Research design and its types
Research design is the blueprint of a study: what data to collect, from whom, how and when. The main types are exploratory, descriptive and experimental.
Main types of research design (Kothari)
| Type | Purpose | Features | Nutrition example |
|---|---|---|---|
| Exploratory (formulative) | Gain new ideas; sharpen the problem | Flexible; literature survey, experience survey, case studies | Group discussion on why girls dislike iron tablets |
| Descriptive | Describe a group or situation accurately | Fixed plan; surveys; guards against bias | Survey of the food habits of college students |
| Diagnostic | Find how often something occurs and what it is linked with | Same design needs as descriptive; looks for associations | Is low iron intake linked with anaemia in girls? |
| Hypothesis-testing (experimental) | Test cause and effect | Researcher gives a treatment; uses a control group and randomisation | Fortified rice in randomly chosen schools, usual rice in the others; compare haemoglobin after 6 months |
Experimental designs (Kothari)
Designs by contact and time (Ranjit Kumar)
| Basis | Design | Nutrition example |
|---|---|---|
| Number of contacts | Cross-sectional: one contact | One-time survey of anaemia in 300 girls |
| Number of contacts | Before-and-after: two contacts | Weight before and after 12 weeks of diet counselling |
| Number of contacts | Longitudinal: many contacts over time | Growth of the same infants measured every month for two years |
| Reference period | Retrospective: looks back | Past infant-feeding practices of mothers of malnourished children |
| Reference period | Prospective: looks forward | Follow pregnant women to record their babies' birth weight |
| Nature of investigation | Experimental, non-experimental, quasi-experimental | Randomised supplement trial; a survey; a study partly experimental and partly non-experimental (other books: a trial without random allocation) |
Where authors differ
Kothari sorts designs by purpose: exploratory, descriptive/diagnostic and hypothesis-testing. Ranjit Kumar sorts them by number of contacts, reference period and nature of investigation. Nutrition and public-health books use epidemiological names: cross-sectional, case-control, cohort and randomised controlled trial (RCT). These are different ways of sorting the same designs. In the exam, say which author you follow.
Most student projects are descriptive cross-sectional surveys, such as diet and anaemia in college girls. Testing a new recipe or supplement is experimental. The strongest proof that a diet therapy works comes from a randomised controlled trial.
2 marks: “What is a research design?” A research design is the blueprint of a study: the plan for collecting, measuring and analysing data so that the research question is answered with the least bias.
5 marks: “Write a note on experimental research design.” Define it (treatment, control group, randomisation), list Kothari's informal designs and name the formal ones, with a supplement-trial example.
10 marks: “What is a research design? Explain its types with suitable examples.” Definition, features of a good design, Kothari's three types with examples, experimental designs, and Ranjit Kumar's classification by contact and time.
Kothari's order: Explore to find the question, Describe to know the facts, Experiment to prove the cause.