Sociology – II
Course Objective:
• Basic & Applied Sociological
Research
• Public Opinion and Social
Research Data Collection &
analysis
• Key Contemporary Sociological
Challenges
Evaluation Pattern:
• 1 MCQ test (10 questions) (10
marks)
• 1 Group Presentation (10
marks)
• Mid Sem (30 marks) & End Sem
(30 marks)
What is a Survey?
• A unique technique of collection of social data
• Uses questionnaire/schedule as a tool
• Several other techniques also use schedules (structured interviews,
observation, content analysis, etc.)
• What distinguishes Survey from other techniques:
• Form of data
• Method of analysis
Form of Data
• Characterised by a structured or systematic set of data
• data collected on the same variable from more cases
• Each row represents a case & each column a variable
• Cases are comparable since we collect the same information for each case
Case Age Sex Education Caste/class Voting
preference
Person 1
Person 2
Person 3
Person 4
Person 5
Person 6
Techniques of surveys
• The techniques of data collection may vary between surveys
• Data may be collected through a questionnaire and then the
responses can be copied to the data grid
• We may collect data through observation and fill the grid
• We can interview someone and gather data, and then extract to grid
• But questionnaires are highly structured and hence provide a
straightforward way of collecting data
The case in the data grid
• The case may not always be a person.
• The case is the unit of analysis, and it can be a class, a group, a
village, a city, a country, etc.
• But we need to collect information from many cases and on same
variable.
Cases Population Size Area Density Unemployment
Country 1
Country 2
Country 3
Case data grid
• A Survey generates a case data grid
• Determining variables of grid depends upon the research
objective
• cases can be compared based on certain variable
Method of Analysis – the 2nd distinguishing
feature
• One of the functions of a survey is to describe the characteristic of a set of cases
– a description of the cases
• The variable by the case grid provides this information
• But a survey researcher may also want to know the causes of phenomena
• Causes can be located by comparing cases
• By looking at how cases vary on some characteristics
• Like in the first table – whether there is a variation in voting preferences
• Or in the 2nd table – where there is variation in unemployment
• We need to look how these variations are related to other variables of cases
(whether they relate to caste/class, size, population, etc)
Analysis of survey data
• Survey data help us to understand what may cause some
phenomena by looking at:
• variations in that variable across cases and other characteristics that are
systematically linked with it
• Survey data aim to draw causal inferences by careful comparison of
various characteristics of cases
Survey vis-à-vis other research methods
• Survey and case study:
• Case study focuses on a particular case and tries to develop a full and
comprehensive understanding of the case
• Case study does not rely on comparing cases, but fully on
understanding the wholeness of a particular case, including its history
• Survey and experimental method
• The experimental method also produces variable by case grid, but the
focus is to see whether a particular intervention creates a difference
Survey and experimental method
• Example of an experimental method: - A research to know whether a drug cures a disease
• A researcher takes a group of sufferers and then dives them to two similar groups and the drug will be
administered to one
• The recovery rates of drug and non-drug groups would be compared
• In the experimental method, the variation is created by the researcher, but in survey method, naturally
occurring variations are observed
• We can compare between two groups in survey method, but groups may not be similar in other
respects
• But in experimental method, the researcher controls the situation, and the differences that arise are
attributed to the intervention
• Ethical issues become a challenge in experimental method, but not in survey
Survey vis-à-vis other research methods
Case study
Questionnaire
Research
Question
Structured Interview
In-depth interview
Survey
Observation
Content analysis
Experiment
al
Qualitative vs Quantitative research
• Survey research is inherently quantitative & positivist
• In contrast, qualitative research involves – participant observation, unstructured
interviewing, case study, focused group
• Survey is often portrayed as unimaginative, but is well suited to provide factual and
descriptive data – hard evidence
• Qualitative research provides rich data – in-depth, real-life situation
• Lacks generalizability, too much reliant on subjective interpretation, can be replicated
by subsequent researcher, case specific
• We can distinguish between the two on – collecting data and analysing data
Distinction on the basis of data collection
QUANTITATIVE QUALITATIVE
• Structured data • Un-structured data
• Survey may gather qualitative data, • Case study may collect quantitative
but the data grid is filled in a data, but only from one or few
systematic manner cases
• Example – opinions on a scale of 1 • Can not compare cases
to 5
Thus, what distinguishes a survey, is not the nature of data, but the method of data
collection
The logic of analysis in survey is also different
from case study
• The logic of analysis is that variations in one variable is matched with variations in other
variables – co-variations
• Causal analysis is integral part of survey
• Survey data can be used to see if one variable affects other variables
• Ex – if caste, religion, education, etc affect voting or unemployment
• Survey method frequently uses statistical tools
• Thus, “survey method is a structured approach to data collection and analysis and it
relies on a particular logic of analysis”
• Typifying survey research on the basis of quantitative/qualitative distinction is misleading
