module 6 Flashcards
reliability
an indication of the inconsistency or stability of a measuring instrument
error in measurement
a measurement is a combination of the true score and an error score
true score
score on the measuring instrument if there were no error
observed score=true score+measurement error
reliability=true score/(true score+error score)
correlation coefficient
a measure of the degree of relationship between 2 sets of scores
-1.00 to +1.00 (0.00 is no relationship)
positive correlation (direct)
an increase in one variable is related to an increase in another (same with decrease)
negative correlation (inverse)
an increase in one variable is related to a decrease in the other and vice versa
types of reliability
test/retest
alternate-forms
split-half
interrater
test/retest reliability
determined by assessing the degree of relationship between scores on the same test administered on 2 different occasions
- assesses stability of a test over time
- practice effects
alternate-forms reliability
determined by assessing the degree of a relationship between scores on 2 equivalent tests
split-half reliability
determined by correlating scores on one half with scores on other half of a measure
- uses spearman brown and kr20 statistics
interrater reliability
assesses the agreement of observations made by 2 or more raters/judges
- needs to be established by percentage of agreement among the raters
interrater=# of agreements/# of possible agreements x100
validity
indication of whether the instrument measures what it claims to measure
types of validity
content
face
criterion
construct
content validity
extent to which a measuring device covers a representative sample of the domain of behaviors to be measured
face validity
appears valid on its surface