SCALES AND MEASURES
A graduate level introduction and illustrated tutorial to measurement and scaling, including Likert scales, Guttman scales, Mokken scales, Rasch scales and more.
New in the 2013 edition:
New coverage of ordinal coefficient alpha, recommended over Cronbach's alpha for binary and ordinal items.
New introduction to item response theory, with annotated computer output.
New introduction to Rasch analysis, with annotated computer output
New FAQ on standardization vs. rescaling.
Now includes all of "Measurement Levels" as well (which is now discontinued as a separate title).
Table of contents:
Measurement7
Levels of Measurement7
Nominal data7
Ordinal data7
Interval data8
Latent variables9
Constructs9
Latent variables9
Data Levels and measures of central tendency10
Scaling11
Overview11
Constructs vs. categories and benchmarks11
Data-gathering techniques12
Types of techniques12
Likert survey items13
Ranking scales13
Social distance scales13
Paired comparisons13
Q-Sorts14
Fixed resources allocations14
Other methods14
Steps in scale development14
Methods of combining items in a scale17
Overview17
Indexes17
Ordinal scales18
Ordered scales18
Composite scores20
Factors and latent variables20
Multidimensional scales20
Tau Equivalence21
Congeneric v. tau-equivalent v. parallel indicators21
Tau-equivalence in SEM21
Likert items and Likert scales21
Overview21
Data level of Likert scales22
Index of discriminative power (DP)22
Guttman scales23
Overview23
Scale error in Guttman scaling23
Goodness of fit measures for Guttman scales25
Chi-square significance of Guttman scales26
Proximity scales27
Overview27
Example27
Proximity scales vs. Guttman scales28
Mokken scales28
Overview28
Loevinger's H29
Measures of Internal Consistency30
Overview30
Cronbach's alpha30
Ordinal coefficient alpha and ordinal coefficient theta33
Factor analysis36
Rasch scales37
Other internal consistency measures37
Measures of agreement38
Overview38
Cohen's kappa38
Measures of Reliability39
Measures of item discrimination39
Biserial correlation39
Point-biserial correlation40
Item Analysis40
Overview40
Software42
Item analysis output43
Rasch models47
Overview47
Software47
Anchoring48
Rasch analysis output48
Assumptions61
Unidimensionality61
Equality of means and standard deviations61
Frequently asked questions62
Is it acceptable to use ordinal variables such as Likert items in interval-level procedures like regression or path analysis which assume interval data?62
Is it acceptable to use dichotomous variables in procedures like regression or path analysis which assume interval data?64
What is attenuation of effect size/correlation?64
I have categorical data but I want to do interval procedures like factor analysis and regression. What do I do?66
When I have a mixture of categorical and interval independents, what procedure should I use?67
Should I use existing items or create my own?67
How can I find standard scales and instruments?68
When constructing a set of scales used to form a component average, should I standardize or rescale?68
Should I standardize my variables before testing for reliability of scale items with Cronbach's alpha?69
I want to add item scores together to form a scale. How can I test if the items are additive?69
How can I get my SPSS worksheet into Guttman scaling format?70
What software programs compute Guttman scaling coefficients of scalability and reliability?71
What is the 'cluster bloc technique'?71
Selecting the criteria71
Computing indices of agreement71
Establishing non-arbitrary cutoff points72
Forming the cluster block matrix72
What are Thurstone equal-appearing interval scales?72
Q-dispersion73
Collections of scales73
Bibliography82
Pagecount 91
A graduate level introduction and illustrated tutorial to measurement and scaling, including Likert scales, Guttman scales, Mokken scales, Rasch scales and more.
New in the 2013 edition:
New coverage of ordinal coefficient alpha, recommended over Cronbach's alpha for binary and ordinal items.
New introduction to item response theory, with annotated computer output.
New introduction to Rasch analysis, with annotated computer output
New FAQ on standardization vs. rescaling.
Now includes all of "Measurement Levels" as well (which is now discontinued as a separate title).
Table of contents:
Measurement7
Levels of Measurement7
Nominal data7
Ordinal data7
Interval data8
Latent variables9
Constructs9
Latent variables9
Data Levels and measures of central tendency10
Scaling11
Overview11
Constructs vs. categories and benchmarks11
Data-gathering techniques12
Types of techniques12
Likert survey items13
Ranking scales13
Social distance scales13
Paired comparisons13
Q-Sorts14
Fixed resources allocations14
Other methods14
Steps in scale development14
Methods of combining items in a scale17
Overview17
Indexes17
Ordinal scales18
Ordered scales18
Composite scores20
Factors and latent variables20
Multidimensional scales20
Tau Equivalence21
Congeneric v. tau-equivalent v. parallel indicators21
Tau-equivalence in SEM21
Likert items and Likert scales21
Overview21
Data level of Likert scales22
Index of discriminative power (DP)22
Guttman scales23
Overview23
Scale error in Guttman scaling23
Goodness of fit measures for Guttman scales25
Chi-square significance of Guttman scales26
Proximity scales27
Overview27
Example27
Proximity scales vs. Guttman scales28
Mokken scales28
Overview28
Loevinger's H29
Measures of Internal Consistency30
Overview30
Cronbach's alpha30
Ordinal coefficient alpha and ordinal coefficient theta33
Factor analysis36
Rasch scales37
Other internal consistency measures37
Measures of agreement38
Overview38
Cohen's kappa38
Measures of Reliability39
Measures of item discrimination39
Biserial correlation39
Point-biserial correlation40
Item Analysis40
Overview40
Software42
Item analysis output43
Rasch models47
Overview47
Software47
Anchoring48
Rasch analysis output48
Assumptions61
Unidimensionality61
Equality of means and standard deviations61
Frequently asked questions62
Is it acceptable to use ordinal variables such as Likert items in interval-level procedures like regression or path analysis which assume interval data?62
Is it acceptable to use dichotomous variables in procedures like regression or path analysis which assume interval data?64
What is attenuation of effect size/correlation?64
I have categorical data but I want to do interval procedures like factor analysis and regression. What do I do?66
When I have a mixture of categorical and interval independents, what procedure should I use?67
Should I use existing items or create my own?67
How can I find standard scales and instruments?68
When constructing a set of scales used to form a component average, should I standardize or rescale?68
Should I standardize my variables before testing for reliability of scale items with Cronbach's alpha?69
I want to add item scores together to form a scale. How can I test if the items are additive?69
How can I get my SPSS worksheet into Guttman scaling format?70
What software programs compute Guttman scaling coefficients of scalability and reliability?71
What is the 'cluster bloc technique'?71
Selecting the criteria71
Computing indices of agreement71
Establishing non-arbitrary cutoff points72
Forming the cluster block matrix72
What are Thurstone equal-appearing interval scales?72
Q-dispersion73
Collections of scales73
Bibliography82
Pagecount 91