Deck 6: Indexes, Scales, and Typologies

Question 1

Indexes must be made up of:

A)nominal variables.
B)ordinal variables.
C)interval variables.
D)ratio variables.
E)any type of variables.

Answer 1

E

Question 2

An index is constructed by assigning scores to patterns of responses.

Answer 2

True

Question 3

For an index or scale to be considered unidimensional, its component items should be indicators of only one dimension of a concept.

Answer 3

False

Question 4

Jeremy includes items that measure sexism in his index of gender roles. He does this because he believes the two variables are empirically related. As a methodology student, you should tell Jeremy that:

A)it's a great idea, and he should do it.
B)if he includes both the variables in his index he is violating concerns about the general or specific dimensions of an index.
C)if he includes both the variables in his index he needs to examine the bivariate correlations between them to be sure they are empirically correlated.
D)if he includes both the variables in his index he is violating the criteria specifying the need for unidimensionality in index construction.
E)it's a great idea as long as the bivariate correlations among the variables exceed 0.9.

Question 5

Given the following patterns for a Guttman scale on life satisfaction, the coefficient of reproducibility (is): ​ ​

A)1.00
B).95
C).90
D).85
E)cannot be computed from the information given

Question 6

The following items measure the extent to which Americans are willing to associate with extraterrestrials. It's OK if my child marries an extraterrestrial. It's OK to have an extraterrestrial for a friend. It's OK to have an extraterrestrial living on the block. It's OK to have an extraterrestrial living in my state. These items illustrate a: ​

A)Guttman scale
B)Bogardus social distance scale
C)Thurstone scale
D)Likert scale
E)Semantic differential scale

Question 7

Why might you choose a scale over an index?

A)You are using nominal variables.
B)You are using ordinal variables.
C)You find an intensity structure in your data.
D)You are using more than one variable.
E)You find you have missing data points.

Question 8

Professor Thames tried to use a Guttman scale on his data set that he had seen published in a major journal. Thames was upset because the items did not form a Guttman scale on his data. Confused, he asked for your help. You should tell him that:

A)the article that appeared in the journal was probably an error.
B)he should recheck his scaling techniques because he probably made an error.
C)items may form a Guttman scale in one sample but not in another.
D)it was probably not a very good scale.
E)a Guttman scale can only be used in a univariate analysis.

Question 9

Which of the following statements is FALSE while determining the relative strengths of the items of bivariate relationships?

A)Be wary of including items in a composite measure that are not related to one another empirically.
B)One should probably drop any item that is not related to several other items in the composite measure.
C)Variable very strong relationship between two items presents a problem.
D)One must check the bivariate correlations between all items that one plans to include in the composite measure.
E)You should strive for perfectly correlated (1.0)indicators within a composite measure.

Question 10

A Thurstone scale was constructed to measure creativity. The scale scores ranged from 1 to 13. If the scale was properly constructed we could conclude that:

A)a person with a score of 6 was three times as creative as a person with a score of 2.
B)a person with a score of 6 was more creative than a person with a score of 5.
C)a person with a score of 5 was half as creative as someone with a score of 10.
D)you cannot make any conclusions with a Thurstone scale.
E)a person with a score of 6 was half as creative as someone with a score of 10.

Question 11

A classification system derived from the observation of specific variables is known as a(n):

A)external validation.
B)Thurstone scale.
C)Likert scale.
D)typology.
E)index.

Question 12

Professor Swanson developed an index of marital happiness. Several items in Swanson's questionnaire were not a part of the index (i.e., "How many serious quarrels or arguments have you had with your spouse in the past year?"). Swanson compared the index scores of marital happiness with these questions. Swanson was doing:

A)index construction.
B)item analysis.
C)external validation.
D)analysis invalidation.
E)Likert scaling.

Question 13

If two items are perfectly correlated, both should be included in an index.

Question 14

Maria is constructing an index from a range of four responses to a question about job satisfaction. She asks your advice about how to weight the responses toward a score on job satisfaction. Which of the following would be the best advice?

A)Maria should weight the answers with the most responses the most heavily.
B)Maria should weight the extreme cases higher than those in the middle.
C)Maria should weight the cases in the middle higher than those on the extremes.
D)Maria should weight each answer equally unless there's a compelling reason not to.
E)There is no general rule of thumb to give Maria.

Question 15

A(n)_____ summarizes the intersection of two or more variables.

A)Guttman scale
B)Likert scale
C)semantic differential scale
D)index
E)typology

Question 16

Which of the following is true about indexes and scales?

A)They are interval measures.
B)Their attributes form an intensity structure.
C)They rank-order the units of analysis in terms of specific variables.
D)They are ordinal measures of variables.
E)They are nominal measures of variables.

Question 17

Using Likert-type scale items, Professor Wong developed an index to assess job morale. Wong administered the index to 2,000 workers and found that 1 to 4 responses were missing on each of the items. Which of the following should Wong NOT do to correct this situation?

A)Exclude the cases that contained the missing data from the analysis.
B)Assign the middle value to cases with missing data.
C)Analyze the cases that had missing data to interpret their meaning.
D)Assign values at random.
E)Readminister the study.

Question 18

Robert notices that several respondents did not answer the question of whether they took out loans to fund their education. However, later in his survey, Robert finds that respondents told him about the source(s)and the amount(s)of their college loans. In handling the missing data on the question of whether or not they had a college loan, Robert's best approach is to:

A)randomly assign values to the missing data.
B)delete the cases with missing data from the analysis.
C)assign the middle value to the cases with the missing data.
D)use the information about amount and source of loans to assign values.
E)ask the respondents to complete the survey a second time to get accurate responses.

Question 19

Which of the following do indexes and scales NOT have in common?

A)They are ordinal measures of variables.
B)They are based on patterns of responses.
C)They rank-order the units of analysis.
D)They are composite measures.
E)They are based on measurement of more than one variable.

Question 20

Tammy is creating a composite measure to examine life satisfaction and she is trying to decide whether to include the categories of very very dissatisfied and very very satisfied or to simply leave the extreme categories as very dissatisfied and very satisfied. She turns to you for help. Which of the following statements would not help her because it is FALSE?

A)This is a variance issue.
B)Typically, as the possible extremes of an index are extended, fewer cases are to be found in each extreme.
C)The range of measurement in an index is an advantage of an index and therefore she should include the most extreme categories to take advantage of the full range.
D)She needs to think about both the range of her index and whether there will be an adequate number of cases at each point in the index.
E)She needs to consider whether it adds important information to her study to know about the most extreme categories.

Question 21

Thurstone scales use judges in their construction.

Question 22

Describe at least three different scales that are used in social research. How are they similar and different?

Question 23

Likert scales are designed to be _____ in measurement.

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

Question 24

A friend of yours is trying to develop an index to reproduce respondents' patterns of responses. You tell your friend:

A)to construct a Guttman scale.
B)to construct a Likert scale.
C)to construct a Thurstone scale.
D)to construct any one of the above.
E)it cannot be done.

Question 25

Both scales and indexes are interval measures.

Question 26

Testing the validity of a composite measure by using another, proven indicator is known as:

A)external validation.
B)internal validation.
C)index scoring.
D)Thurstone scaling.
E)face validation.

Question 27

Scales are sample dependent, indexes are not.

Question 28

The coefficient of reproducibility refers to the ability of a Guttman scale to reproduce the pattern of individual responses.

Question 29

Typologies are typically nominal composite measures.

Question 30

Explain the steps in index construction. How does a researcher validate the index once it's constructed?

Question 31

Jenny finds that she has data missing on 10-35 cases (out of 2,000 cases)for each variable that she intends to use in her composite measure. What are her options in dealing with the missing data?

Question 32

Typologies can only be constructed on the basis of a single variable.

Question 33

Item analysis allows you to examine the extent to which the composite index is related to other items in the questionnaire that are not part of the index.

Question 34

Which of the following is NOT an important consideration to take into account when building an index?

A)Face validity
B)Balance between general and specific
C)Perfectly correlated components
D)Unidimensionality
E)Variance

Question 35

According to Babbie, as a general guideline, the coefficient of reproducibility for a Guttman scale should be:

A)100 percent
B)90 or 95 percent
C)80 or 85 percent
D)70 percent
E)It doesn't matter

Question 36

Which of the following statements is NOT applicable to typologies?

A)They are typically nominal composite measures.
B)They involve a set of categories or types.
C)They may be used effectively as independent or dependent variables.
D)They are often used when researchers wish to summarize the intersection of two or more variables.
E)They are abstractions that are used to examine reality.

Question 37

What differentiates typologies from scales and indexes? Provide at least two examples of appropriate uses of typologies.

Question 38

Compare and contrast how scales and indexes are used in social research. How do researchers decide which to use?