Questions & Answers, latest update.
Verified.
Power tests
a. use time to prevent perfect scores
b. are concerned with normal performance
c. assess how quickly the test taker can understand the question and choose the right answer
d. prevent perfect scores by including difficult test items - ✔✔-d. prevent perfect scores by including
difficult test items
The ___ is NOT an example of a subjective test.
a. Rorschach inkblot test
b. sentence completion test
c. Mental Status Exam (MSE)
d. MBTI - ✔✔-d. MBTI
Which of the following is NOT true concerning construct validity?
a. It's empirically established by examining the relationship b/t data collected from the instrument and a
specific criterion.
b. It's important to establish when an instrument measures an abstract construct.
c. It's defined as the extent to which an instrument measures a theoretical construct.
d. It can be determined through experimental design or factor analysis. - ✔✔-a. It's empirically
established by examining the relationship b/t data collected from the instrument and a specific criterion.
,If an individual's mean score after multiple test administrations is 84 and the SEM is 5, what can be said
about the distribution of his observed scores?
a. 95% of the time his observed scores will fall b/t 89-79
b. 68% of the time his observed scores will fall b/t 89-79
c. 68% of the time his observed scores will fall b/t 91-74
d. there isn't enough info provided to make a determination - ✔✔-b. 68% of the time his observed
scores will fall b/t 89-79
SEM = Standard Error of Measurement
= Standard Deviation (SD) of an individual's repeated test scores when administered the same test
multiple times
- computed using SD and instrument's reliability coefficient
- the larger the SEM, the lower the reliability of the test
- 68% of all scores fall w/in +/- 1 SEM from the mean; 95% of all scores fall w/in +/- 2 SEM; 99.7% of
scores fall w/in +/- 3 SEM
The developer of a challenging test computes each item's difficulty. He finds that the majority of his test
items have an item difficulty score ranging from .90 to .95. What can be said about the test?
a. He needs to make his test items more difficult.
b. He needs to make his test items easier.
c. The test questions aren't consistent.
d. He doesn't need to make any corrections to the individual items. - ✔✔-a. He needs to make his test
items more difficult.
Average item difficulty b/t .90-.95 = average item being answered correctly by 90-95% of test takers
- item difficulty calculated by dividing number of test takers answering correctly by total number of test
takers > calculation expressed as decimal point call p value ranging from 0-1.0
- test developers generally want average item difficulty to have average p value of .50 > since 50% of
test takers would miss item, that p value yields most variation in test score distribution
, What percentage of a normally distributed population's scores fall b/t +/- 3 standard deviations?
a. 95%
b. 99%
c. 68%
d. 2% - ✔✔-b. 99%
empirical 68-95-99.7 normal curve rule = 68% of all scores fall w/in +/- 1 SD from the mean; 95% of all
scores fall w/in +/- 2 SD; 99.7% of scores fall w/in +/- 3 SD
if a student scored a 45 on an exam with a mean of 35 and a standard deviation of 5, what would be the
student's z-score?
a. +2
b. -1
c. 0
d. +1.5 - ✔✔-a. +2
___ is NOT a diagnostic achievement test.
a. Key Math Diagnostic Test
b. Test of Adult Basic Education
c. Wide Range Achievement Test
d. Metropolitan Achievement Test - ✔✔-d. Metropolitan Achievement Test
common diagnostic achievement tests
- Key Math Diagnostic Test (Key Math 3) = assesses math-related learning disabilities
- Test of Basic Adult Education (TABE) = assesses basic skills adults need to participate fully in society