research methods exam 2

survey/poll

method of asking people questions face to face, otp, written questionnaires or online

threats to construct validity

• question formats

• well worded questions

• accurate responses

open ended questions

allows people to answer in whatever way they choose and how much they elaborate

likert scale

rating scale to reflect degree of agreement

forced choice

choose the best of 2 or more options

semantic differential format

numeric scale anchored by adjectives

common pitfalls in question wording

• leading questions

• double barreled

• negatively worded questions

• question order

everyday and major discrimination scale

• Drs. Jackson and Williams opted to develop simple and objective questions

memory can be faulty

confidence does not reliably indicate memory accuracy

lotus and palmer experiment

use of different verbs to see how that changes someone perception

behavioral observations

researcher watches people/animals and systematically records their actions

3 threats to behavioral observations

• observer bias

• observer effects

• reactivity

observer bias

observers’ expectations influence their interpretation of the participants behaviors or the outcome of the study

observer effects

influence the behavior of those they are observing to conform to their expectations through subtle, likely unintentional manipulations

how to lessen observer bias

codebooks

how to lessen observer effects

have high interrater reliability or a masked/blind design

reactivity

change in behavior when study participants know another person is watching

how to lessen reactivity

-unobtrusive design

-acclimate

-measure indirectly

double blind trials

neither the patient nor dr know whether they are receiving/giving the active tx or placebo

population of interest

entire set of people or things in which you are interested in

sample

smaller set of people or things taken from the population

examples of biased/unrepresentative samples

• convenience samples

• purposive samples

• snowball samples

• quota samples

convenience sample

sampling only those individuals who are easiest to contact/access

purposive samples

interested in certain kinds of people so only recruit those types of participants in a non-random way

snowball samples

participants recommend others for the study

quota samples

set a quota for each category and use nonrandom sampling until quotas are filled

examples of unbiased/representative samples

• simple random sampling

• cluster samples

• multistage cluster

• stratified random sampling

simple random sampling

putting every members name/ID of your population of interest in a pool and then randomly selecting a deermined number to include in your sample

cluster samples

clusters of participants within a population of interest are randomly selected and then all individuals in each selected cluster are used

multistage cluster

randomly select 4 preschools, then from there a random sample of people is chosen

stratified random sampling