Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
22 Cards in this Set
- Front
- Back
sample vs. population
|
sample:subset of ppl who represent the population by participating in the study
population: the group of ppl that we'd like our study to represent/generalize to |
|
category data
|
frequency counts, qualititative
every participant put into category |
|
ranked
|
participant's position relative to others
1st, 2nd, 3rd |
|
scaled
|
quantitative data that doesnt depend on other participant's scores
each score stands on its own |
|
self-report measures
|
respond to questions from an interviewer or questionnaire
reactivity may be problem if ppl know what they are being measured free-format self report: stimulus material, reponses indicate feelings w/out constraints. ex projective measures |
|
fixed-format self report measures
|
presented w/ questions, response alternatives provided
more structured than free format likert scale |
|
likert scale
|
most popular, personaltiy and attitude research
series of items Rosenbergs self-esteem scale |
|
behavioral measures
|
potential for less reactivity
coding system, more than 1 rater beh measured in terms of frequency, duration, intensity, proximity, speed |
|
psychophysiological measures
|
assess the physiological functioning of the body's nervous and endocrine systems
may reduce reactivity brain activity:EEg, MRI, CAT stress; blood test, saliva test |
|
frequency distribution
|
=histogram
-normal(bell shape) -bimodal (2 lumps) -negatively skewed (more scores tail off at low end) -positively skewed(more scores tail off to high end) |
|
classical test theory
|
X=T+E
x:observed score t:true score e:error goal is X=T |
|
reliability
|
index is correlation
>.70 is acceptable types: test-retest,parellel forms, inter-rater, internal consistnecy |
|
Test-Retest reliabilty
|
stability of a measure over time
administer same measure to same folks twice calculate correlation of scores from time 1 to time 2 |
|
parellel forms reliability
|
consistency of scores from 1 version of a measure w/another version
administer 2 versions of same measure to same folks calculate correlation of scores from form1 and form 2 |
|
Inter-Rater reliability
|
consistency with which ppl provide rating
2 rater provide ratings of same ppl calculate correlation of scores given by rater 1 and 2 |
|
Internal consistency reliability
|
consistnecy of responses to items w/in a measure
split-half:divide measure into 2 (odd/even) calculate correlation |
|
coefficient alpha
|
average of all possible split halves
look at consistency of reponses across items |
|
construct validity
|
how well an instrument measures an abstract quality; cant be determined directly
|
|
content validty
|
logic/conceptual evidence
concerned w/ sampling the domain the construct represents ex.test w/ questions about peripheral stuff |
|
Convergent and discriminant validty
|
statistical evidence
measure variable of interest and related (convergent) and unrelated(disc) variables convergent:correlation should be positive discriminant:correlation should be zero |
|
scale
|
series of items, tap into unidemnsional construct
items should be intercorrelated |
|
reponse bias
|
deviant responding(random)
careless responding(mistakes) consistent resp(same rating for all items) omitting items extremity response(end points only) social desirability(to look good) aquiensence(agree with everything) |