Multidimensional Scaling and | Verified Test Bank Chapter 21

Test Bank

CHAPTER 21 Multidimensional Scaling and Conjoint Analysis

True-False

1. Multidimensional scaling addresses the general problem of

positioning objects in a perceptual space.

2. Consider the following MDS solution, in which A, B, C and

represent brands of a product class.

Therefore, it can be concluded that B is sweeter than D but less creamy than A.

3. Multidimensional scaling involves two steps. First, objects need to

be positioned. Second, the dimensions upon which customers perceive

or evaluate objects must be determined.

4. MDS Incorporated has sent out questionnaires to respondents to rank

five brands of toothpaste on their ability to clean and whiten, to prevent decay, and to freshen the mouth. MDS Incorporated is using an attribute-based approach.

5. Two of the approaches used to reduce the number of attributes on which brands of

product are ranked are discriminant analysis and multiple regression analysis.

6. When there is a need to reduce the number of attributes in a study,

factor analysis has an advantage over discriminant analysis in that

a significance test can be done of the output of factor analysis.

7. A few respondents who are unfamiliar with the objects to be mapped should be included

in generating a perceptual map to ensure adequate representation of the population.

8. A source of error results with an attribute-based MDS if the list of

attributes is not accurate and complete.

9. The product "widget" is marketed by ten companies under at least 14 brand names.

It is generally agreed that consumers perceive and evaluate the widgets as a whole. Under the circumstances, the results of an attribute-based MDS solution will be more valid than a non-attribute-based MDS solution.

10. If there are 10 brands of a certain product on which the researcher

wants similarity judgments from a respondent, as many as 45 paired

judgments can be required. This maximum is determined by the formula n(n - 1) / 2.

11. The greater the number of objects to be mapped, the smaller the

chance of a possibly unique mapping solution.

12. In a scaling solution, the most frequent trade-off is between

minimizing the number of dimensions in the solution and maximizing the degree of fit.

13. Non-attribute-based similarity MDS solutions have the advantage of

producing dimensions that are easy to interpret.

14. Two objects could be very different in a similarity-based perceptual

map, but could be regarded as very similar in a preference-based perceptual map.

15. If the input to MDS is binary data, the determination of whether an attribute is

associated with a brand or an object is done through correspondence analysis

16. In MDS, when the number of dimensions increases, the stress value increases

17. An ideal object is one the customer would prefer over all other

objects included in the space. An ideal object need not actually exist.

18. Before introducing a new product, Innovators Incorporated conducted

a preference study. The data were analyzed by using an ideal-point

MDS solution. The company then designed and introduced a product

which was extremely "close" to the target market's ideal point.

However, since a new product has been introduced into the space,

another study should be done since the ideal point may have been relocated.

19. In the economic theory of consumer behavior, a basic underlying

assumption is that of non-satiation (the consumer would always like

to have more). Under this assumption, the ideal object would be

represented by an ideal vector or direction rather than an ideal point in the space.

20. Conjoint analysis provides a qualitative measurement of the relative

importance of one attribute as compared with another.

21. When preferences for various attributes are in conflict, conjoint

analysis cannot be used.

22. When each attribute is considered independently, respondents tend to

indicate that only a few attributes are important.

23. In conjoint analysis, respondents are given product concepts on

cards and asked to describe the attributes they attach to each concept.

24. A trade-off is made by giving up some amount of a less important

attribute to get more of a more important attribute.

25. In theory, before respondents can knowledgeably make trade-offs,

they must compare different attributes and evaluate the desirability

of the various levels of each attribute.

26. The greater the difference between the highest and lowest valued

levels of an attribute, the less important the attribute.

27. When utilities are summed for each of the concepts being judged, the

rank order of these sums should match the respondent's rank ordering

of preference as closely as possible.

28. If all of the possible levels of an attribute have the same utility,

the attribute is not important in influencing overall attitude.

29. One of the applications of conjoint analysis is to the creation of

new products with significant consumer appeal relative to competitive alternatives.

30. In the full profile approach, respondents may be asked to rank order

cards with complete product or service configurations.

31. A limitation in the use of conjoint analysis is the necessity of

dividing each attribute level into discrete levels.

32. The pair-wise trade-off approach produces concepts that are more

realistic than those of the full profile approach.

33. Respondents who are overwhelmed by a ranking task will tend to

ignore variations in the less important attributes.

34. Utility measurements from conjoint analysis can be used to develop

marketing simulations.

35. The independent variable in conjoint analysis is the preference

judgment that a respondent makes about a new concept.

36. A basic assumption of conjoint analysis is that people evaluate a concept by adding

up their evaluations of the individual attribute levels of that concept.

37. Conjoint analysis assumes interaction between attributes.

38. Conjoint analysis, although intuitively appealing, has been slow to

gain acceptance in the marketing research community.

39. Conjoint analysis has been found to have reliability problems in

that different implications tend to arise if the full profile method

is used instead of the trade-off approach.

40. The output of MDS is the location of the objects on the dimensions, termed as

perceptual map

41. Discriminant analysis does not provide a test of statistical significance

42. Discriminant analysis identifies clusters of attributes on which objects differ,

similar to factor analysis

Multiple Choice

1. The objective of multidimensional scaling is
2. predict buying or usage of product that may still be in concept stage
3. to address the general problem of positioning objects in perceptual space
4. group individuals or objects into groups
5. none of the above
6. Number of dimensions in an MDS cannot be decided using
7. ease of use
8. external validity of solution
9. interpretability of data
10. none of the above

3. Approaches to MDS involve

1. the study of attribute data.

2. the study of similarity, non-attribute data.

3. the study of preference, non-attribute data.

a. 1

b. 2

c. 3

d. 1, 2, and 3

4. MDS computer algorithms for mapping similarity data

1. convert ranks by degree of similarity into distances.

2. try to use the least number of dimensions to accurately

represent the input rankings.

3. try to construct a map so that similar objects are at a maximum distance.

4. always produce a unique map for the input similarity rankings.

a. 1

b. 2

c. 3

d. 1 and 2

5. Consider the following solution from attribute-based data:

Which of the following is not true?

a. Brand A is perceived to be more similar to Brand B than to Brand D.

b. The price dimension is a better discriminant than the alcoholic strength dimension.

c. Brand C is more expensive than Brand E.

d. Brand D is less strong than Brand E.

6.

Table 1

Table 2

Note: Both tables have been derived from the same data

The similarity scale was of the type

1 2 3 4 5 6 7 8 9 10

Which of the following statements is true?

1. Table 1 retains approximately as much information as

Table 2 when input to an algorithm.

2. Table 2 contains nonmetric information.

3. The average customer perceives brands D and B as the most similar brands.

4. The average customer perceives Brands D and E as the most similar.

a. 1

b. 2

c. 3

d. 1 and 3

7.

I₁ and I₂ represent the ideal points of two market segments. A, B, C, and D represent available brands.

1. B is preferred to D by segment 2.

2. B is preferred to D by segment 1.

3. C is preferred to B by segment 1.

4. C is preferred to B by segment 2.

a. 1

b. 2

c. 3

d. 2 and 4

8. Multidimensional scaling attempts to

a. map consumers' perceptions and preferences for objects or their attributes.

b. provide spatial relationships of objective data.

c. map and transform distances.

d. evaluate the positive and negative aspects of existing brands of a product.

9. When using similarity-based MDS techniques, one of the major

problems is interpreting the dimensions. The methods for doing this include

1. trial and error.

2. interpretation based on the researcher's insight and experience.

3. correlating the physical characteristics of an object with the object's position on the dimension.

4. increasing the number of dimensions.

a. 1

b. 2

c. 3

d. 2 and 3

10. Which of the following is an appropriate use of conjoint analysis?

a. Where the alternative products have a number of attributes with multiple levels

b. Where most of the feasible combinations of attribute levels do not presently exist

c. Where the range of possible attribute levels can be expanded beyond those presently available

d. All of these are appropriate uses of conjoint analysis.

11. Which of the following situations would present difficulty for conjoint analysis?

a. A manager wants to present potential customers with a range of

attribute levels beyond those available in products on the market.

b. Most of the feasible attribute levels for the product being studied do not exist.

c. The product being studied is a low-priced, low-risk product,

which consumers tend to pick up and buy with little thought.

d. None of the above.

12. In order to be able to make trade-offs, the respondents must theoretically be able to

1. compare different attributes.

2. evaluate the desirability of the various levels of each attribute.

3. evaluate all the different products on the market on all attributes.

a. 1

b. 2

c. 1, 2, and 3

d. 1 and 2

13. Which of the following statements is not true?

a. Conjoint analysis is termed an analysis of interdependence technique.

b. Conjoint analysis requires a respondent to evaluate a concept in terms of overall liking, intention to buy, or rank order of preference compared to other concepts.

c. Using conjoint analysis, a researcher can identify optimal levels

of attributes for new products.

d. Conjoint analysis decomposes the consumer's overall judgment into utilities that represent the worth of each level of each attribute relative to the other levels.

14. Working with MDS includes all the following problems except:

a. when more than two or three dimensions are needed, the usefulness is reduced.

b. perceptual mapping has not been proven to be reliable across different methods.

c. the interpretation of dimensions is difficult since several attributes are involved.

d. maps are not usually based on groups that are aggregated with respect to their familiarity with products, usage levels and attitudes.

15. Which of the following statements concerning the full profile

approach is not true?

a. It yields a somewhat higher predictive validity than does the trade-off approach.

b. The concept evaluation task can employed either a ranking or a rating scale.

c. Fewer judgments have to be made by the respondent than if

two-attribute, trade-off approach were used.

d. All of these are true.

16. Given the following utility graph developed from a conjoint

analysis, which statement is true?

Price $ $ $ 30 25 20 Miles per

3 4 6 gallon

0 5 0

0 0 0

0 0 0

a. The utility of a car costing $3,000 and doing 30 mpg is .54.

b. Price is a less important attribute than mpg.

c. The utility of a car costing $3,000 and doing 30 mpg is less than 1.

d. None of the above are true.

17. An approach in which the respondents are given cards that describe complete

product or service configurations is called

1. Trade off approach
2. Full profile approach
3. Conjoint analysis
4. Direct approach