Deck 13: Measuring Your Study Variables

Question 1

Which of these situations would be least vulnerable to the social desirability bias?

A) You ask therapists whether they believe their treatment of clients for anxiety is usually effective.
B) You conduct a client satisfaction survey where clients respond to questions anonymously.
C) You ask a client in a therapy session if he or she believes the therapy is helping.
D) You give a group of clients the Beck Depression Inventory before and after therapy and they respond anonymously.

Answer 1

You give a group of clients the Beck Depression Inventory before and after therapy and they respond anonymously.

Question 2

Objective data are more likely than biased data to be ______.

A) rejected
B) disapproved
C) heard
D) viewed as cherry-picking

Answer 2

heard

Question 3

What is the critical issue in measurement?

A) accuracy
B) expressiveness
C) vulnerability
D) dependency

Answer 3

accuracy

Question 4

You need to know the level of measurement of your study variables in order to ______.

A) determine if your measurement tool is critical
B) select the appropriate statistic to test your hypothesis
C) find the appropriate study sample for your study
D) discover the most useful tool to measure your variables

Question 5

For which type of client are you more likely to find an individualized scale more useful than a standardized scale?

A) a middle-aged woman who is seeking treatment for depression
B) a young man being treated for PTSD
C) a young child with limited verbal ability
D) a senior citizen being treated for anxiety

Question 6

Tests of reliability and validity for measurement tools are designed to reduce ______ error.

A) measurement
B) sampling
C) generalization
D) statistical

Question 7

Which of the following statements about measurement error is true?

A) You can expect no measurement error if you use a good tool for measurement.
B) You can expect no measurement error if you have a random sample.
C) You can expect a good deal of measurement error if you employ a quasi-experimental research design.
D) You can expect some level of measurement error whenever you measure human variables.

Question 8

Which of the following is a general type of error in measurement?

A) nonrandom
B) random
C) elusive
D) nondeterminant

Question 9

______ is/are mechanisms for preventing measurement error.

A) Tests of reliability and validity
B) A good knowledge base and random sample
C) A proper statistic for testing the study hypothesis
D) A set of conclusions that are consistent with the data analyzed in the study

Question 10

In measurement, reliability refers to ______

A) variability
B) enhancement
C) consistency
D) turnover

Question 11

The tool you use to measure the client's target behavior can be ______.

A) valid even though it is not reliable
B) reliable even though it is not valid
C) valid and reliable without being credible
D) credible without being reliable or valid

Question 12

A test of ______ validity is the weakest means of reducing measurement error.

A) face
B) criterion
C) content
D) construct

Question 13

Which of the following is true?

A) There are mechanisms for testing for reliability and validity for quantitative research but not for qualitative research.
B) There are mechanisms for testing for reliability and validity both for quantitative research and qualitative research.
C) There are only methods of testing for reliability in quantitative research.
D) There are only methods of testing for validity in quantitative research.

Question 14

A key advantage of the published standardized scale over the individualized scale is that the published standardized scale has ______

A) been designed for qualitative measurement
B) been reviewed by the top experts in the field
C) neither passed nor failed various tests of reliability and validity because it is unique in what it measures
D) typically been designed by an expert and has passed one or more tests of reliability or validity

Question 15

According to the text, your most important guide when you are seeking a measurement tool is the ______

A) process objective for the treatment that is being delivered
B) analysis of the behavior to be measured
C) definition of the behavior to be measured
D) criticism of the behavior to be measured

Question 16

Suppose your outcome objective is to improve knowledge of how to find employment. Consider what your tool would look like. Which of the following would be the best set of categories to give to clients on the tool?

A) (1) true or (2) false
B) (1) none of the time, (2) some of the time, or (3) most of the time
C) (1) disagree, (2) undecided, or (3) agree
D) (1) very unlikely, (2) likely, or (3) very likely

Question 17

One of the questions to consider when seeking a published standardized scale for adults is whether the tool is ______

A) incongruent with various ways to measure your variable
B) consistent with the way you have defined the variable being measured
C) revolutionary in its treatment of the concept being measured
D) simple enough for children to read

Question 18

When you have a measurement tool that contains options like strongly agree, agree, undecided, disagree, and strongly disagree, you will typically find that the statements are different in the way they measure the study variable. This means that when you tally the score for this scale, you must ______ on the scale.

A) direct score all the items
B) engage in optional scoring of items
C) consistently score each item
D) reverse-score some of the items

Question 19

Which of the following would be most useful in determining the practical significance in an evaluative study?

A) a p value less than .50
B) information on the reliability and validity of the measurement tool employed
C) statistical significance and thresholds of functioning
D) the likelihood that the data can be explained by chance

Question 20

According to the text, what is one of the special challenges in measurement in evaluative research when you are evaluating the progress of a single client using a single-subject design?

A) You do not have any information in single-subject research about whether the intervention is the cause of the measured client growth rather than something else.
B) You will not know whether the measurement tools are reliable or valid.
C) The time it takes for the client to complete the measurement tool may be too intrusive into the service process.
D) You cannot draw conclusions from your study if you are measuring only one client.

Question 21

One of the limitations of the Session Rating Scale and the Outcome Rating Scale is that they do not allow the client a great range of self-expression in describing the outcomes they seek.

Question 22

The Beck Depression Inventory is an example of an individualized rating scale.

Question 23

Measurement error is the distance between the results from a measurement tool and the truth about the concept being measured.

Question 24

Standardized scales are tailored to the specific needs of the client.

Question 25

Finding a published scale for measuring your study variables is usually preferable to developing your own scale.

Question 26

If you computed the correlations between scores on one self-esteem scale and the scores on another scale, where higher scores represent higher self-esteem for both scales, you would expect to find a negative correlation coefficient.

Question 27

When you develop a scale that measures study variables, the options for response should be both mutually exclusive and exhaustive.

Question 28

The alpha coefficient (Cronbach's ?) is a coefficient that indicates the internal consistency of a scale by examining the correlations of different items with one another.

Question 29

Test-retest reliability is a test of reliability that examines whether a given tool is consistent when administered to the same group of people at two points in time.

Question 30

Because practical significance is a matter of opinion, the researcher does not need to explain the basis for asserting that it has been achieved.

Question 31

Suppose that you have two clients in treatment, one is an adult female being treated for depression and another is a 6-year-old child being treated because of poor motivation in school. For each of these situations, indicate whether you would be inclined to seek a standardized scale for measuring client progress or develop your own individualized scale. Explain why.

Question 32

What is meant by reliability and validity? Can a scale be reliable if it is not valid? Can it be valid if it is not reliable? Explain.

Question 33

In evaluative research, what is meant by statistical significance and practical significance? Can you have statistical significance if you do not have practical significance? Can you have practical significance if you do not have statistical significance?

Question 34

Suppose that you need to develop a scale that measures political opinions. Would you have items on this scale with response options like (1) true or (2) false, or would it have options like (1) strongly disagree, (2) disagree, (3) undecided, (4) agree, and (5) strongly agree? Explain why you chose this form for the measurement of opinions.

Question 35

What do face validity and content validity mean? How would each be examined?