Malhotra 09 tif - research techniques PDF

Preview

Summary

research techniques ...

Description

Chapter 9 Measurement & Scaling: Non-comparative Scaling Techniques True/False Questions 1. Comparative techniques are comprised of continuous & itemized rating scales. (False, moderate, page 255) 2. Non-comparative scales are often referred to as monadic scale. (True, moderate, page 256) 3. Respondents using a non-comparative scale employ whatever rating standard seems appropriate to them. (True, moderate, page 256) 4. Burger King used Perception Analyzer to measure responses to a series of “slice-of-life” commercials. (False, moderate, page 257) 5. In an itemized rating scale, respondents are provided with a scale that has a number or brief description associated with each category. (True, moderate, page 257) 6. Itemized rating scales are widely used in marketing research & form basic components of more complex scales. (True, moderate, page 257) 7. Typically, each Likert scale item has seven response categories, ranging from “strongly disagree” to “strongly agree.” (False, moderate, page 258) 8. A total (summated) score can be calculated for each respondent by summing across items. (True, easy, page 258) 9. Profile analysis involves determining average respondent ratings for each item. (True, moderate, page 258) 10. Semantic differential scale is also referred to as a summated scale. (False, moderate, page 258) 11. Semantic differential is a five-point rating scale with endpoints associated with bipolar labels that have semantic meaning. (False, moderate, page 259) 12. Stapel scale is usually presented horizontally. (False, easy, page 261) 13. An advantage of Stapel scale is it can be administered over telephone. (True, difficult, page 261) 14. Of three itemized rating scales considered, semantic differential scale is used least. (False, moderate, page 261) 15. Researcher must make four major decisions when constructing non-comparative itemized rating scales. 115

(False, moderate, page 261) 16. Smaller number of scale categories, finer discrimination among stimulus objects that is possible. (False, moderate, page 262) 17. When determining number of scale categories to use in a non-comparative itemized rating scale, nature of object is relevant. (True, moderate, page 262) 18. When determining number of scale categories to use in a non-comparative itemized rating scale, if individual responses are of interest, or data will be analyzed by sophisticated statistical techniques, five or more scale categories may be required. (False, difficult, page 262) 19. Likert scale is a balanced rating scale with an odd number of categories & a neutral point. (True, easy, page 262) 20. A forced rating scale forces respondents to express an opinion because “no opinion” or “no knowledge” options are not provided. (True, easy, page 263) 21. In situations where respondents are expected to have no opinions, as opposed to simply being reluctant to disclose it, accuracy of data may be improved by a non-forced scale that includes a “no opinion” category. (True, easy, page 263) 22. It has been found that providing a verbal description for each scale category consistently improves accuracy or reliability of data. (False, moderate, page 263) 23. Non-comparative itemized rating scales with weak adjectives as anchors (1=generally disagree, 7=generally agree) result in less variable & more peaked response contributions. (False, difficult, page 263) 24. A construct is theory being measured. (False, moderate, page 263) 25. Scale development process is an iterative one. (True, easy, page 265) 26. XO = XT + XS + XR represents pure score model. (False, difficult, page 266) 27. Reliability refers to extent to which a scale produces valid results if repeated measurements are made. (False, moderate, page 267) 28. Systematic sources of error do have an adverse impact on reliability because they affect measurement in a constant way & do not lead to inconsistency. (True, moderate, page 267) 29. Reliability can be defined as extent to which measures are free from random error, XR. (True, easy, page 267) 116

30. When assessing test-retest reliability, higher correlation coefficient, between two measurements, greater reliability. (True, moderate, page 267) 31. In alternative-forms reliability, same respondents are measured at two different times, usually one to three weeks apart, with a different scale form being administered each time. (False, difficult, page 268) 32. With alternative forms reliability, a low correlation may reflect either an unreliable scale or nonequivalent forms. (True, moderate, page 268) 33. An important property of coefficient alpha is that its value tends to increase with an increase in number of scale items. (True, moderate, page 268) 34. Coefficient alpha assists in determining whether averaging process used in calculating coefficient beta is masking any inconsistent items. (False, moderate, page 268) 35. Perfect validity requires that there be no measurement error, therefore: (X O = XT, XR = 0, XS = 0). (True, difficult, page 269) 36. Given its subjective nature, content validity alone is a sufficient measure of validity of a scale. (False, moderate, page 269) 37. Construct validity is simplest & easiest type of validity to establish. (False, moderate, page 269) 38. Using several scale items to measure characteristic of interest provides more accurate measurement than a single-item scale. (True, easy, page 271) 39. Semantic differential scale may be said to be pan-cultural or free of cultural bias. (True, moderate, page 271) 40. Researcher can bias scales by either biasing wording of statements (Likert type scales), scale descriptors, or other aspects of scale. (True, easy, page 272)

Multiple Choice Questions 41. In a _____, respondents rate objects by placing a mark at appropriate position on a line that runs from one extreme of criterion variable to other. a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (c, easy, page 256)

117

42. How would you rate Sears as a department store? Version 1 Probably worst - - - - - - -I - - - - - - - - - - - - - - - - - - -

- - - Probably best

Version 2 Probably worst - - - - - - -I - - - - - - - - - - - - - - - - - - - - - - Probably best 0 10 20 30 40 50 60 70 80 90 100 Above scales are all examples of a _____. a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (a, moderate, page 256) 43. Scores assigned to continuous rating scales by researcher are typically treated as _____ data. a. Nominal b. Ordinal c. Ratio d. Interval (d, moderate, page 256) 44. Which of following statements does not pertain to non-comparative scales? a. Comparative scales are often referred to as monadic scales. b. Respondents using a non-comparative scale employ whatever rating standard seems appropriate. c. Data must be interpreted in relative terms & have only ordinal or rank order properties. d. Non-comparative techniques consist of continuous & itemized rating scales. (c, moderate, page 256) 45. Which non-comparative scale has advantage of being easy to construct & disadvantage of cumbersome scoring unless scoring is computerized (Table 9.1)? a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (c, difficult, page 257) 46. Perception Analyzer utilizes a _____. a. Continuous rating scale b. Staple scale. c. Semantic differential scale d. Likert scale (a, moderate, page 257) 47. Which scale is not an itemized rating scale? a. Stapel scale b. Semantic differential scale c. Likert scale d. Continuous rating scale (d, easy, page 257)

118

48. A _____ is a measurement scale with five response categories ranging from “strongly disagree” to “strongly agree,” which requires respondents to indicate a degree of agreement or disagreement with each of a series of statements related to stimulus objects. a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (b, easy, page 258) 49.

Neither Agree Strongly nor Strongly Disagree Disagree disagree Agree agree 1. Sears has poor in-store service. 1 2X 3 4 5 2. I like to shop at Sears. 1 2X 3 4 5 Above scale is an example of a _____. a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (d, moderate, page 258)

50. Which non-comparative scale is analyzed using profile analysis? a. Likert scale b. Semantic differential scale c. Stapel scale d. All of above (d, difficult, pages 258, 260, 261) 51. Which itemized rating scale takes longer to complete than other itemized rating scales because respondents have to read each statement? a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (b, difficult, page 259) 52. A _____ is a seven point rating scale with endpoints associated with bi-polar labels that have semantic meaning. a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (a, easy, page 259) 53. Sears is: Powerful —:—:—:—:-X-:—:—: Weak Unreliable —:—:—:—:—:-X-:—: Reliable Above scale is an example of a _____ scale. a. Continuous rating b. Stapel 119

c. Semantic differential d. Likert (c, easy, page 260) 54. _____ is known for its versatility & is very popular with marketing researchers. a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (c, difficult, page 260) 55. Which non-comparative scale is widely used in comparing brand, product, & company images? a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (a, moderate, page 260) 56. A _____ is a scale for measuring attitudes that consists of a single adjective in middle of an evennumbered range of values, from -5 to +5, without a neutral point (zero). a. Semantic differential scale b. Likert scale c. Continuous rating scale d. Stapel scale (d, moderate, page 261) 57. Which scale asks respondent to indicate how accurately or inaccurately each term describes object by selecting an appropriate numerical response category? a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (b, difficult, page 261) 58. Data obtained by using a Stapel scale can be analyzed in same way as a_____. a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (c, difficult, page 261) 59. _____ is confusing & difficult to apply. It is least used of itemized scales. a. Continuous rating scale b. Stapel scale c. Semantic differential scale d. Likert scale (b, difficult, page 261) 60. Which of following statements is not a consideration when making non-comparative itemized rating scale decisions? a. Number of scale categories to use b. Forced versus non-forced choice c. Balanced versus unbalanced scales 120

d. All are considerations (d, moderate, page 261) 61. Which statement is not true if deciding number of scale categories to use in a non-compensatory itemized rating scale? a. Traditional guidelines suggest that appropriate number of categories should be seven plus or minus two: between five & nine. b. Smaller number of scale categories, finer discrimination among stimulus objects that is possible. c. If respondents are not very knowledgeable or involved with task, fewer categories should be used. d. How data are to be analyzed & used should also influence number of categories. (b, difficult, page 262) 62. Which statement is not true when deciding on whether to use balanced or unbalanced scales when developing a non-comparative itemized rating scale? a. Scale should be balanced to obtain objective data. b. In a balanced scale, number of favorable & unfavorable categories are equal. c. If distribution of responses is likely to be skewed, either positively or negatively, a balanced scale with more categories in direction of skewness may be appropriate. d. If an unbalanced scale is used, nature & degree of unbalance in scale should be taken into account in data analysis. (c, difficult, page 262) 63. Which statement is not true when deciding on whether to use an odd or even number of categories when developing a non-comparative itemized rating scale? a. With an odd number of categories, middle scale position is generally designated neutral or impartial. b. Decision to use an odd or even number of categories depends on whether some of respondents may be neutral on response being measured. c. A rating scale with an even number of categories should be used if researcher wants to force a response. d. All of above statements are true. (d, easy, pages 262-263) 64. Deciding whether to present scales as vertical or horizontal is related to which of non-comparative itemized rating scale decisions? a. Number of scale categories b. Physical form or configuration c. Odd or even number of categories d. Nature & degree of verbal description (b, easy, page 263) 65. _____ is first step in developing a multi-item scale. _____ is last step. a. Generate an initial pool of item; Prepare final scale b. Develop a theory; Prepare a final scale c. Develop a theory; Develop a purified scale d. Generate an initial pool of items; Develop a purified scale (b, moderate, page 265) 66. Validity can be assessed by examining all of following except: a. Item validity b. Content validity c. Criterion validity d. Construct validity 121

(a, moderate, page 266) 67. Which of following is not an approach to assess multi-item scale reliability? a. Test-retest reliability b. Construct reliability c. Alternative forms reliability d. Internal consistency reliability (b, moderate, page 266) 68. _____ is variation in information sought by researcher & information generated by measurement process employed. a. Systematic error b. Measurement error c. Random error d. Variable error (b, difficult, page 266) 69. XO = XT + XS + XR In true score model shown above, XT represents: a. Random error b. Observed score or measurement c. True score of characteristic d. Systematic error (c, moderate, page 266) 70. Situational factors, such as presence of other people, noise, & distractions & mechanical factors, such as poor printing, overcrowding of items in questionnaire, & poor design are both _____ in measurement. a. Random error b. Potential sources of reliability c. Potential sources of error d. Systematic error (c, difficult, page 267) 71. _____ represents stable factors that affect observed score in same way each time measurement is made, such as mechanical factors (see Fig. 9.6). a. Systematic error b. Measurement error c. Random error d. Variable error (a, moderate, page 267) 72. _____ is not constant. It represents transient factors that affect observed score in different ways each time measurement is made, such as transient personal or situational factors. a. Systematic error b. Measurement error c. Random error d. Variable error (c, moderate, page 267) 73. _____ is extent to which a scale produces consistent results if repeated measurements are made on characteristic. 122

a. Validity b. Generalizability c. Reliability d. None of above (c, difficult, page 267) 74. A measure is perfectly reliable if: a. XO = 0 b. XT = 0 c. XS = 0 d. XR = 0 (d, moderate, page 267) 75. _____ is an approach for assessing reliability in which respondents are administered identical sets of scale items at two different times under as nearly equivalent conditions as possible. a. Internal consistency reliability b. Split-half reliability c. Test-retest reliability d. Alternative-forms reliability (c, moderate, page 267) 76. There are several problems associated with test-retest approach to determining reliability. If measuring respondents’ attitude toward low-fat milk may cause them to become more health conscious & develop a more positive attitude toward low-fat milk, then there is a problem with: a. Time interval between testing. b. Initial measurement altering characteristic being measured. c. It being impossible to make repeated measurements. d. First measurement having a carryover effect to second or subsequent measurements. (b, difficult, page 267) 77. _____ is an approach for assessing reliability that requires two equivalent forms of scale to be constructed & then same respondents are measured at two different times. a. Internal consistency reliability b. Split-half reliability c. Test-retest reliability d. Alternative-forms reliability (b, moderate, page 268) 78. Which of following is not a problem with alternative-forms reliability? a. Results will depend on how scale items are split. b. It is time consuming & expensive to construct an equivalent form of scale. c. It is difficult to construct two equivalent forms of a scale. d. Both b & c are correct. (a, difficult, page 268) 79. _____ is an approach for assessing internal consistency of set of items when several items are summated in order to form a total score for scale. a. Internal consistency reliability b. Split-half reliability c. Test-retest reliability d. Alternative-forms reliability (a, easy, page 268) 123

80. _____ is a form of internal consistency reliability in which items constituting scale are divided into two halves & resulting half scores are correlated. a. Internal consistency reliability b. Split-half reliability c. Test-retest reliability d. Alternative-forms reliability (b, easy, page 268) 81. _____ is a measure of internal consistency reliability that is average of all possible split-half coefficients resulting from different splittings of scale items. a. Coefficient delta b. Coefficient alpha c. Coefficient beta d. Coefficient eta (b, moderate, page 268) 82. _____ is extent to which differences in observed scale scores reflect true differences among objects on characteristics being measured, rather than systematic or random errors. a. Validity b. Generalizability c. Reliability d. None of above (a, difficult, page 269) 83. _____ is a type of validity, sometimes called face validity, that consists of a subjective but systematic evaluation of representativeness of content of a scale for measuring task at hand. a. Construct validity b. Content validity c. Criterion validity d. Internal consistency validity (b, difficult, page 269) 84. A scale designed to measure store image would be considered inadequate if it omitted any of major dimensions (quality, variety, assortment of merchandise, etc.). This inadequacy would be reflected in _____ of scale. a. Construct validity b. Content validity c. Criterion validity d. Internal consistency validity (b, difficult, page 269) 85. _____ is a type of validity that examines whether measurement scale performs as expected in relation to other variables selected as meaningful criteria. a. Construct validity b. Content validity c. Criterion validity d. Internal consistency validity (c, difficult, page 269)

124

86. _____ is assessed when data on scale being evaluated on criterion variables are collected at same time. a. Convergent validity b. Predictive validity c. Concurrent validity d. Discriminant validity (c, moderate, page 269) 87. _____ is a type of validity that addresses question of what construct or characteristic scale is measuring. An attempt is made to answer theoretical questions of why a scale works & what deductions can be made concerning theory underlying scale. a. Construct validity b. Content validity c. Criterion validity d. Internal consistency validity (a, easy, page 269) 88. _____ is a measure of construct validity that measures extent to which scale correlates positively with other measures of same construct. a. Convergent validity b. Discriminant validity c. Nomological validity d. Concurrent validity (a, difficult, page 269) 89. _____ is a type of construct validity that assesses extent to which a measure does not correlate with other constructs from which it is supposed to differ. a. Convergent validity b. Discriminant validity c. Nomological validity d. Concurrent validity (b, difficult, page 269) 90. _____ is extent to which scale correlates in theoretically predicted ways with measures of different but related constructs. a. Convergent validity b. Discriminant validity c. Nomological validity d. Concurrent validity (c, difficult, page 269) 91. Which statement is not true regarding relationship between reliability & validity? a. If a measur...