Problem 5

Question 1

probability sampling

Answer 1

each member has (ideally equal) probability of being in sample

Question 2

simple random sampling

Answer 2

randomly chosen person, equal chance for everyone; no bias

Question 3

stratified sampling

Answer 3

dividing population in segments, select sample of equal size from each segment; high representativeness, maybe overrepresentation of one population

Question 4

proportionate sampling

Answer 4

equal proportions of segments in sample

Question 5

systematic sampling

Answer 5

sampling every xth element after random start

Question 6

cluster sampling

Answer 6

identifying natural occurring groups

Question 7

multistage sampling

Answer 7

identifying clusters & selecting sample from them

Question 8

representativeness

Answer 8

sample closely matches the characteristics of population; without sampling results can’ be generalized

Question 9

sampling error

Answer 9

extent to which characteristics of sample differ from those of population

Question 10

sample size

Answer 10

influenced by acceptable amount of sampling error & magnitude of difference you expect to find,

• -> large sample: expecting large amount of uncontrolled variables, small acceptable error
• -> small sample: high controls over variables

Question 11

internal validity

Answer 11

ability of design to test the hypothesis it was designed to test

threats:
- history: specific events occur between observations
- maturation: performance changes due to age, mental state
- testing: testing prior to the treatment changes how subjects respond in post-treatment testing
- instrumentation: unobserved changes in observer criteria or instrument calibration confound the effect of the treatment
- biased selection of subjects: subjects with extreme scores more closer to mean
- experimental morality: loss of participants

Question 12

external validity

Answer 12

results can be generalized

threats:
- reactive testing: pre-test affects participants’ reaction to an experimental variable, participants’ responses are unrepresentative of the general population
- interactions between participant selection biases & independent variable: observed effects may apply only to the participants included in the study
- reactive effects of experimental arrangements: highly artificial experimental situations used in some research & the participant’s knowledge that he or she is a research participant (= participants bias)
- multiple treatment interference: when participant are exposed to multiple experimental treatments, exposure to early treatments affects responses to later treatments