Notes on research design for UGC NET Paper 1 Research Aptitude, in 4 short sub-topics. Each one gives the concept, the facts to memorise and the traps examiners set.
4 sub-topics
8 memory points
4 exam-trap notes
3 min read
Topic 12 of 29
What this topic covers
4 sub-topics, in the order to study them. Tap one to jump to it.
The research design is the plan that connects the research questions to the data and the analysis; it specifies the type of study, the sample, the tools, the timing and the analytical procedures. Its purpose is to obtain valid answers economically and to control variance. Kothari groups designs as exploratory (flexible, to formulate a problem), descriptive and diagnostic (accurate description, little flexibility) and experimental (hypothesis testing). Fisher’s three principles of experimental design are replication, randomisation and local control, and Kerlinger’s MAX-MIN-CON principle guides variance control. A research proposal or synopsis sets out the design in advance: title, background, review, objectives, hypotheses, methodology, sampling, tools, analysis plan, schedule and references.
Memorise
Design = blueprint. Fisher: replication, randomisation, local control.
Exploratory designs are used to formulate a problem or hypotheses when knowledge is limited. Selltiz and colleagues named three common methods: a literature survey, an experience survey of knowledgeable people, and the analysis of insight-stimulating examples (case studies). Mixed-methods designs, as classified by Creswell, include the convergent parallel design (both strands collected together and merged), the explanatory sequential design (quantitative first, qualitative to explain), and the exploratory sequential design (qualitative first, then a quantitative test). Notation such as QUAN → qual shows the order and the priority of the strands.
Memorise
Exploratory methods: literature survey, experience survey, insight-stimulating cases.
Explanatory sequential = QUAN then qual; exploratory sequential = qual then QUAN.
Analyse / traps
In “explanatory” sequential, the qualitative strand explains numbers, which is the reverse of the exploratory sequence.
In quasi-experimental designs the researcher manipulates a treatment but cannot randomly assign participants, as in the non-equivalent control group design, the interrupted time-series design and the regression-discontinuity design. True experimental designs use random assignment. Fisher’s formal designs include the completely randomised design (units assigned wholly at random), the randomised block design (units grouped into homogeneous blocks and randomised within blocks), the Latin square (two blockings) and factorial designs (several factors studied together with their interactions). Notation: R for random assignment, X for the treatment and O for an observation.
Memorise
Quasi-experiment: treatment but no randomisation.
CRD, RBD, Latin square, factorial: formal experimental designs.
Analyse / traps
If groups were formed by existing classrooms rather than random assignment, the design is quasi-experimental.
Construct validity asks whether the study’s measures and treatments really represent the theoretical constructs intended; it is threatened by vague definitions of the construct, relying on a single measure (mono-operation bias) or a single method (mono-method bias). Variables can be further classified: latent variables are not directly observable (intelligence, attitude) and are inferred from manifest or observed indicators such as test scores; dichotomous variables have two categories (pass/fail); dummy variables code categories as 0 and 1 for regression; composite variables combine several items into one score.
Last updated: . Notes are a revision aid; check official UGC sources for any rule or scheme that changes over time. Spotted a mistake? Write to netlawverse@gmail.com.