Deck 5: The Process of Measurement

Question 1

Which of the following levels of measurement will permit us to make statements about proportions and ratios

A)ordinal
B)interval
C)ratio
D)both interval and ratio

Answer 1

C

Question 2

All of the following are measurement techniques used in the social sciences EXCEPT:

A)verbal reports.
B)sampling validity.
C)observation.
D)archival records.

Answer 2

B

Question 3

A research study measures height by having observers judge people's heights and classify them into one of three categories: tall height, medium height, and short height. The resulting measure would most clearly be at which level of measurement

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

Answer 3

B

Question 4

Operational definitions are at which of the following levels in the research process

A)the theoretical level
B)the revised nominal definition level
C)the research level
D)the most abstract level

Question 5

The major reason for using a scale for measurement rather than a single item is that:

A)scales are usually quicker to administer.
B)scales enable us to measure variables in a more precise and accurate fashion.
C)scales enable us to measure many variables at the same time rather than just a single variable.
D)scales are more concrete and less abstract than are single item measures.

Question 6

Measurement is a part of the scientific process that involves moving from _______________ to _______________.

A)the concrete...the theoretical
B)the theoretical...the concrete
C)concepts...hypotheses
D)theories...hypotheses

Question 7

In the formula presented to show the logic of measurement in the social sciences, X = T + E, the "E" stands for:

A)error.
B)experiment.
C)evaluation.
D)effect.

Question 8

A single indicator of a variable is called:

A)an item.
B)a scale.
C)an index.
D)a measurement.

Question 9

Some researchers argue that attitude scales-with response alternatives such as "strongly agree," "agree," "disagree," and "strongly disagree"-can be considered to be at the interval level of measurement because:

A)such scales involve categories with a fixed order.
B)such scales have sufficient items to make for valid and reliable measurement.
C)such scales constitute continuous variables.
D)people perceive the distance between choice alternatives on such scales to be equal.

Question 10

Which of the following levels of measurement possess mutually exclusive categories, a fixed order to the categories, and equal spacing between the units of measurement

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

Question 11

Which of the following is the formula used to show the logic of measurement in the social sciences

A)X = T - E
B)T = X + E
C)X = T + E
D)E = X + T

Question 12

The concept of socioeconomic status possesses all of the following characteristics related to its level of measurement EXCEPT:

A)mutually exclusive categories.
B)equal spacing between categories.
C)a fixed order to the categories.
D)all of the other answers are characteristics of socioeconomic status.

Question 13

Which of the following variables would be considered nominal variables at the theoretical level

A)sex
B)ethnicity
C)socioeconomic status
D)sex and ethnicity

Question 14

A research project asks people to state the number of times that they have been arrested. This variable would be considered at which level of measurement

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

Question 15

An "indicator" in both research and practice is best defined as:

A)an observation that is assumed to be evidence of the attributes or properties of something.
B)the process of describing abstract concepts by assigning numbers or other symbols to them in accordance with rules.
C)a reconceptualization of a theoretical definition on the basis of difficulties encountered in operationalizing it.
D)a variable that has been operationalized by combining a number of items into a composite score.

Question 16

Nominal measures classify observations in which of the following ways

A)into mutually exclusive and exhaustive categories
B)into ordered categories
C)into categories with equal spacing between the intervals
D)into categories having a meaningful zero point

Question 17

The text refers to "measurement" as the process of:

A)validating indicators.
B)creating error.
C)operationalizing concepts.
D)creating indexes.

Question 18

The level of measurement that will allow researchers to perform common arithmetic operations on the numbers is:

A)nominal.
B)ordinal.
C)interval.
D)all of the answers are levels of measurement that will allow researchers to perform common arithmetic operations on the numbers.

Question 19

Which of the following statements would be consistent with the nonpositivist critique of measurement

A)The social sciences should strive to quantify all measurement.
B)Social science measurement often involves the construction of something that we believe exists in the world.
C)All the phenomena that social scientists measure exist objectively in the world and are discovered by our measuring devices.
D)While interval and ratio variables can be objectively measured, nominal and ordinal measures cannot.

Question 20

The term "level of measurement" refers to:

A)whether research will be qualitative or quantitative in nature.
B)whether research will be correlational or experimental in nature.
C)the rules that define permissible mathematical operations on a measuring device.
D)whether a variable can be considered continuous or discrete.

Question 21

Of the following ways of demonstrating validity, which is the weakest demonstration of validity

A)content validity.
B)criterion validity.
C)concurrent validity.
D)construct validity.

Question 22

The reliability test that is subject to the multiple-testing effect is:

A)the split half approach.
B)the Spearman Brown approach.
C)the test retest approach.
D)Cronbach's alpha approach.

Question 23

When developing a measurement tool for use with members of a minority group, the text recommends that researchers do which of the following

A)Get assistance from members of the culture in developing the measuring tools.
B)Immerse themselves in the culture of the group being studied.
C)Perform validity and reliability checks using members of the minority group.
D)All of the answers represent things that are recommended for researchers to do.

Question 24

One of the advantages of the test retest method of assessing reliability is:

A)it can be used with many measures.
B)it is not subject to the multiple testing effect.
C)it is quicker than the other tests of reliability.
D)it can be used even though the values of variables may have changed between the test administrations.

Question 25

In order to improve the validity and reliability of measurement devices, the text recommends all of the following techniques EXCEPT:

A)more extensive conceptual development.
B)use a lower level of measurement.
C)better train those who will apply the measurement devices.
D)all of the answers are techniques that are recommended.

Question 26

The two principles that serve as a basis for the techniques for estimating the reliability of a measure are:

A)induction and deduction.
B)stability and equivalence.
C)stability and known groups.
D)equivalence and jury opinion.

Question 27

In selecting an indicator to measure a variable, a researcher should:

A)strive for the highest level of measurement possible.
B)be guided predominantly by the theoretical nature of the variable.
C)avoid the nominal level of measurement.
D)both strive for the highest level of measurement possible, and be guided predominantly by the theoretical nature of the variable.

Question 28

Which of the following methods of testing reliability gives the clearest indication of reliability

A)internal consistency approaches
B)the test retest method
C)the multiple testing effects method
D)the multiple forms approach

Question 29

Of the following, the most complex form of validity is:

A)criterion validity.
B)concurrent validity.
C)construct validity.
D)predictive validity.

Question 30

Which of the following would be the best demonstration of the validity of a scale for the measurement of suicide potential

A)Ask mental health experts to assess whether it measures suicide potential accurately.
B)Correlate the scale results with the actual occurrence of suicides.
C)Ask people who have attempted suicide whether it measures suicide potential accurately.
D)Administer the scale to those who have attempted suicide to see if they score appropriately on it.

Question 31

Error that is consistent and patterned is called:

A)random error.
B)systematic error.
C)unreliable data.
D)replicable error.

Question 32

One inherent weakness of concurrent validity is that:

A)it depends on the subjective opinions of judges.
B)it has low levels of reliability.
C)it assumes that the existing measures used for comparison purposes are valid.
D)it depends on the accuracy of prior tests for face validity.

Question 33

Which of the following is most clearly NOT a discrete variable

A)social class position
B)religious affiliation
C)race
D)sex

Question 34

According to Research in Practice 5.2, which of the following statements is true regarding measurement in human service practice

A)Practitioners need not be concerned about issues of validity.
B)Practitioners need to assess the validity and reliability of the tools they use to measure things.
C)Practitioners need to be concerned about validity but not reliability.
D)Measurement tools used in practice have become valid by the time practitioners use them.

Question 35

When testing the reliability of a measuring device, a solution to the problem of multiple-testing effects is to:

A)use the test retest method.
B)divide the test group into two groups, one that is tested twice and one that is tested only once.
C)combine the results from the two administrations and then randomly divide them into two separate test groups.
D)use a known groups comparison as a way of assessing the extent of the multiple testing effect.

Question 36

The crime statistics of the Federal Bureau of Investigation consistently report less violent crime than actually occurs. This is an example of:

A)test-retest reliability.
B)concurrent validity.
C)random error.
D)systematic error.

Question 37

Researchers should strive for the highest level of measurement possible because:

A)higher levels of measurement are generally more theoretically sound.
B)higher levels of measurement are generally more precise.
C)higher levels of measurement are generally more valid.
D)higher levels of measurement are generally more reliable.

Question 38

The "known groups" approach to validity is a variation on:

A)concurrent validity.
B)predictive validity.
C)construct validity.
D)discriminant validity.

Question 39

When choosing which of several indicators to use in a research project, which of the following would play a role in the decision

A)consideration of which indicator is theoretically preferred
B)consideration of which indicator would permit the most sophisticated mathematical operations to be performed on it
C)consideration of the level of measurement of the indicator
D)all of the answers would play a role in the decision

Question 40

variables are variables that have a finite number of distinct and separate values or categories.

A)Nominal
B)Discrete
C)Continuous
D)Valid

Question 41

Describe two methods of assessing the validity of measurement devices and elaborate on the advantages and disadvantages of each.

Question 42

Describe two methods of assessing the reliability of measurement devices and elaborate on the advantages and disadvantages of each.

Question 43

What techniques are recommended in the chapter to improve the validity and reliability of a measuring device that shows unacceptably high levels of error

Question 44

Compare and contrast the positivist and nonpositivist approaches to measurement. Be sure to include a discussion of the nonpositivist's critique of positivist measurement theory.

Question 45

Describe in detail the procedures involved in applying the split half approach to reliability.

Question 46

Discuss ways in which computer technology can enhance measurement in human service research.

Question 47

Differentiate between random and systematic error in measurement.Which is more problematic for researchers and why Relate each type of error to the formula used in classical measurement theory.

Question 48

Identify the four levels of measurement and give the distinguishing characteristics of each.

Question 49

In the area of measurement, researchers confront two major issues: validity and reliability. Differentiate between validity and reliability.

Question 50

A researcher is studying the effectiveness of an intensive treatment program for young drug users. One dependent variable used in the study is the self reported, estimated dollar value of goods stolen during the year following treatment. For the ten participants and ten controls, the data are as follows:
Indicate whether these data are discrete or continuous, and specify the highest level of measurement appropriate for these data. In analyzing these data, which level do you think would be best Defend your choice.