Notes: Variables, Sampling, Hypothesis, Reliability & Validity | Sociology Mains Optional for UPSC 2024 PDF Download

 Page 1


 
 
3. Research Methods and 
Analysis 
 
C.Variables, Sampling, 
Hypothesis, Reliability and 
Validity.  
  
Page 2


 
 
3. Research Methods and 
Analysis 
 
C.Variables, Sampling, 
Hypothesis, Reliability and 
Validity.  
  
 
 
CHOOSING A SAMPLE: 
Sample is a part of larger population, representative of that population and chosen 
as a cross section of larger group 
 
Stages of sampling: 
1) Identifying target population 
2) Obtaining/producing a sampling frame (source of data) eg census, electoral 
register 
3) Sample should have same proportion of people having relevant characteristics 
4) Generalize the result 
 
Types of sampling 
 
A. PROBABILITY SAMPLING 
1. Random sampling: Each unit has equal chance of being chosen (less costly 
and time consuming) 
2. Systematic Sampling: A random starting point but next members are chosen 
on a fixed interval. Say, every 10
th
 or 20
th
 item to be selected. 
 
Both techniques rely on law of averages that eventually representative sample will be 
formed. 
 
3. Stratified Random sampling: division of sample frames into groups to ensure 
sample is representative. Division is based on variable which researcher wants 
to control. 
 
B. NON- PROBABILITY SAMPLING 
1. Quota Sampling: quota is filled for particular category and till then no responses 
will be collected. It is quicker and cheaper with no need to produce to 
sampling frame. Everybody doesn’t have chance to get selected. Difficult to 
fill quotas of minority. 
2. Multistage sampling: samples from sample when population is large. Ex -opinion 
polls 
3. Snowballing sampling: involves using of personal contacts of one sample entity 
to bring in others. Ex criminals bringing their friends. Specific and used when 
other techniques cant be used. 
4. Volunteer sampling  
 
Non-representative sampling: 
5. Convenience sampling – easily accessible and available.  
6. Purposive sampling- Deliberately selected sample on the basis of certain variables. 
Eg: Sample for study of domestic violence from areas where incidents are higher in 
frequency. 
 
  
Page 3


 
 
3. Research Methods and 
Analysis 
 
C.Variables, Sampling, 
Hypothesis, Reliability and 
Validity.  
  
 
 
CHOOSING A SAMPLE: 
Sample is a part of larger population, representative of that population and chosen 
as a cross section of larger group 
 
Stages of sampling: 
1) Identifying target population 
2) Obtaining/producing a sampling frame (source of data) eg census, electoral 
register 
3) Sample should have same proportion of people having relevant characteristics 
4) Generalize the result 
 
Types of sampling 
 
A. PROBABILITY SAMPLING 
1. Random sampling: Each unit has equal chance of being chosen (less costly 
and time consuming) 
2. Systematic Sampling: A random starting point but next members are chosen 
on a fixed interval. Say, every 10
th
 or 20
th
 item to be selected. 
 
Both techniques rely on law of averages that eventually representative sample will be 
formed. 
 
3. Stratified Random sampling: division of sample frames into groups to ensure 
sample is representative. Division is based on variable which researcher wants 
to control. 
 
B. NON- PROBABILITY SAMPLING 
1. Quota Sampling: quota is filled for particular category and till then no responses 
will be collected. It is quicker and cheaper with no need to produce to 
sampling frame. Everybody doesn’t have chance to get selected. Difficult to 
fill quotas of minority. 
2. Multistage sampling: samples from sample when population is large. Ex -opinion 
polls 
3. Snowballing sampling: involves using of personal contacts of one sample entity 
to bring in others. Ex criminals bringing their friends. Specific and used when 
other techniques cant be used. 
4. Volunteer sampling  
 
Non-representative sampling: 
5. Convenience sampling – easily accessible and available.  
6. Purposive sampling- Deliberately selected sample on the basis of certain variables. 
Eg: Sample for study of domestic violence from areas where incidents are higher in 
frequency. 
 
  
 
 
ADVANTAGES of Sampling: 
1. Saves time and money 
2. Ease of supervision as smaller size 
3. Increases the accuracy of study (as proper focus) 
4. A more detailed study can be conducted with the small amount of 
resources 
 
DISADVANTAGES: 
1. Selection of proper size of sample is a difficult job, sample may not be truly 
representative. 
2. Improper selection of sampling technique may affect the whole process 
3. Faulty of biased selection will lead to inaccurate results 
4. Validity is not as accurate as in census method 
 
 
Karl Popper:  
Researcher should try to falsify their theories by looking at untypical examples. 
 
Herbert Blumer: 
Study best informed members of social groups rather than cross section of a group 
 
Applications- NFHS, ASER 
  
Page 4


 
 
3. Research Methods and 
Analysis 
 
C.Variables, Sampling, 
Hypothesis, Reliability and 
Validity.  
  
 
 
CHOOSING A SAMPLE: 
Sample is a part of larger population, representative of that population and chosen 
as a cross section of larger group 
 
Stages of sampling: 
1) Identifying target population 
2) Obtaining/producing a sampling frame (source of data) eg census, electoral 
register 
3) Sample should have same proportion of people having relevant characteristics 
4) Generalize the result 
 
Types of sampling 
 
A. PROBABILITY SAMPLING 
1. Random sampling: Each unit has equal chance of being chosen (less costly 
and time consuming) 
2. Systematic Sampling: A random starting point but next members are chosen 
on a fixed interval. Say, every 10
th
 or 20
th
 item to be selected. 
 
Both techniques rely on law of averages that eventually representative sample will be 
formed. 
 
3. Stratified Random sampling: division of sample frames into groups to ensure 
sample is representative. Division is based on variable which researcher wants 
to control. 
 
B. NON- PROBABILITY SAMPLING 
1. Quota Sampling: quota is filled for particular category and till then no responses 
will be collected. It is quicker and cheaper with no need to produce to 
sampling frame. Everybody doesn’t have chance to get selected. Difficult to 
fill quotas of minority. 
2. Multistage sampling: samples from sample when population is large. Ex -opinion 
polls 
3. Snowballing sampling: involves using of personal contacts of one sample entity 
to bring in others. Ex criminals bringing their friends. Specific and used when 
other techniques cant be used. 
4. Volunteer sampling  
 
Non-representative sampling: 
5. Convenience sampling – easily accessible and available.  
6. Purposive sampling- Deliberately selected sample on the basis of certain variables. 
Eg: Sample for study of domestic violence from areas where incidents are higher in 
frequency. 
 
  
 
 
ADVANTAGES of Sampling: 
1. Saves time and money 
2. Ease of supervision as smaller size 
3. Increases the accuracy of study (as proper focus) 
4. A more detailed study can be conducted with the small amount of 
resources 
 
DISADVANTAGES: 
1. Selection of proper size of sample is a difficult job, sample may not be truly 
representative. 
2. Improper selection of sampling technique may affect the whole process 
3. Faulty of biased selection will lead to inaccurate results 
4. Validity is not as accurate as in census method 
 
 
Karl Popper:  
Researcher should try to falsify their theories by looking at untypical examples. 
 
Herbert Blumer: 
Study best informed members of social groups rather than cross section of a group 
 
Applications- NFHS, ASER 
  
 
 
Reliability 
- if other researchers using the same method on same material produce the 
same results 
- if reliability can be established results can be generalized. 
- Generally speaking, quantitative methods are more reliable than qualitative 
method. 
 
Validity 
A valid statement gives true measurement/description and explanation of what it 
claims to measure.  
It is accurate reflection of social reality.  
Data can be reliable without being valid. 
 
Alan Bryman outlines four types of validity 
1) Measurement validity/ Construct validity: 
Whether a measure that is employed really measures what it claims. 
Eg If IQ tests really measure intelligence 
 
2) Internal validity:  
If one thing is said to cause another, this explanation will be internally valid, if 
causal relationship is true.  
Ex Pierre Bourdieu- Different cultural settings produce different educational 
achievements.  
 
3) External validity:  
If study can be generalized to situations other than study itself. 
 
4) Ecological validity: 
How accurately a research mirrors natural setting or real experience.  
Lab experiments may lack ecological validity. 
 
Validation by respondents (of what they actually meant) overcome problems of 
validity 
Practicality: less time consuming and less personal commitments 
Ethics: researcher will choose topics where informed consent and confidentiality are 
possible 
  
Page 5


 
 
3. Research Methods and 
Analysis 
 
C.Variables, Sampling, 
Hypothesis, Reliability and 
Validity.  
  
 
 
CHOOSING A SAMPLE: 
Sample is a part of larger population, representative of that population and chosen 
as a cross section of larger group 
 
Stages of sampling: 
1) Identifying target population 
2) Obtaining/producing a sampling frame (source of data) eg census, electoral 
register 
3) Sample should have same proportion of people having relevant characteristics 
4) Generalize the result 
 
Types of sampling 
 
A. PROBABILITY SAMPLING 
1. Random sampling: Each unit has equal chance of being chosen (less costly 
and time consuming) 
2. Systematic Sampling: A random starting point but next members are chosen 
on a fixed interval. Say, every 10
th
 or 20
th
 item to be selected. 
 
Both techniques rely on law of averages that eventually representative sample will be 
formed. 
 
3. Stratified Random sampling: division of sample frames into groups to ensure 
sample is representative. Division is based on variable which researcher wants 
to control. 
 
B. NON- PROBABILITY SAMPLING 
1. Quota Sampling: quota is filled for particular category and till then no responses 
will be collected. It is quicker and cheaper with no need to produce to 
sampling frame. Everybody doesn’t have chance to get selected. Difficult to 
fill quotas of minority. 
2. Multistage sampling: samples from sample when population is large. Ex -opinion 
polls 
3. Snowballing sampling: involves using of personal contacts of one sample entity 
to bring in others. Ex criminals bringing their friends. Specific and used when 
other techniques cant be used. 
4. Volunteer sampling  
 
Non-representative sampling: 
5. Convenience sampling – easily accessible and available.  
6. Purposive sampling- Deliberately selected sample on the basis of certain variables. 
Eg: Sample for study of domestic violence from areas where incidents are higher in 
frequency. 
 
  
 
 
ADVANTAGES of Sampling: 
1. Saves time and money 
2. Ease of supervision as smaller size 
3. Increases the accuracy of study (as proper focus) 
4. A more detailed study can be conducted with the small amount of 
resources 
 
DISADVANTAGES: 
1. Selection of proper size of sample is a difficult job, sample may not be truly 
representative. 
2. Improper selection of sampling technique may affect the whole process 
3. Faulty of biased selection will lead to inaccurate results 
4. Validity is not as accurate as in census method 
 
 
Karl Popper:  
Researcher should try to falsify their theories by looking at untypical examples. 
 
Herbert Blumer: 
Study best informed members of social groups rather than cross section of a group 
 
Applications- NFHS, ASER 
  
 
 
Reliability 
- if other researchers using the same method on same material produce the 
same results 
- if reliability can be established results can be generalized. 
- Generally speaking, quantitative methods are more reliable than qualitative 
method. 
 
Validity 
A valid statement gives true measurement/description and explanation of what it 
claims to measure.  
It is accurate reflection of social reality.  
Data can be reliable without being valid. 
 
Alan Bryman outlines four types of validity 
1) Measurement validity/ Construct validity: 
Whether a measure that is employed really measures what it claims. 
Eg If IQ tests really measure intelligence 
 
2) Internal validity:  
If one thing is said to cause another, this explanation will be internally valid, if 
causal relationship is true.  
Ex Pierre Bourdieu- Different cultural settings produce different educational 
achievements.  
 
3) External validity:  
If study can be generalized to situations other than study itself. 
 
4) Ecological validity: 
How accurately a research mirrors natural setting or real experience.  
Lab experiments may lack ecological validity. 
 
Validation by respondents (of what they actually meant) overcome problems of 
validity 
Practicality: less time consuming and less personal commitments 
Ethics: researcher will choose topics where informed consent and confidentiality are 
possible 
  
 
 
Concepts 
Concepts are mental constructs or logical abstractions through which social scientists 
make sense of social reality. They are basically tools with which we think ,criticize, 
argue and explain. They serve as medium of communication among researchers. 
Variables 
Variables are specific characteristics or attributes of the more general concepts, or 
more specifically the attributes of events, objects and things that are observed and 
controlled by the researcher. They are empirical properties which take up one value 
or another. 
Earl R. Babbie- Variable is a logical set of attributes. 
Use- Variables help formulate hypothesis. In any social experiment , the researcher 
needs to identify the variables and then establish which of them are dependent and 
which are independent. 
Durkheim study of suicide-used multivariate or variable analysis.He studied the effect 
if independent variables like religion, gender and marital status on dependent 
variable viz suicide. 
Types: 
1.Dependent and Independent 
Dependent Variable: 
The variable that depends on other factors that are measured. These variables are 
expected to change as a result of an experimental manipulation of the independent 
variable or variables. It is the presumed effect. 
Independent Variable: 
The variable that is stable and unaffected by the other variables you are trying to 
measure. It refers to the condition of an experiment that is systematically manipulated 
by the investigator. It is the presumed cause. 
2.Qualitative and Quantitative 
Qualitative- variables which take non-numerical value 
Example- Gender, Religion 
Quantitative-variables that take up a numerical value 
Example-Age, Income 
3.Continuous and discreet variables 
Discreet variables have a definite value. Continuous Variables can’t be expressed as 
a particular value. Example: Studying in which class will elicit a number whereas the