Populations and samples

Question 1

why do we sample?

Answer 1

unpractical/too costly/too time consuming to survey a whole population

Question 2

advantages of simple random samples

Answer 2

• remove researcher bias
• represents a whole population effectively if using a large sample
• less complicated than stratified or quota sampling
• doesn’t require background info at individual or aggregate level
• doesn’t require a complex system of stratification

Question 3

convenience sampling definition

Answer 3

-non-probability sample made up of people who are easy to reach

Question 4

opportunistic sampling

Answer 4

-taking the opportunity when it arises

Question 5

theoretical/purposive sampling

Answer 5

selecting cases with a particular rationale (e.g. maximum diversity) to develop the theory as it emerges (simultaneous manner). closely associated with grounded theory methodology

Question 6

snowball sampling

Answer 6

used for hard to reach groups. get one contact and they introduce you to others e.g. criminal underworld

Question 7

quota sampling definition

Answer 7

-based on selecting cases opportunistically until particular quotas have been filled

Question 8

negatives of quota sampling

Answer 8

not random as some groups/people easier to ask than others

Question 9

stratified random sampling

Answer 9

stratas are formed based on members’ shared attributes or characteristics. a random sample from each strata is taken in s number proportional to the stratum’s size when compared to the population. these sub-sets of the strata form a random sample

Question 10

negatives of stratified random sampling

Answer 10

• requires individual level data

- can get complicated to produce/time consuming

Question 11

stages of generating a simple random sample

Answer 11

1. identify population of interest
   -often ignored by researchers
   -you cannot generalise beyond this population
2. generate sampling frame
   -uses individual level data
   -each case must be identifiable and contactable
   e.g. UOL student records
3. select sample size
   -based off the most complex bit of analysis rather than population size
   -large samples always preferable as more likely to reflect the population we’re interested in
4. select sample
   -either using random number generator or a randomised sampling frame
   -

Question 12

sampling frame defintion

Answer 12

a selection of cases from the population from which the sample is taken

Question 13

where do sampling frames come from?

Answer 13

generated by the researcher, encompassing the population of study. often too many cases to use all the data so a sample frame effectively limits this

Question 14

examples of sampling frames

Answer 14

• telephone directory

- electoral register