Types of Criminological Research (UGC-NET Unit VI)

Exploratory research applies when a topic is too new or too poorly mapped for a structured instrument, such as an emerging form of online fraud. It relies on unstructured interviews, small purposive samples and case studies, and its output is a set of working hypotheses and vocabulary, not a tested claim; because the sample is not randomly drawn, findings do not generalise to a wider population.

Descriptive research takes a phenomenon that is already reasonably well defined and profiles its extent using structured instruments applied to a large or representative sample, such as victimisation surveys or official crime counts. It answers who, what, where and how much, and a well-designed descriptive study reports a margin of error and distinguishes a real change in a rate from one caused by a shift in recording practice.

The two are often chained: an exploratory phase of interviews generates a working typology of a problem, and a descriptive survey then measures how common each type is across a wider population. A NET question that describes small unstructured interviews generating hypotheses is exploratory; one that describes a large structured survey counting a rate is descriptive.

Explanatory research, also called analytical research on the syllabus, moves past description to test why a pattern exists. Where a descriptive study reports that a burglary rate is higher in certain neighbourhoods, an explanatory study tests a specific candidate cause, such as whether unemployment or residential turnover predicts the difference, typically through the quantitative and qualitative methods a researcher chooses, most often regression models or longitudinal designs.

The core discipline explanatory research imposes is separating correlation from causation: two variables can move together because a third factor drives both, or because the direction of causation runs the opposite way to the one assumed. Explanatory theories such as strain theory and differential association each propose a specific mechanism and generate a testable prediction that later studies can support, qualify or contest.

Because explanatory research usually works with data the researcher did not control, it can never fully rule out an unmeasured confounding variable, even after controlling for several alternative explanations. That limitation is exactly what experimental research, covered next, is built to overcome.

An experimental design isolates a causal effect by manipulating a variable directly and comparing outcomes across a treatment group and a control group. The defining feature is random assignment: when a random process decides who receives the treatment, the two groups should be statistically equivalent on every other characteristic before the intervention starts, so any later difference in outcome can be attributed to the treatment.

True random assignment is often impossible or unethical in criminal justice settings, such as a court randomly assigning sentence severity purely for research purposes. Quasi-experimental designs respond to that constraint by comparing groups that differ on the variable of interest for reasons outside the researcher's control, such as a policy change that took effect in one jurisdiction but not a comparable neighbouring one, and using statistical matching to approximate a randomised comparison.

A single trial establishes an effect in one setting and one period; a policy conclusion needs replication across settings before it generalises. For NET purposes, the presence or absence of random assignment is the fastest way to tell an experimental design apart from an explanatory one that merely compares existing groups.

Doctrinal research is not empirical in the sense the previous four types are; it does not collect data about behaviour. It analyses primary legal sources, statutes and the judgments that interpret them, to state what the law currently requires, using close textual reading rather than fieldwork. Criminology needs it because a descriptive count of an offence is uninterpretable without a doctrinal account of what conduct the governing statute actually covers, and that account changes when the statute changes.

A second, cross-cutting classification sorts studies by purpose rather than design: basic research tests a theoretical claim with no immediate practical client, while applied research is commissioned to answer a specific agency's question. Evaluation research is the most common applied form in criminal justice, measuring whether an existing programme achieved its intended outcome, and it borrows its comparison logic directly from the experimental and quasi-experimental designs above.

The basic-applied distinction is about motive, not rigour. NET questions sometimes pair a design label (descriptive, analytical, experimental, exploratory, doctrinal) with a purpose label (basic or applied) in the same item, and the two axes must be scored independently: a study can be exploratory in design and basic in purpose, or descriptive in design and applied in purpose.