Deck 11: The Process of Test Development

Question 1

The free response format, as opposed to multiple choice, is characterized by

A) increased difficulty in scoring
B) high variability in examinees' responses
C) a depth of thought not otherwise detectable
D) all of the above

Answer 1

D

Question 2

In order to interpret scores accurately, detailed information regarding _______ must be included.

A) the norming process
B) criterion validation studies
C) the test development
D) both b and c

Answer 2

A

Question 3

Concern over ______ in item writing involves the possibility of promoting sexist attitudes.

A) length of items
B) vocabulary
C) wording of items
D) gender-specific language

Answer 3

D

Question 4

A critical step in the test development process is

A) using a variety of samples for norming
B) keeping the test manual up-to-date
C) establishing cutoff scores
D) analyzing weak distractors

Question 5

________ generally contain items that are transparent with respect to what they are attempting to measure.

A) Rational scales
B) Empirical scales
C) Factor-analytic scales
D) Orthogonal scales

Question 6

People who are incapable of completing the test accurately, or who lack motivation, may exhibit ______ on a test.

A) random responding
B) dissimulation
C) social desirability
D) all of the above

Question 7

The factor-analytic approach to scale development yields instruments

A) with scales that do not vary no matter what method of factor analysis is used
B) with scales that always translate to clinically useful concepts
C) with psychometric properties generally superior to those constructed by the rational method
D) none of the above

Question 8

When developing norms, one must

A) define the target population
B) select the sample
C) standardize conditions
D) all of the above

Question 9

The type of response format used may be dictated by the ______ of the test.

A) content
B) length
C) purpose
D) scoring system

Question 10

One particular drawback of items generated from theory is that

A) examinees often respond to such items as they would like to appear
B) examinees often respond to such items as they actually perceive themselves
C) there is often no common theme that binds the items together
D) such items are often less subject to examinee distortion

Question 11

In developing norms, it is important to

A) select samples that represent the target population
B) obtain a representative sample from the general population
C) obtain educational backgrounds on all samples
D) all of the above

Question 12

Conscious efforts of the test taker to produce a distorted picture is

A) random responding
B) dissimulation
C) response style
D) biased responding

Question 13

For information regarding the purpose, development, reliability and validity of a test, one should consult

A) past test users
B) past research in which the test was used
C) the test manual
D) none of the above contains all of this information

Question 14

The most popular response format involves the use of multiple-choice items due to

A) the ease in scoring
B) low error in measurement
C) ease in item writing
D) both a and b

Question 15

The most practical solution to the need to standardize the testing procedure is to

A) standardize the physical environment
B) standardize the administration process
C) use a correction formula
D) not bother trying; it cannot be done

Question 16

A general tendency to agree or disagree with statements without attending to the content of the statements themselves is a

A) response set
B) random responding
C) response style
D) dissimulation

Question 17

________ scales allow the test user to discriminate one group from another.

A) Rational
B) Empirical
C) Factor-analytic
D) Orthogonal

Question 18

Revisions should be made

A) when item content becomes dated
B) every ten years
C) when test builds a substantial data base due to frequent use
D) both a and c

Question 19

Test developers circumvent the problem of items becoming out-dated or obsolete by

A) writing items that are timeless
B) replacing them with new items with similar item characteristics
C) rewriting items and developing new norms
D) doing nothing as it is a trivial problem

Question 20

The method most likely to be used when developing large-scale norms is

A) relaxed random sampling
B) standardization of conditions
C) a variation on cluster sampling
D) strict random sampling

Question 21

Tests should be normed

A) nationally
B) regionally
C) locally
D) either nationally, regionally, or locally, depending on the purpose of the testing

Question 22

Which of the following statements is true?

A) Tests which have not been revised in the last ten years should not be used.
B) Tests with dated content should never be used.
C) Newer tests are not necessarily better than their older counterparts.
D) Always choose the newest available measure of a construct.