PSYC 426 final exam with answers
3 components of validity - ANSWERS✔✔CI: Construct of Interest
SE: Systematic Error of Measurement
RE: Random Error of Measurement
example of CI and SE influence on test - ANSWERS✔✔you have a test of math reasoning (CI)
scores can be influenced by ability to understand instructions, vocab (SE)
classical test theory of validity definition/formula defining validity - ANSWERS✔✔validity is the
proportion of the relevant construct over the total variance
𝜎𝐶𝐼2 /𝜎𝑋2
Classical Test Theory Validity - ANSWERS✔✔ox2 = ot2 + oe2
ot= oci2 + ose2
ox2 = oci2 + ose2 + ore2
validity splits true score variance into construct of interest variance and systematic error variance
(what you want to measure and what you don't want to measure)
Validity is about figuring out what you are really measuring
,why do we need a reliable measure to have a valid measure - ANSWERS✔✔there will be lots of
measurement error if we dont
types of validity (5) - ANSWERS✔✔face validity
content validity
criterion related validity (many)
experimental validity
construct validity
face validity - ANSWERS✔✔test seems to be measuring what it is supposed to
extent to which a test is subjectively viewed as covering the construct it is supposed to assess
not real validity, a test doesn't really need this
What is the usefulness of face validity - ANSWERS✔✔cooperation, acceptation
what are the limits of face validity - ANSWERS✔✔bias, social desirability
content validity (what is it and how is it assessed) - ANSWERS✔✔The test covers key aspects of the
construct it aims to assess (includes a representative sample of the target behaviors) - important
This form of validity is typically assessed by experts. Ask to rate items, suggest some items, not
assessed by numers
,following proper principles when developing a test
criterion related validity types - ANSWERS✔✔concurrent
predictive
congruent
convergent
discriminant
discriminative
criterion related validity - ANSWERS✔✔Established via the comparison of test scores with
"objective" criterion assumed to provide some "true" reflection of the underlying construct.
concurrent validity - ANSWERS✔✔your test + criterion is administered simultaneously
assessed with correlation
predictive validity - ANSWERS✔✔the criterion test is administered after yours
assessed with regression
congruent validity - ANSWERS✔✔the criterion test assess the same construct as yours
convergent validity - ANSWERS✔✔the criterion assess a related construct to yours (depression and
anxiety)
discriminant validity - ANSWERS✔✔the criterion assess a construct that is either the opposite of
yours (negative correlation) OR unrelated to yours (no correlation)
discriminative validity - ANSWERS✔✔the criterion is categorical and you want to predict group
membership
, 2 ways to assess discriminative validity - ANSWERS✔✔Mean comparisons: Groups (serving as the
criterion) are compared based on their test scores treated as continuous variables (norm-referenced
or not).
For example, one could compare engineers and musicians scores on a test of "musical abilities" using
a t-test (or ANOVA, MANOVA, etc.).
Chi-Square: Groups (serving as the criterion) are compared based on their test scores treated as
categorical variables (criterion-referenced or not).
For example, one could compare the frequency of individuals receiving a diagnosis of bipolar
disorder on the basis of their test scores in a group of psychology students and a group of psychiatric
patients.
proportion of variance related to the construct (in context of assessing convergent and congruent
validity) - ANSWERS✔✔rxy2= CI2/ox2
proportion of systematic error (in context of assessing convergent and congruent validity) -
ANSWERS✔✔1-rxy2
oes2/ox2
why is random error not important (in context of assessing convergent and congruent validity) -
ANSWERS✔✔because it's unrelated to anything else, follows normal distribution
when predictive congruent/convergent validity is assessed using a regression, how is validity
interpreted - ANSWERS✔✔There is always a discrepancy between the observed score on the
criterion (Y) and the score that is predicted (Y') based on the test scores (X), unless validity is perfect
(which never happens).
the extent to which there is a difference = the index of validity
prediction error - ANSWERS✔✔the difference between observed score on criterion (Y) ans predicted
score(Y') based on test scores (x)
3 components of validity - ANSWERS✔✔CI: Construct of Interest
SE: Systematic Error of Measurement
RE: Random Error of Measurement
example of CI and SE influence on test - ANSWERS✔✔you have a test of math reasoning (CI)
scores can be influenced by ability to understand instructions, vocab (SE)
classical test theory of validity definition/formula defining validity - ANSWERS✔✔validity is the
proportion of the relevant construct over the total variance
𝜎𝐶𝐼2 /𝜎𝑋2
Classical Test Theory Validity - ANSWERS✔✔ox2 = ot2 + oe2
ot= oci2 + ose2
ox2 = oci2 + ose2 + ore2
validity splits true score variance into construct of interest variance and systematic error variance
(what you want to measure and what you don't want to measure)
Validity is about figuring out what you are really measuring
,why do we need a reliable measure to have a valid measure - ANSWERS✔✔there will be lots of
measurement error if we dont
types of validity (5) - ANSWERS✔✔face validity
content validity
criterion related validity (many)
experimental validity
construct validity
face validity - ANSWERS✔✔test seems to be measuring what it is supposed to
extent to which a test is subjectively viewed as covering the construct it is supposed to assess
not real validity, a test doesn't really need this
What is the usefulness of face validity - ANSWERS✔✔cooperation, acceptation
what are the limits of face validity - ANSWERS✔✔bias, social desirability
content validity (what is it and how is it assessed) - ANSWERS✔✔The test covers key aspects of the
construct it aims to assess (includes a representative sample of the target behaviors) - important
This form of validity is typically assessed by experts. Ask to rate items, suggest some items, not
assessed by numers
,following proper principles when developing a test
criterion related validity types - ANSWERS✔✔concurrent
predictive
congruent
convergent
discriminant
discriminative
criterion related validity - ANSWERS✔✔Established via the comparison of test scores with
"objective" criterion assumed to provide some "true" reflection of the underlying construct.
concurrent validity - ANSWERS✔✔your test + criterion is administered simultaneously
assessed with correlation
predictive validity - ANSWERS✔✔the criterion test is administered after yours
assessed with regression
congruent validity - ANSWERS✔✔the criterion test assess the same construct as yours
convergent validity - ANSWERS✔✔the criterion assess a related construct to yours (depression and
anxiety)
discriminant validity - ANSWERS✔✔the criterion assess a construct that is either the opposite of
yours (negative correlation) OR unrelated to yours (no correlation)
discriminative validity - ANSWERS✔✔the criterion is categorical and you want to predict group
membership
, 2 ways to assess discriminative validity - ANSWERS✔✔Mean comparisons: Groups (serving as the
criterion) are compared based on their test scores treated as continuous variables (norm-referenced
or not).
For example, one could compare engineers and musicians scores on a test of "musical abilities" using
a t-test (or ANOVA, MANOVA, etc.).
Chi-Square: Groups (serving as the criterion) are compared based on their test scores treated as
categorical variables (criterion-referenced or not).
For example, one could compare the frequency of individuals receiving a diagnosis of bipolar
disorder on the basis of their test scores in a group of psychology students and a group of psychiatric
patients.
proportion of variance related to the construct (in context of assessing convergent and congruent
validity) - ANSWERS✔✔rxy2= CI2/ox2
proportion of systematic error (in context of assessing convergent and congruent validity) -
ANSWERS✔✔1-rxy2
oes2/ox2
why is random error not important (in context of assessing convergent and congruent validity) -
ANSWERS✔✔because it's unrelated to anything else, follows normal distribution
when predictive congruent/convergent validity is assessed using a regression, how is validity
interpreted - ANSWERS✔✔There is always a discrepancy between the observed score on the
criterion (Y) and the score that is predicted (Y') based on the test scores (X), unless validity is perfect
(which never happens).
the extent to which there is a difference = the index of validity
prediction error - ANSWERS✔✔the difference between observed score on criterion (Y) ans predicted
score(Y') based on test scores (x)
