Donna Harrington
- Published in print:
- 2008
- Published Online:
- January 2009
- ISBN:
- 9780195339888
- eISBN:
- 9780199863662
- Item type:
- book
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195339888.001.0001
- Subject:
- Social Work, Research and Evaluation
Measures that are reliable, valid, and can be used across diverse populations are vital to social work research, but the development of new measures is an expensive and time-consuming process. An ...
More
Measures that are reliable, valid, and can be used across diverse populations are vital to social work research, but the development of new measures is an expensive and time-consuming process. An array of existing measures can provide a cost-effective alternative, but in order to take this expedient step with confidence, researchers must ensure that the existing measure is appropriate for the new study. Confirmatory factor analysis (CFA) is one of the ways to do so. CFA has four primary functions—psychometric evaluation of measures, construct validation, testing method effects, and testing measurement invariance. This book provides an overview of the method, step-by-step guides to creating a CFA model and assessing its fit, and explanations of the requirements for using CFA, as well the book underscores the issues that are necessary to consider when using multiple groups or equivalent and multilevel models. Real-world examples, screenshots from the Amos software program that can be used to conduct CFA, and reading suggestions for each chapter form part of the book.Less
Measures that are reliable, valid, and can be used across diverse populations are vital to social work research, but the development of new measures is an expensive and time-consuming process. An array of existing measures can provide a cost-effective alternative, but in order to take this expedient step with confidence, researchers must ensure that the existing measure is appropriate for the new study. Confirmatory factor analysis (CFA) is one of the ways to do so. CFA has four primary functions—psychometric evaluation of measures, construct validation, testing method effects, and testing measurement invariance. This book provides an overview of the method, step-by-step guides to creating a CFA model and assessing its fit, and explanations of the requirements for using CFA, as well the book underscores the issues that are necessary to consider when using multiple groups or equivalent and multilevel models. Real-world examples, screenshots from the Amos software program that can be used to conduct CFA, and reading suggestions for each chapter form part of the book.
Donna Harrington
- Published in print:
- 2008
- Published Online:
- January 2009
- ISBN:
- 9780195339888
- eISBN:
- 9780199863662
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195339888.003.0004
- Subject:
- Social Work, Research and Evaluation
This chapter examines how to determine whether a confirmatory factor analysis model fits well, including a discussion of the various fit indices available, which ones to use, and thresholds for ...
More
This chapter examines how to determine whether a confirmatory factor analysis model fits well, including a discussion of the various fit indices available, which ones to use, and thresholds for determining acceptable fit. Assessment of model fit involves considering a number of indices of model fit, including absolute fit, parsimony correction, comparative fit, and predictive fit indices. Recommendations for identifying acceptable model fit are presented, and methods of finding sources of poor fit were discussed. Model revision, including the use and testing of nested models, modification indices, localized areas of strain, and specification search, is discussed. The chapter also addresses how to revise a model that does not fit well, including incorporating theory-based changes and the use of modification indices. Finally, a detailed confirmatory factor analysis (CFA) example is presented that includes a discussion of all the aspects of specifying, testing, assessing, and revising the model.Less
This chapter examines how to determine whether a confirmatory factor analysis model fits well, including a discussion of the various fit indices available, which ones to use, and thresholds for determining acceptable fit. Assessment of model fit involves considering a number of indices of model fit, including absolute fit, parsimony correction, comparative fit, and predictive fit indices. Recommendations for identifying acceptable model fit are presented, and methods of finding sources of poor fit were discussed. Model revision, including the use and testing of nested models, modification indices, localized areas of strain, and specification search, is discussed. The chapter also addresses how to revise a model that does not fit well, including incorporating theory-based changes and the use of modification indices. Finally, a detailed confirmatory factor analysis (CFA) example is presented that includes a discussion of all the aspects of specifying, testing, assessing, and revising the model.
Will Bridewell, Stuart R. Borrett, and Pat Langley
- Published in print:
- 2009
- Published Online:
- September 2009
- ISBN:
- 9780195381634
- eISBN:
- 9780199870264
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195381634.003.0011
- Subject:
- Psychology, Cognitive Psychology
Scientific modeling is a creative activity that can benefit from computational support. This chapter reports five challenges that arise in developing such aids, as illustrated by PROMETHEUS, a ...
More
Scientific modeling is a creative activity that can benefit from computational support. This chapter reports five challenges that arise in developing such aids, as illustrated by PROMETHEUS, a software environment that supports the construction and revision of explanatory models. These challenges include the paucity of relevant data, the need to incorporate prior knowledge, the importance of comprehensibility, an emphasis on explanation, and the practicality of user interaction. The responses to these challenges include the use of quantitative processes to encode models and background knowledge, as well as the combination of AND/OR search through a space of model structures with gradient descent to estimate parameters. This chapter reports our experiences with PROMETHEUS on three scientific modeling tasks and some lessons we have learned from those efforts. This chapter concludes by noting additional challenges that were not apparent at the outset of our work.Less
Scientific modeling is a creative activity that can benefit from computational support. This chapter reports five challenges that arise in developing such aids, as illustrated by PROMETHEUS, a software environment that supports the construction and revision of explanatory models. These challenges include the paucity of relevant data, the need to incorporate prior knowledge, the importance of comprehensibility, an emphasis on explanation, and the practicality of user interaction. The responses to these challenges include the use of quantitative processes to encode models and background knowledge, as well as the combination of AND/OR search through a space of model structures with gradient descent to estimate parameters. This chapter reports our experiences with PROMETHEUS on three scientific modeling tasks and some lessons we have learned from those efforts. This chapter concludes by noting additional challenges that were not apparent at the outset of our work.
