Deck 3: The Measurement of Behavior

Question 1

A perfectly reliable measure

A)has a reliability of .00.
B)has no measurement error.
C)is valid.
D)has as much true score variance as error variance.

Answer 1

B

Question 2

If you wanted to reduce measurement error in a study, which would you be most likely to do?

A)Use items with low item-total correlations.
B)Use face valid measures.
C)Measure all participants under the same conditions.
D)All of these would reduce measurement error.

Answer 2

C

Question 3

A researcher who asks people to answer questions on questionnaires is using a(n)_________ measure.

A)observational
B)self-report
C)ratio
D)affective

Answer 3

B

Question 4

A researcher counting the number of times a rat presses a bar is using a(n)______ measure.

A)observational
B)self-report
C)cognitive
D)physiological

Question 5

High school class graduation rank is measured on a(n)_________ scale.

A)interval
B)nominal
C)ratio
D)ordinal

Question 6

Researchers generally want two administrations of the same test to

A)correlate at least .05.
B)correlate at least .50.
C)correlate at least .70.
D)correlate 1.00.

Question 7

Intelligence is measured on a(n)__________ scale.

A)ordinal
B)nominal
C)interval
D)ratio

Question 8

If scores on a test are consistent over time, the test has

A)interitem reliability.
B)test-retest reliability.
C)interrater reliability.
D)item-total reliability.

Question 9

Which type of measure tends to require sophisticated equipment?

A)Behavioral
B)Psychometric
C)Archival
D)Physiological

Question 10

The ratio of true score variance to total variance reflects a measure's

A)reliability.
B)measurement error.
C)validity.
D)variability.

Question 11

A participant's observed score consists of

A)reliability and validity.
B)convergent and discriminant validity.
C)transient and stable variables.
D)measurement error and true score.

Question 12

Which of the following does not belong with the others?

A)Interitem reliability
B)Cronbach's alpha coefficient
C)Item-total correlation
D)Test-retest reliability

Question 13

The measurement of weight is on a(n)_______ scale.

A)ordinal
B)nominal
C)ratio
D)interval

Question 14

As a rule of thumb, a measure is considered sufficiently reliable if

A)it is valid.
B)there is no measurement error.
C)at least half of the total variance is true-score variance.
D)its correlation is 1.00.

Question 15

Which scale of measurement provides the greatest amount of information about a construct?

A)Ordinal
B)Nominal
C)Ratio
D)Interval

Question 16

The relationship between measurement error and reliability is

A)inverse.
B)direct.
C)curvilinear.
D)none; measurement error and reliability are not related.

Question 17

Measuring a construct with several different kinds of measures is called

A)converging operations.
B)scaling.
C)strong inference.
D)convergent validity.

Question 18

The field that is devoted to the study of psychological measurement is called

A)statistics.
B)evaluation.
C)psychometrics.
D)phrenology.

Question 19

If all participants' observed scores equal their true scores,

A)the measure is valid.
B)measurement error is high.
C)reliability is higher than validity.
D)measurement error is zero.

Question 20

Political preference (Democrat, Republican, Independent)is measured on a(n)___________ scale.

A)ordinal
B)nominal
C)trichotic
D)interval

Question 21

What does a psychometrician do?

Question 22

Why is interitem reliability not a concern when using formative measures?

A)The individual items do not measure the same construct.
B)They are observational measures, so no items are being used.
C)All formative measures are perfectly reliable.
D)When using formative measures, total variance equals true score variance.

Question 23

Why do researchers prefer, when possible, to use measures that are on a ratio scale of measurement?

Question 24

A measure that measures what it appears to measure has good ________ validity.

A)criterion-related
B)face
C)construct
D)convergent

Question 25

Concurrent is to ______ as predictive is to

A)reliability; validity.
B)behavior; construct.
C)criterion-related; construct.
D)now; later.

Question 26

To assess a measure's construct validity, researchers

A)examine its correlations with future behaviors.
B)test hypotheses about it using controlled experiments.
C)see how strongly it correlates with measures of related constructs.
D)use all of the tactics listed to assess construct validity.

Question 27

Which of the following is most important in assessing the adequacy of the SAT in admitting students to college?

A)Construct validity
B)Concurrent validity
C)Face validity
D)Predictive validity

Question 28

A measure that does not correlate with measures with which it should not be related has

A)convergent validity.
B)discriminant validity.
C)concurrent validity.
D)no validity.

Question 29

Which of the following is impossible?

A)A valid measure is unreliable.
B)A reliable measure is valid.
C)A reliable measure is not valid.
D)An unreliable measure is not valid.

Question 30

What kind of reliability would be relevant in a study in which two observers recorded people's behaviors during a laboratory interaction?

A)Interpersonal
B)Convergent
C)Interrater
D)Observational

Question 31

What are the three primary types of measures that behavioral researchers use?

Question 32

When a particular measure is not equally valid for everyone, ______ occurs.

A)discriminant validity
B)low reliability
C)psychometric contamination
D)test bias

Question 33

Describe each of the four scales of measurement, and give an example of each.

Question 34

Researchers usually consider a measure to have adequate interitem reliability if Cronbach's alpha coefficient

A)is zero.
B)exceeds .05.
C)exceeds .50.
D)exceeds .70.

Question 35

Test bias is demonstrated when

A)the average score for one group is lower than the average of another group.
B)predictive validity is different for different groups.
C)the means of the scores for two groups do not differ.
D)scores on the test do not predict the target outcome for one group.

Question 36

Why do researchers use converging operations?

Question 37

Which of the following is not a hypothetical construct?

A)Dominance
B)Stuttering
C)Id
D)Creativity

Question 38

Which of the following is true of using shoe size as a measure of intelligence?

A)It is a reliable measure.
B)It is an unreliable measure.
C)It is a valid measure.
D)It is unreliable but valid.

Question 39

A measure that correlates with other measures as it should has

A)convergent validity.
B)discriminant validity.
C)interitem validity.
D)concurrent validity.

Question 40

An item-total correlation is the correlation between

A)an item on a scale and the sum of the other items on that scale.
B)an item on a scale and the total variance.
C)scores on a multi-item scale and a set of other scales.
D)two items from a larger questionnaire.

Question 41

Can a measure be valid yet unreliable? Explain.

Question 42

Why is Cronbach's alpha coefficient the preferred index of interitem reliability?

Question 43

Why is interitem reliability a concern for reflective measures but not for formative measures?

Question 44

Describe what reliability is from the standpoint of variance.

Question 45

How can you tell whether a test is biased?

Question 46

Distinguish between construct and criterion-related validity.

Question 47

What is a hypothetical construct? Give two examples.

Question 48

The score we obtain for a participant is composed of two components. What are they?

Question 49

What does it indicate about a measure if interrater reliability is low?

Question 50

List four things that can decrease the reliability of a measure.

Question 51

Can a measure be reliable but not valid? Explain.

Question 52

In examining the criterion-related validity of employment tests, which type of validity-predictive or concurrent-is most important? Explain.

Question 53

Distinguish among test-retest, interitem, and interrater reliability. For what kind of measure is each type of reliability relevant?

Question 54

Distinguish between convergent and discriminant validity.

Question 55

Why is face validity sometimes important?