Slide 1 : RESEARCH DESIGN
PROF.P.VIJAYA PRAKASH
DEPARTMENT OF ANTHROPOLOGY
ANDHRA UNIVERSIT
VISAKHAPATNAM-530003
Slide 2 : RESEARCH DESIGN A research design is more than just the data collection methods used
It refers to the logic of how these data will be collected
It involves both qualitative and quantitative data
What kind of data will help to the research question
How the data are going to answer the question Selection of an APPROPRIATE RESEARCH DESIGN is essential which is capable of answering research question
Slide 3 : Common kinds of Research Designs: 1. Experiments 2. Surveys 3. Observations 4. Documentary 5. Participatory
Slide 4 : 2. Surveys Surveys aim to describe or explain the characteristics or opinions of a population through the use of a representative sample.
(small local samples t large scale national surveys)
Surveys can be taken up under four headings: Factual: Total enumeration of population (material situation rather than attitude or opinion)
Attitudinal: Gaining data on attitudes (public opinion on issues) – A policy is justified by ‘what the public demand (-Srikrishna Commission)
Social-psychological: Concern with small group behaviour (focus on the relationship between attitude and behaviour)
Explanatory: To test hypotheses derived from theories (to explain social processes)
Slide 5 : Social Surveys: Research seeks “statistical evidence” for a theory rather than “proof”
Surveys aim to show causal relationships (positivistic tradition) Characteristics:
Standardization of the survey depends on how questionnaire is designed, administered and analyzed
Replicability of the survey increases confidence leads to ‘reliability’ and ‘validity’
Representativeness of the population – statistically significant
Slide 6 : Sampling: A sample is a portion or a subset of a large group called population
SAMPLING IS CENTRAL TO SURVEY DESIGN Probability samples: Total list is ‘Sample Frame’ from which a sample is randomly selected (Mathematical probability – Statistical error)
Multistage cluster sampling (sampling at different levels and stages)
Stratified random sample (stratification according to characteristics such as age, group, gender, type of housing etc.)
Systematic random sampling (selecting a random number (5th 10th etc.)
Slide 7 : Non-Probability Samples: If there is no ‘Sampling Frame’ (total list)
Particularly market research – the statistical accuracy of probability sample Purposive sampling – according to known characteristics (politicians / trade union leaders etc.) “FIT FOR PURPOSE”
Snowball sampling – homeless people, drug users etc. (initial contact may be made with a member of the population who will lead the research to other members of the same population – Generalization is not really legitimate, mainly because we have no idea of the size of the population – yet the method has often proven valuable) Sampling decisions are therefore:
Bound up closely with the survey’s target population
The unit of analysis may not be an individual, a household (a political party, a school, and so on)
Slide 8 : Types of Questionnaires: Self-Completion questionnaire (mail)
Telephone survey, and
Face-to-face interview schedule Designing and Testing questionnaire: Clear in your own mind what the question is for?
To whom it is to be answered by
How you are intend them to interpret Pilot survey: Initial fieldwork to verify the questions to be posed, to take account of criticism, clarity and problems
Slide 9 : Types of questions: Classification questions – personal, demographic face sheet information (age, income, housing …etc.) explanatory variables
Factual questions – facts or opinions (materialistic)
Opinion questions – (a person in front of a group of friends may well elicit a different answer)
Open and closed question:
* open questions give respondents a greater freedom to answer the question in a way that suits their interpretation
* Closed questions limit the number of possible answers Coding: Questionnaires are pre-coded to allow the classification of the responses into analysable and meaningful categories.
Slide 10 : 4. Documentary
Slide 11 : 5. Participatory Research Design:
Slide 12 :
Slide 13 : Sensitize yourself (every individual is born equal)
You ‘believe and trust’ that they (individual or group) possess knowledge
Capable of excellent ‘communication skill’, not oral / oratory, but in a manner others understand
Develop the art of learning from simple things (Newton-Apple)
Possess tremendous concentration to develop rapport to tune the informant to recollect the exact information to share with you Components Hardware (Techniques) Software (Tools)
Slide 14 : Components Hardware (Techniques) Software (Tools) Maps (Social map; Resource map; Demographic map; Family history
Diagrams (Venn diagrams; Transect walks; Time line; Flow diagram; Seasonal analysis etc.)
Matrix (Pair-wise; Issue/Situation/Ranking/Variable Matrix)
Others: (Key informants; Focus group discussions etc.)
Slide 15 :
Slide 16 :
Slide 17 :
Slide 18 :
Slide 19 :
Slide 20 :
Slide 21 :
Slide 22 :