Deck 6: Choosing the Best Measure for Your Study

Question 1

When choosing between several measures, you should

A) choose the one with the highest validity
B) choose the one with the highest sensitivity
C) choose the one with the highest level of measurement
D) can't tell from the information given

Answer 1

can't tell from the information given

Question 2

Sensitivity can be improved by

A) increasing reliability
B) increasing validity
C) increasing the number of scale points
D) all of the above

Answer 2

all of the above

Question 3

Sensitivity can be improved by all but one of the following. Which is it?

A) increasing reliability
B) adding scale points
C) using a higher scale of measurement
D) increasing validity

Answer 3

using a higher scale of measurement

Question 4

A measure has the following weaknesses. Which is the LEAST serious threat to its validity?

A) vulnerability to observer bias
B) vulnerability to social desirability bias
C) susceptibility to demand characteristics
D) insensitivity

Question 5

Generally, which measure would probably have the most sensitivity?

A) a "yes/no" measure
B) a 3-point rating scale
C) a 7-point rating scale
D) a 100-point rating scale

Question 6

Which of the following would NOT increase sensitivity?

A) using a behavior that was resistant to change
B) using scientific equipment
C) measuring response latencies rather than counting number of responses
D) adding scale points to a rating scale

Question 7

Which of the following statements are true?

A) the more valid a measure, the more sensitive it will tend to be
B) in choosing a measure, the goal of sensitivity may conflict with the goal of validity
C) the more complex the behavior, the more sensitive it will tend to be
D) both a and b

Question 8

Which is most harmful to a measure's sensitivity?

A) low reliability
B) low validity
C) low scale of measurement
D) a wide range of scores

Question 9

Which is most harmful to a measure's sensitivity?

A) low reliability
B) subject biases
C) low level of measurement
D) using simple behaviors

Question 10

Which of the following measures is LEAST likely to be sensitive?

A) a latency measure
B) a 7-point rating scale
C) a noise measure that relies on a sound meter
D) a measure that produces a limited range of scores

Question 11

Which of the following is GOOD advice?

A) Don't use a latency measure
B) Don't pilot test your measure
C) If you use a rating scale, don't use anything less than an 11-point rating scale
D) Don't use a measure that produces a limited range of scores

Question 12

Random error

A) makes scores vary
B) decreases sensitivity
C) both a and b
D) neither a nor b

Question 13

All of the below are true EXCEPT

A) not all numbers are equally descriptive
B) not all measures produce the same kinds of numbers
C) a higher scale of measurement is needed to answer some research questions than to answer others.
D) nominal numbers are on more sophisticated level of measurement than ordinal numbers

Question 14

Which of the following is NOT a nominal number?

A) your social security number
B) your car's license plate number
C) a basketball player's uniform number
D) your age

Question 15

If we only need to know whether two groups differ from one another, we need at least ______ level measurement.

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

Question 16

If we want to classify people into different types, we would probably use _____ numbers.

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

Question 17

If we want to know if one group has MORE of a quality than another group, we can't use a measure that produces ____ numbers.

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

Question 18

If we want to know if one group has MORE of a quality than another group, our measure must produce numbers that are at least

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

Question 19

When a professor ranks essay tests from best to worst, she is placing the tests on a(n) _______ scale.

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

Question 20

____ numbers allow you to tell that one group has more of a quality than another group, but does not let you know how much more of the quality it has.

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

Question 21

Ranked data are

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

Question 22

Which of the following measurement scales has equal scale units along the entire scale?

A) ordinal
B) interval
C) ratio
D) b and c

Question 23

Rating scale data are assumed to be

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

Question 24

The simplest level of measurement is

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

Question 25

The most complex level of measurement is

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

Question 26

Shifts to more complex levels of measurement mean that your numbers contain

A) more information
B) less information
C) more sensitivity
D) less sensitivity

Question 27

If movies are rated on a scale from no stars (terrible) to four stars (outstanding), most psychologists would assume that this measurement is

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

Question 28

If a professor correlates hours of that students work with the number of academic units they are carrying during the current term, measurement of academic units carried should be:

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

Question 29

The property of ratio measurement that makes it different from other scales is

A) equal intervals
B) order
C) classification
D) a true zero

Question 30

If we want to determine whether people who have therapy are twice as happy as those who don't have therapy, we need ___ scale of measurement.

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

Question 31

Which type of number is a psychological measure LEAST likely to produce?

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

Question 32

A race horse, wearing number three, finishes second, with a time of 1.50 minutes. That statement provides almost pure examples of each scale of measurement EXCEPT

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

Question 33

Six is twice three only when these numbers emerge from a _________ scale of measurement.

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

Question 34

Measurements of time are at the _______ scale level.

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

Question 35

If you ranked 12 students according to their physical attractiveness, you would most likely generate a(n) ______ scale.

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

Question 36

All of the following are nominal numbers except

A) a telephone number
B) the number of students in a class
C) a social security number
D) the 19 on a quarterback's jersey

Question 37

The highest level of measurement is a(n) ______ scale, which has a true zero.

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

Question 38

In order to say that Sue is twice as intelligent as Joe, intelligence must be measured with a _______ scale.

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

Question 39

If you are trying to say that one treatment is half as effective as another, you need ____ data.

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

Question 40

All valid measures can provide data that are at least ____

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

Question 41

If we want to know if one group has changed more than another group, we need at least ____ scale measurement.

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

Question 42

Scores on a measure range from 1 to 7. If the difference between how a person scoring a "1" and a person scoring a "3" feels is psychologically the same as the difference between how a person scoring a "4" and a person scoring a "6" feels, your scale provides ____ scale measurement.

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

Question 43

A measure with face validity

A) correlates highly with measures of the same construct
B) has internal validity
C) has construct validity
D) appears, to the average person, to measure what it is supposed to measure

Question 44

A measure with face validity

A) has internal validity
B) may be vulnerable to the social desirability bias
C) produces reliable results
D) has construct validity

Question 45

A measure of how much time a couple spends looking into each other's eyes is:

A) a ratio measure of gazing time; and a ratio measure of love
B) a ratio measure of gazing time; perhaps only an ordinal measure of love
C) an interval measure of gazing time; and a ratio measure of love
D) an interval measure of gazing time; and an interval measure of love

Question 46

A measure's ability to detect small differences is called reliability.

Question 47

Nominal scale numbers are the least informative type of number used in research.

Question 48

Ratio scale numbers are the most informative type of number used in research.

Question 49

Ordinal scale numbers permit you to talk about specific amounts of a quality

Question 50

Interval scale numbers permit you to talk about how many times more of a quality one participant has relative to another.

Question 51

Both interval and ratio numbers let you say how much more of a quality one participant has than another.

Question 52

A measure's validity is always more important than its sensitivity and level of measurement.

Question 53

Your most important consideration in selecting a measure is whether the use of the measure is ethical.