Data Collection Through Sampling (Exam 1)

Question 1

sampling frame

Answer 1

list of all individuals belonging to the population

Question 2

sampling design

Answer 2

describes exactly how to choose a sample from the sampling frame (or from the general idea of the population if sampling frame not available)

Question 3

biased sampling designs

Answer 3

• design either over/under emphasizes some characteristics of the population based on the procedure used to select individuals from the sample
• all individuals in the population don’t have an equal chance of being sampled
• flaw: the sample will NOT be representative

Question 4

unbiased sampling designs

Answer 4

• all individuals in the population have an equal chance of being sampled
• on average, the sample will be representative of the population

Question 5

convenience sample

Answer 5

• a sample obtained by selecting individuals in the population that are easiest to reach (so likely share characteristics)
• produce unrepresentative results
• individuals in the population who are easiest to access

Question 6

voluntary response sample

Answer 6

• consists of the people who choose to respond to a broad invitation (either don’t get neutral/unbiased opinions participating in study or only get polarizing views from those who care a lot), results in too many strong opinions
• under-samples individuals who are neutral, over-samples strong-beliefs people
• individuals in the population who opt into the sample

Question 7

simple random sample

Answer 7

• unbiased
• selects individuals from the sampling frame through pure randomization
• most basic approach
• basically like a random number generator
• most likely to be representative (but due to chance, it might not be)
• we assume that we have a sampling frame listing all individuals in the population from which we randomly make selections

Question 8

stratified random sample

Answer 8

• unbiased
• separates the population into mutually exclusive groups (strata) and then draws simple random samples from each stratum
• we assume that we have a sampling frame listing all individuals in the population from which we randomly make selections
• choose the strata based on characteristics known beforehand that are believed to influence the variable(s) of interest in the study
• sample proportionately

Question 9

stratum

Answer 9

• a subset of individuals in the population who are grouped because they share a common characteristic believed to affect the variable of interest
• ex: stratify be region, age, gender, etc

Question 10

cluster sampling

Answer 10

• if we can’t obtain a sampling frame that includes all individuals in the population, we hope to be able to obtain a sampling frame of mutually exclusive groups (“clusters”) that includes all individuals in the sampling frame
• can randomly sample these clusters and gather data from ALL individuals in the selected cluster

Question 11

cluster

Answer 11

• believed to be a representative group from the population, not grouped by any feature believed to affect the variable of interest, mutually exclusive groups
• ex: neighborhoods, dorms

Question 12

multistage sampling

Answer 12

• combines multiple cluster samples in stages
• we can randomly sample these clusters, then randomly sample small sub-clusters (within original clusters), etc… until we are able to actually select a reasonable number of individuals belonging to our population