Download presentation
Presentation is loading. Please wait.
1
McGraw-Hill/Irwin McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All rights reserved.
2
Chapter 12 Advanced Measurement Designs for Survey Research
3
1. Understand attitudes and their components. 2. Describe attitude and behavior scales and assess their strengths and weaknesses. 3. Recommend when noncomparative and comparative scale designs are appropriate. 4. Summarize measurement scale design issues. 5. Compare the trilogy and affect global approaches for explaining attitudes. Learning Objectives 12-3
4
Components of attitudes Cognitive Cognitive Affective Affective Behavioral (conative) Behavioral (conative) Components of attitudes Cognitive Cognitive Affective Affective Behavioral (conative) Behavioral (conative) Attitude... a learned predisposition to act in a consistent positive or negative way to a given object, idea, or set of information. Nature of Attitudes and Marketplace Behaviors 12-4
5
Components of Attitudes Cognitive – beliefs, perceptions, knowledge about objects. Cognitive Affective – emotional feelings toward and object. Affective Behaviora l – intended of actual behavioral response. 12-5
6
Beliefs Examples... Your university Examples... Your university... is a prestigious place to get a degree.... has excellent professors.... is a good value for the money.... needs more and better computers. 12-6
7
Emotions Examples... You love your university. You hate your classes. You like your professors. You think living in the dorm is fun and exciting. 12-7
8
... an observable outcome driven by the interaction of a person’s cognitive component (beliefs) and affective component (emotional strength of beliefs) as they relate to a particular object. Behavior Example: Your decision to return to your university for your sophomore year. 12-8
9
A General Construct Development/Scale Measurement Process 12-9
10
10 Question-Response Formats Three Types ScaledScaled Open-endedOpen-ended Closed-endedClosed-ended 12-10
11
Open-ended = no response options for the respondents. Unprobed Unprobed Probed Probed Close-ended = have response options that can be answered quickly and easily. Dichotomous = only two response options. Dichotomous = only two response options. Multiple Category = more than two response options. Multiple Category = more than two response options. Scaled-response = measures the attributes of some construct under study using a scale. Unlabeled = only the endpoints of the scale are identified. Unlabeled = only the endpoints of the scale are identified. Labeled = all scale positions are identified. Labeled = all scale positions are identified. Open-ended = no response options for the respondents. Unprobed Unprobed Probed Probed Close-ended = have response options that can be answered quickly and easily. Dichotomous = only two response options. Dichotomous = only two response options. Multiple Category = more than two response options. Multiple Category = more than two response options. Scaled-response = measures the attributes of some construct under study using a scale. Unlabeled = only the endpoints of the scale are identified. Unlabeled = only the endpoints of the scale are identified. Labeled = all scale positions are identified. Labeled = all scale positions are identified. Question-Response Formats 12-11
12
Types of Attitude Scales LikertLikertSemanticDifferentialSemanticDifferentialBehaviorIntentionBehaviorIntention 12-12
13
Scale Types Defined LikertScalesLikertScalesSemanticDifferentialSemanticDifferential BehavioralIntentionBehavioralIntention... ask respondents to indicate the extent to which they either agree or disagree with a series of mental or behavioral belief statements about a given object.... use bipolar adjectives and adverbs (competitive/noncompetitive, helpful/not helpful, high quality/low quality, dependable/undependable) as the endpoints of multiple scales.... collect information on the likelihood that people will demonstrate some type of predictable behavior regarding the purchase of a product or service. 12-13
14
Rating Scale Formats AttitudeScalesLikertLikert Semantic Differential Behavioral Intention 12-14
15
Likert Scales... ask respondents to indicate the extent to which they agree or disagree with a series of mental or behavioral belief statements about a given object; e.g., person, event, product, store, etc.... ask respondents to indicate the extent to which they agree or disagree with a series of mental or behavioral belief statements about a given object; e.g., person, event, product, store, etc. 12-15
16
Semantic Differential Scale... a scale format that asks respondents to express their attitudes and/or feelings about a given object; e.g., person, store, product, service, etc. 12-16
17
Behavioral Intention Scale... asks respondents to indicate the likelihood they will behave in some way regarding a product, service, event, object, etc.... asks respondents to indicate the likelihood they will behave in some way regarding a product, service, event, object, etc. 12-17
18
Example of a Partial Modified Likert Scale 12-18
19
Semantic Differential Scale for Tiger Woods as a Credibility Spokesperson 12-19
20
Semantic Differential Scale Used by Bank of America That Defines Each Scale Descriptor 12-20
21
Semantic Differential Scale for Midas Auto Systems Experts 12-21
22
Midas Auto Systems’ Performance Profile Compared with Two Competitors 12-22
23
Behavior Intention Scale for Determining Attendance at Performing Arts Events in Vail, Colorado 12-23
24
Scaled-responses – Labeled Scaled-response questions... utilize a scale selected by the researcher to measure the attributes of some construct under study. Labeled scales – use a scale in which all of the scale points are identified with some description Labeled scales – use a scale in which all of the scale points are identified with some description oAdvantages Enables degree of intensity/feelings to be expressed Enables degree of intensity/feelings to be expressed Simple to administer and code Simple to administer and code Respondents can relate to scale Respondents can relate to scale oDisadvantages Scale may be “forced” or overly detailed Scale may be “forced” or overly detailed Labels may be difficult to develop if more than 5- point scales Labels may be difficult to develop if more than 5- point scales Scaled-response questions... utilize a scale selected by the researcher to measure the attributes of some construct under study. Labeled scales – use a scale in which all of the scale points are identified with some description Labeled scales – use a scale in which all of the scale points are identified with some description oAdvantages Enables degree of intensity/feelings to be expressed Enables degree of intensity/feelings to be expressed Simple to administer and code Simple to administer and code Respondents can relate to scale Respondents can relate to scale oDisadvantages Scale may be “forced” or overly detailed Scale may be “forced” or overly detailed Labels may be difficult to develop if more than 5- point scales Labels may be difficult to develop if more than 5- point scales 12-24
25
The nature of the property being measured; e.g., gender = dichotomous; how you feel about chocolate = scale. The nature of the property being measured; e.g., gender = dichotomous; how you feel about chocolate = scale. Previous research studies – use format if objective is to compare results. Previous research studies – use format if objective is to compare results. Data collection mode – influences the type of scale that can be used. Data collection mode – influences the type of scale that can be used. Respondent ability – kids, less educated may have difficulty with scaled responses. Respondent ability – kids, less educated may have difficulty with scaled responses. The desired scale level – higher level scales permit more sophisticated statistical analysis. The desired scale level – higher level scales permit more sophisticated statistical analysis. Considerations in Choosing a Question-Response Format 12-25
26
26 Basic Concepts in Measurement Measurement – determining how much of a property is possessed by an object. Properties: specific features or characteristics of an object that can be used to distinguish it from another object... Properties: specific features or characteristics of an object that can be used to distinguish it from another object... Objective properties are physically verifiable. Objective properties are physically verifiable. Subjective properties are mental constructs. Subjective properties are mental constructs. Properties: specific features or characteristics of an object that can be used to distinguish it from another object... Properties: specific features or characteristics of an object that can be used to distinguish it from another object... Objective properties are physically verifiable. Objective properties are physically verifiable. Subjective properties are mental constructs. Subjective properties are mental constructs. 12-26
27
Basic Concepts in Measurement ScalesScales OriginOrigin DistanceDistance OrderOrder DescriptionDescription 12-27
28
Description: the use of a descriptor, or label, to stand for each “unit” on the scale: “Yes” “No”, “male” “female”,etc. All levels of measurement have description. Description: the use of a descriptor, or label, to stand for each “unit” on the scale: “Yes” “No”, “male” “female”,etc. All levels of measurement have description. Order: the relative sizes of the descriptors are known so we can say one is “greater / less-than” the other. Order: the relative sizes of the descriptors are known so we can say one is “greater / less-than” the other. Distance: differences between the descriptors are known; e.g., there is a $1 difference between $4 and $5. There is a 10 degree difference between 90 and 100 degrees. Distance: differences between the descriptors are known; e.g., there is a $1 difference between $4 and $5. There is a 10 degree difference between 90 and 100 degrees. Origin: there is a true, natural zero; e.g., there is a zero level of dollars, market share, sales, etc. Origin: there is a true, natural zero; e.g., there is a zero level of dollars, market share, sales, etc. Scales: the approach used to measure concepts (constructs). Scales: the approach used to measure concepts (constructs). Basic Concepts in Measurement 12-28
29
Other Types of Scales ComparativeComparativeNoncomparativeNoncomparative 12-29
30
Noncomparative Rating Scales Used in Marketing Research 12-30
31
Comparative Rating Scales Used in Marketing Research 12-31
32
Examples of Comparative Rating Scales Rank Order PairedComparisonPairedComparisonConstantSumConstantSum 12-32
33
Comparative Rating Scales Used in Marketing Research 12-33
34
... enable respondents to compare their responses to each other by indicating their first preference and so forth until all the desired responses are placed in rank order. Rank order scales 12-34
35
... create a pre-selected group of traits, product characteristics, or features that are paired against one another into two groups; respondents are asked to select which pair is more important to them. PairedComparisonscales 12-35
36
Constant Sum Scales Format requires a person to evaluate each separate attribute/feature relative to all the others listed. Respondents are asked to allocate a given number of points, usually 100, among several attributes or features based on their importance. Most appropriate in self-administered surveys. Require a lot of mental energy on the part of the respondent. 12-36
37
Single-Item or Double-Item Scales? Used when the data requirements focus on collecting data about only one attribute of the object or construct being investigated Measurement is based on responses to several items simultaneously rather than measuring just one item. Single-Item Scales Double-Item Scales 12-37
38
In developing constructs... Must resolve reliability and validity issues of scales. Must assess the dimensionality of the construct under investigation. When choosing single-item versus multiple-item scales... 12-38
39
Recap of Measurement Design Issues oConstruct Development Issues Constructs should be clearly defined. Constructs should be clearly defined. Avoid double-barreled questions/constructs. Avoid double-barreled questions/constructs. oScale Measurement Issues Instructions for both respondents and interviewers should be carefully developed. Instructions for both respondents and interviewers should be carefully developed. Use clear wording and avoid ambiguity. Use clear wording and avoid ambiguity. Avoid “leading” phrases or words. Avoid “leading” phrases or words. Make sure items are phrased unidimensionally (relating to a single concept). Make sure items are phrased unidimensionally (relating to a single concept). Make sure the descriptors are relevant to the type of data being sought. Make sure the descriptors are relevant to the type of data being sought. oConstruct Development Issues Constructs should be clearly defined. Constructs should be clearly defined. Avoid double-barreled questions/constructs. Avoid double-barreled questions/constructs. oScale Measurement Issues Instructions for both respondents and interviewers should be carefully developed. Instructions for both respondents and interviewers should be carefully developed. Use clear wording and avoid ambiguity. Use clear wording and avoid ambiguity. Avoid “leading” phrases or words. Avoid “leading” phrases or words. Make sure items are phrased unidimensionally (relating to a single concept). Make sure items are phrased unidimensionally (relating to a single concept). Make sure the descriptors are relevant to the type of data being sought. Make sure the descriptors are relevant to the type of data being sought. 12-39
40
Follow process guidelines for development of scales, including pre-testing. Follow process guidelines for development of scales, including pre-testing. Develop and use the most appropriate scales possible. Develop and use the most appropriate scales possible. Avoid bias Avoid bias Follow process guidelines for development of scales, including pre-testing. Follow process guidelines for development of scales, including pre-testing. Develop and use the most appropriate scales possible. Develop and use the most appropriate scales possible. Avoid bias Avoid bias Ethical Responsibilities of Researcher Ethical Responsibilities of Researcher 12-40
41
Used before questioning to identify qualified respondents. Used to instruct respondents on which questions to answer – instructions must be very clear. Screening Questions Skip Questions 12-41
42
Other Considerations – Rating Scales Noncomparative Rating Scales Noncomparative... requires a judgment without reference to another object, person, or concept.... requires a judgment without reference to another object, person, or concept.... requires a judgment comparing one object, person, or concept against another. Comparative Rating Scales Comparative 12-42
43
Other Scale Measurement Issues... oSingle-item or multiple-item scales oClear wording oScreening questions oSkip questions oSingle-item or multiple-item scales oClear wording oScreening questions oSkip questions 12-43
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.