cycles of planning, acting and reflecting, is associated with:
1. Thomas Kuhn
2. Robert Merton
3. Karl Popper
4. Kurt Lewin
Show answer
Answer: D — Kurt Lewin
Lewin coined "action research" in 1946. Popper is associated with falsifiability, Kuhn with paradigms, and Merton with middle-range theory.
A research team independently tests 200 null hypotheses, all of which are in fact true, each at a significance level of 0.05. How many of them should it expect to reject wrongly? Type a number.
Numerical answer — type the value.
Show answer
Answer: 10
α is the probability of a Type I error on each test of a true null, so the expected number of false rejections is 200 × 0.05 = 10. This is why running many tests and reporting only the "significant" ones misleads.
Which of the following correctly match a type of research with its description? Select all that apply.
1. Longitudinal — studying the same phenomenon over a period of time
2. Exploratory — gaining familiarity with a little-understood problem and formulating hypotheses
3. Explanatory — testing cause-and-effect relationships
4. Cross-sectional — following one cohort for twenty years
Show answer
Answer: A — Longitudinal — studying the same phenomenon over a period of time; B — Exploratory — gaining familiarity with a little-understood problem and formulating hypotheses; C — Explanatory — testing cause-and-effect relationships
The first three are correct. Following a cohort for twenty years is longitudinal; cross-sectional research observes at a single point in time.
"Grounded theory", in which theory is built inductively from data rather than tested deductively, was developed by:
1. Max Weber and Gunnar Myrdal
2. Barney Glaser and Anselm Strauss
3. Donald Campbell and Julian Stanley
4. Karl Popper and Thomas Kuhn
Show answer
Answer: B — Barney Glaser and Anselm Strauss
Glaser and Strauss's The Discovery of Grounded Theory (1967) set out the approach. Campbell and Stanley are known for their work on experimental and quasi-experimental designs.
Which of the following is a directional hypothesis?
1. Districts with longer-serving Collectors achieve higher coverage of the scheme
2. The tenure of Collectors has no effect on scheme coverage
3. The tenure of Collectors is related to scheme coverage
4. What explains variation in scheme coverage across districts?
Show answer
Answer: A — Districts with longer-serving Collectors achieve higher coverage of the scheme
A directional hypothesis specifies which way the relationship runs ("higher"). The second is non-directional, the third is the null hypothesis, and the fourth is a research question, not a hypothesis.