Surveys and sampling methods

Question 1

Surveys are

Answer 1

a means of collecting primary data from large numbers of people, and are most commonly carried out using questionnaires or structured interviews

Question 2

Who uses surveys?

Answer 2

• favoured by positivists as they produce quantitative statistical data
• also used by governments (eg census), market researchers, election pollsters

Question 3

Example of using surveys - Townsend ‘Poverty in the United Kingdom’ (1979)

Answer 3

• produced a mass of statistical data
• used questionnaires about the causes and extent of poverty
• done by trained interviewers

Question 4

Sample definition

Answer 4

a smaller representative group drawn from the survey population

Question 5

Representative sample definition

Answer 5

a smaller group drawn from the survey population, of which it contains a good cross-section, such as the right proportions of people of different ethnicities, ages, social classes, or genders. The information obtained from a representative sample should provide roughly the same results as if the whole survey population had been questioned

Question 6

Survey population definition

Answer 6

the whole group being studied, and will depend on the hypothesis of the researcher

Question 7

Sampling frame definition

Answer 7

a list of the names of all those included in the survey population from which the sample is selected

Question 8

The representativeness of a sample is affected by

Answer 8

• sample size
• the sampling frame (eg the electoral register, school registers etc)
• the sampling method

Question 9

Sampling method definition

Answer 9

a technique sociologists use to select representative individuals to study from the survey population

Question 10

Example of a sampling method: Random sampling

Answer 10

• every individual has equal chance of being selected
• commonly done by numbering the names in the sampling frame and using a random number generator
• may result in an unrepresentative sample

Question 11

Example of a sampling method: Systematic sampling

Answer 11

• names are selected from the sampling frame at regular intervals (eg every 10th name)
• risk of being unrepresentative

Question 12

Example of a sampling method: Stratified random sampling

Answer 12

• stratifies the sampling frame into smaller groups (eg based on ethnicity) according to their proportion of the population, then randomly selects members of this smaller group
• makes the sample more representative

Question 13

Example of a sampling method: Quota sampling

Answer 13

• selecting people that fit into certain categories according to their proportion in the survey population as a whole
• eg choosing 10 middle class people, 10 working class and 1 upper class
• not necessarily representative - may be done by stopping people in the street, but this would exclude those at work or home
• may be researcher bias as they choose participants

Question 14

Example of a sampling method: Multistage/cluster sampling

Answer 14

• selecting a sample in various stages, each time selecting a sample from the previous sample until the final sample is finished
• eg in a national survey of school students, you could take a sample of schools, then a sample of students within the schools selected

Question 15

Example of a sampling method: Snowball sampling

Answer 15

• used when a sampling frame is difficult to access or doesn’t exist
• identify a couple of people who fit the desired characteristics and ask them to introduce you to more
• not random or representative - rely on volunteers recommending other volunteers so could be biased

Question 16

Example of using snowball sampling - Laurie Taylor ‘In The Underworld’ (1984)

Answer 16

• wanted to investigate the lifestyles of criminals
• knew a convicted criminal, who agreed to put him in touch with other criminals willing to cooperate with his research

Question 17

Non-representative sampling

Answer 17

in some cases a non-representative sample may be useful, eg if studying specifically young middle-class couples, it would be more useful to have more young middle-class couples in the sample

Question 18

Pilot survey definition

Answer 18

a small-scale practice survey carried out before the final survey to check for any possible problems (eg unclear questions) to avoid wasting time or resources in the the final survey

Question 19

Problems with the social survey

Answer 19

• Validity - statistical data may not present a genuine image, could fail to accurately depict motives and behaviour, use categories imposed by sociologists
• Generalisation - how far the findings can be generalised to the whole survey population
• Reliability - whether it can be repeated by a different researcher on different participants to give similar results

Question 20

The stages of a survey

Answer 20

1. state the hypothesis
2. operationalise concepts
3. define survey population and draw up sampling frame
4. select a sample
5. decide research method (questionnaire, interview etc)
6. do a pilot survey and evaluate
7. conduct final full-scale survey
8. analyse and interpret findings
9. prepare a report on findings