Deck 10: Attitude Measurement

Question 1

Which of the following statements about attitudes is FALSE?

A) Attitudes represent a person's ideas or convictions, with regard to a specific object or idea.
B) Attitude plays a key role in many marketing models that describe consumer behavior.
C) Although attitude is a key concept, it is generally regarded as unrelated to actual purchase behavior.
D) In general, practitioners have a strong interest in consumer attitudes and attitude research.
E) Attitudes are relatively stable.

Answer 1

Although attitude is a key concept, it is generally regarded as unrelated to actual purchase behavior.

Question 2

A company might be interested in assessing the attitudes of

A) its employees.
B) its customers.
C) its suppliers.
D) all of the above.
E) a and b only.

Answer 2

all of the above.

Question 3

Which of the following is FALSE?

A) In measurement theory, it is not possible to directly measure objects themselves.
B) It is easy to determine which properties of the scale of numbers apply in a given measurement situation.
C) When numbers are assigned to represent attributes of objects, the relationship between the numbers may not be the same as the relationship between the attributes, which affects what one can say about the numbers.
D) We defined measurement as the "assignment of numerals to represent quantities of attributes." For nominal scaling, the rule is "do not assign the same numeral to different classes or different numerals to the same class."
E) Ratio scales allow more powerful modes of analysis than ordinal scales.

Answer 3

It is easy to determine which properties of the scale of numbers apply in a given measurement situation.

Question 4

Which of the following objects can be directly measured?

A) a person
B) a car
C) an attitude
D) all of the above
E) none of the above

Question 5

In measurement we measure

A) objects.
B) quantities of objects.
C) quantities of attributes of objects.
D) only constructs with visual referents.
E) only constructs with non-visual referents.

Question 6

Which of the following statements is NOT true of an ordinal scale?

A) It has the property of identity.
B) It has the property of order.
C) The mean is an appropriate measure of average.
D) All of the above are true of an ordinal scale.
E) a and c only are not true of an ordinal scale.

Question 7

Any monotonic positive transformation of the numerals can be performed on which type of scale?

A) nominal
B) ordinal
C) interval
D) a and b
E) positive monotonic transformations can be applied to any scales

Question 8

Which of the following is FALSE?

A) A positive linear transformation of the form y = a + bx can only be performed on a ratio scale if one is to preserve the properties of the scale.
B) A ratio scale differs from an interval scale in that the ratio scale has a natural zero.
C) Determination of absolute zero is difficult with measurements in marketing. It is debatable whether many of the constructs used in marketing even possess this property.
D) a and b.
E) a, b, and c.

Question 9

Which of the following is FALSE?

A) The observation approach to attitude determination rests on the presumption that a subject's behavior is conditioned by his or her attitudes.
B) The most common approach to measuring attitudes has been self-report.
C) One of the disadvantages of the equal-appearing interval method of scaling is that a large number of judges are required during the process of developing the scale.
D) One of the advantages of the equal-appearing interval method of scaling is that scale construction and scale use take place at the same time.
E) The directions to subjects for the Likert scale are the same when constructing it as when using it.

Question 10

Which of the following is NOT true when a scale is a ratio scale?

A) Transformations of the form y = bx are permissible
B) The scale is invariant for all transformations of the form y = a - bx
C) If "a" is greater than "b" and "b" is greater than "c," then "a" is greater than "c"
D) If "a" equals "b" and "b" equals "c," then "a" equals "c"
E) The mode and the median are permissible measures of central tendency

Question 11

Which scale(s) is (are) undistorted under a linear transformation of the form y=a+bx?

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

Question 12

Which of the following is used as a measure of dispersion with the method of equal-appearing intervals?

A) the variance
B) the interquartile range
C) the median
D) the Q-value
E) b and d

Question 13

Which one of the following statements regarding attitude measurement is FALSE?

A) The physiological reaction approach makes use of some unstructured or partially structured stimuli such as word association tests or sentence completion tests.
B) The observation of behavior approach rests on the presumption that a subject's behavior is conditioned by his attitudes and thus we can use his observed behavior to infer his attitudes.
C) The performance of "objective" tasks approach rests on the presumption that a subject's performance of a specific assigned task will depend upon his attitudes.
D) The self-report techniques rely on the subject's own responses to a set of statements.
E) Self-report measures of attitude have their own sources of error.

Question 14

When developing a scale using the method of equal-appearing intervals, the analyst should

A) generate a series of statements and classify the statements a priori as favorable or unfavorable.
B) select a small number of important attributes of the object.
C) generate a list of statements that reflect the attributes of the object and which span the favorable-unfavorable continuum.
D) generate a large number of statements that represent an intermediate degree of favorableness to the object.
E) select a small number of statements to be presented to the subjects on separate cards.

Question 15

After the judges have completed sorting the statements with the method of equal-appearing intervals, the researcher uses the scale values and measures of dispersion to

A) select a subset of judges to develop the final instrument.
B) select a number of judges to act as a panel.
C) determine which judges did not take the tasks seriously.
D) select a number of statements which need to be improved.
E) select a subset of statements to serve as a final instrument.

Question 16

If, when choosing the method of equal-appearing intervals, one is forced to choose between two statements of equal scale values, he would choose the statement that

A) had the smallest Q value.
B) had the largest Q value.
C) was favorable to the object and yet had the smallest Q.
D) was unfavorable to the object and yet had the largest Q.
E) The Q value would not be used in this case.

Question 17

Which of the following statements about Thurstone equal-appearing interval scales is FALSE?

A) The final scale typically contains 20-22 statements that span the favorability continuum.
B) The subject indicates his degree of agreement or disagreement with each statement.
C) Judges are asked to sort the initial pool of statements into categories based on the degree of favorability of each statement.
D) Statements with low Q values are preferred over statements with high Q values.
E) The task for subjects when developing the scale is different than their task in using the scale.

Question 18

Which of the following about the Thurstone scale is TRUE?

A) The judges used to evaluate the statements for a Thurstone scale judge the statements on whether or not they agree or disagree with them.
B) If a respondent agrees with statements in a Thurstone equal-appearing interval scale that have an average score of 4, the respondent is said to have an unfavorable attitude toward the object.
C) When using a Thurstone scale to measure a subject's attitude toward an object, the subject indicates his degree of agreement or disagreement with each statement and the respondent's total score is computed by summing his scores from all of the statements.
D) The Q value in the Thurstone scale refers to the scale value of a particular statement.
E) They are all false.

Question 19

Which of the following is FALSE?

A) With the method of equal-appearing intervals, the final scale should have statements that cover the range of favorable to unfavorable.
B) The final form of the Thurstone (equal-appearing interval) scale typically contains approximately 20-22 statements that are relatively equal in scale value with interquartile ranges as small as practical.
C) The Thurstone scale does not allow respondents to express the intensity of their feelings.
D) a and b.
E) a, b, and c.

Question 20

The method of summated ratings overcomes which criticism of the method of equal-appearing intervals?

A) Difficulties in developing a large number of statements.
B) The costs due to the size of the judgment sample required with equal-appearing intervals.
C) The difficulty of generating total scores by which to array subjects.
D) The non-scientific nature of item analysis with the method of equal-appearing intervals.
E) The lack of ability for a subject to express his intensity of feeling with an equal-appearing interval scale.

Question 21

Which of the following is TRUE?

A) The basic format of the scale for the summated ratings method is the same in both construction and use.
B) For both Thurstone and Likert scales, respondents are asked to indicate their degree of agreement or disagreement with each statement.
C) Statements are classified as being favorable or unfavorable a priori with equal-appearing intervals and summated rating scales.
D) a and b.
E) a, b, and c.

Question 22

The method of summated ratings

A) does not require a screening or judgment sample as does the method of equal-appearing intervals.
B) not only requires a screening sample but further requires the screening sample to be representative of the larger group of subjects of interest.
C) requires the analyst to classify each statement a priori as favorable or unfavorable.
D) a and c.
E) b and c.

Question 23

Which of the following statements about Likert summated rating scales is FALSE?

A) The subject indicates his/her degree of agreement or disagreement with each statement.
B) Statements are classified a priori as being favorable or unfavorable.
C) The sample used to screen statements need not be representative of the population on whom the scale is to be used.
D) Members of the screening sample respond to each statement with reference to a particular attitude object.
E) The essential task assigned subjects is the same when using the scale as it is when developing the scale.

Question 24

Item analysis

A) tells the researcher that a statement with a high item-to-total product moment correlation should be eliminated because it is likely to be redundant.
B) is useful in the initial process of developing statements for a Likert scale.
C) is based on the premise that there should be consistency in the responses of a given individual.
D) a and b.
E) a, b, and c.

Question 25

Item analysis is

A) based on the proposition that there is consistency in the response patterns of the subjects when the items are ambiguous.
B) a procedure for increasing the reliability and validity of a test by increasing the number of items.
C) a procedure of determining the degree of favorableness and unfavorableness of items.
D) a procedure to determine whether the total test discriminates in the same way the individual items discriminate.
E) a procedure whereby each test item is evaluated to determine whether that item effectively discriminates among subjects and therefore should be included in the final version of the instrument.

Question 26

Which of the following is FALSE?

A) Marketers have often failed to engage in scale purification procedures when designing semantic differential scales.
B) Some semantic differential scale items should be reversed so that respondents can more easily complete the scales.
C) A snake diagram can readily communicate the perceived position of one or more objects based on semantic differential scale items.
D) As with a Likert scale, the semantic differential scale allows a respondent to express the intensity of his or her feeling toward an object.
E) Scale purification procedures for semantic differential scales are like those for summated rating scales.

Question 27

There are 50 items in a summated rating scale and the items are scored 1=strongly disagree and 5=strongly agree for positive statements while the reverse scoring is used for negative statements. Subject A has a total score of 150 toward the attitude object "x." This means that

A) subject A has a positive attitude toward "x" because his score exceeds the "middle" score of 125.
B) subject A has a negative attitude toward "x" because his score exceeds the "middle" score of 125.
C) subject A is indifferent to "x" because his score is in the midrange of possible scores.
D) subject A has a more favorable attitude toward "x" than the average subject.
E) none of the above.

Question 28

The three basic uncorrelated dimensions which account for a large proportion of variances in the ratings of objects by the semantic differential method are

A) objective, subjective, and indifferent.
B) length, breadth, and width.
C) semantic, pragmatic, and syntactic.
D) cognition, perception, and meaning.
E) evaluation, potency, and activity.

Question 29

When using the semantic differential technique, marketers have tended to modify the basic approach. These modifications have NOT included

A) generalization of new items rather than using items from the basic list of adjective pairs.
B) the use of adjective phrases rather than simple antonyms.
C) the generation of total scores employing a procedure similar to that employed in the method of summated ratings.
D) the generation of evaluation, potency, and activity scores by which objects could be compared.
E) all of the above modifications have been made by marketers when employing the semantic differential technique.

Question 30

The product-moment correlations of ten Likert items with the total scores are as follows
Item $\quad$ Product -moment correlation
A $\quad$ $\quad$ $\quad$ .92
B $\quad$ $\quad$ $\quad$ .01
C $\quad$ $\quad$ $\quad$ -.01
D $\quad$ $\quad$ $\quad$ -.05
E $\quad$ $\quad$ $\quad$ .70
F $\quad$ $\quad$ $\quad$ -.92
G $\quad$ $\quad$ $\quad$ -.98
H $\quad$ $\quad$ $\quad$ .70
I $\quad$ $\quad$ $\quad$ .85
J $\quad$ $\quad$ $\quad$ .02
Which of the above statements are the best for inclusion in a scale?

A) A, E, H, I
B) C, D, F, G
C) B, J
D) A, B, E, H, I, J
E) A, E, F, G, H, I

Question 31

Which of the following pairs is not appropriate for anchoring a semantic differential scale item in a marketing application?

A) Good-bad
B) Unbelievable-believable
C) able to attract attention-unable to attract attention
D) Fluffy-hard
E) all of the above could be used

Question 32

The semantic differential adjective pair "helpful-unhelpful" is an example of the____ dimension.

A) potency
B) activity
C) evaluation
D) structural
E) none of the above

Question 33

Which of the following is FALSE?

A) A Likert scale is an example of a comparative rating scale.
B) A semantic differential scale is an example of an itemized scale.
C) When using graphic rating scales, individuals indicate their rating by placing a check at the appropriate point on a line that runs from one extreme of the attribute to the other.
D) Itemized scales are similar to graphic scales in that individuals make their judgments independently.
E) Comparative scales involve relative judgments.

Question 34

Which of the following is FALSE?

A) While the level of measurement is not overly important from a statistical point of view, the level of measurement is crucial for interpretation of results.
B) A halo effect occurs when there is carryover from one judgment to another in the scaling procedure.
C) Item analysis is a procedure whereby each test item is evaluated to determine whether that item effectively discriminates among subjects and therefore should be included in the final version of the instrument.
D) By establishing the reliability of a scale, a researcher also establishes its validity.
E) The validity of a scale is always an inference.