Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- book
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.001.0001
- Subject:
- Social Work, Research and Evaluation
Structural Equation Modeling (SEM) has long been used in social work research, but the writing on the topic is typically fragmented and highly technical. The chapters demonstrate two SEM programs ...
More
Structural Equation Modeling (SEM) has long been used in social work research, but the writing on the topic is typically fragmented and highly technical. The chapters demonstrate two SEM programs with distinct user interfaces and capabilities (Amos and Mplus) with enough specificity that readers can conduct their own analyses without consulting additional resources. Examples from social work literature highlight best practices for the specification, estimation, interpretation, and modification of structural equation models. Unlike most sources on SEM, this book provides clear guidelines on how to evaluate SEM output and how to proceed when model fit is not acceptable. Oftentimes, confirmatory factor analysis and general structure modeling are the most flexible, powerful, and appropriate choices for social work data.Less
Structural Equation Modeling (SEM) has long been used in social work research, but the writing on the topic is typically fragmented and highly technical. The chapters demonstrate two SEM programs with distinct user interfaces and capabilities (Amos and Mplus) with enough specificity that readers can conduct their own analyses without consulting additional resources. Examples from social work literature highlight best practices for the specification, estimation, interpretation, and modification of structural equation models. Unlike most sources on SEM, this book provides clear guidelines on how to evaluate SEM output and how to proceed when model fit is not acceptable. Oftentimes, confirmatory factor analysis and general structure modeling are the most flexible, powerful, and appropriate choices for social work data.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0002
- Subject:
- Social Work, Research and Evaluation
This chapter discusses a number of theoretical and statistical concepts and principles that are central to SEM. It introduces SEM notation and equations in the context of more familiar graphics and ...
More
This chapter discusses a number of theoretical and statistical concepts and principles that are central to SEM. It introduces SEM notation and equations in the context of more familiar graphics and terminology. It explains the role of matrices in SEM analyses.Less
This chapter discusses a number of theoretical and statistical concepts and principles that are central to SEM. It introduces SEM notation and equations in the context of more familiar graphics and terminology. It explains the role of matrices in SEM analyses.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0005
- Subject:
- Social Work, Research and Evaluation
This chapter describes how social workers use general structural equation models (general SEMs) and explains how to specify and test them. General SEMs include the measurement model of latent ...
More
This chapter describes how social workers use general structural equation models (general SEMs) and explains how to specify and test them. General SEMs include the measurement model of latent variables and their indicators, as well as the structural model of directional relationships among latent variables. A measurement model becomes a general SEM when some or all of the correlational relationships among latent variables in the measurement model are respecified as directional relationships based on the researcher's substantive knowledge of the topic (i.e., theory and past research).Less
This chapter describes how social workers use general structural equation models (general SEMs) and explains how to specify and test them. General SEMs include the measurement model of latent variables and their indicators, as well as the structural model of directional relationships among latent variables. A measurement model becomes a general SEM when some or all of the correlational relationships among latent variables in the measurement model are respecified as directional relationships based on the researcher's substantive knowledge of the topic (i.e., theory and past research).
Elinor Scarbrough and Eric Tanenbaum (eds)
- Published in print:
- 1998
- Published Online:
- November 2003
- ISBN:
- 9780198292371
- eISBN:
- 9780191600159
- Item type:
- book
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/0198292376.001.0001
- Subject:
- Political Science, Reference
This volume is a collection of commissioned articles by 16 experts in social science methodology, each contribution introducing experienced social scientists to more advanced analytic techniques. The ...
More
This volume is a collection of commissioned articles by 16 experts in social science methodology, each contribution introducing experienced social scientists to more advanced analytic techniques. The contributions explain the theoretical underpinnings of a particular technique, and illustrate the approach with a worked example. The techniques covered are the basic regression model and its extensions, linear structural equation modelling, log‐linear and latent class models, multi‐level modelling, and three extensions to modelling time series data. In these contributions, statistical notation is kept to a minimum; where necessary, it is consigned to footnotes or an appendix. Three final contributions introduce new developments in rational choice theory and discourse analysis.Less
This volume is a collection of commissioned articles by 16 experts in social science methodology, each contribution introducing experienced social scientists to more advanced analytic techniques. The contributions explain the theoretical underpinnings of a particular technique, and illustrate the approach with a worked example. The techniques covered are the basic regression model and its extensions, linear structural equation modelling, log‐linear and latent class models, multi‐level modelling, and three extensions to modelling time series data. In these contributions, statistical notation is kept to a minimum; where necessary, it is consigned to footnotes or an appendix. Three final contributions introduce new developments in rational choice theory and discourse analysis.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0008
- Subject:
- Social Work, Research and Evaluation
This chapter distills elements of the best practices highlighted in previous chapters and provides concluding remarks about how to become a skillful and critical researcher with structural equation ...
More
This chapter distills elements of the best practices highlighted in previous chapters and provides concluding remarks about how to become a skillful and critical researcher with structural equation modeling. Among these are that a sound SEM analysis should be guided by a theoretical model, and any subjective decision made by a user should have theoretical justification and rationale; when facing multiple choices in modeling, other things being equal, the user should choose the model that is most parsimonious; and the user should conduct sensitivity analyses to check violations of assumptions embedded in a model.Less
This chapter distills elements of the best practices highlighted in previous chapters and provides concluding remarks about how to become a skillful and critical researcher with structural equation modeling. Among these are that a sound SEM analysis should be guided by a theoretical model, and any subjective decision made by a user should have theoretical justification and rationale; when facing multiple choices in modeling, other things being equal, the user should choose the model that is most parsimonious; and the user should conduct sensitivity analyses to check violations of assumptions embedded in a model.
Judea Pearl
- Published in print:
- 2011
- Published Online:
- September 2011
- ISBN:
- 9780199574131
- eISBN:
- 9780191728921
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780199574131.003.0033
- Subject:
- Mathematics, Logic / Computer Science / Mathematical Philosophy
This chapter presents a general theory of causation based on the Structural Causal Model (SCM) described by Pearl (2000a). The theory subsumes and unifies current approaches to causation, including ...
More
This chapter presents a general theory of causation based on the Structural Causal Model (SCM) described by Pearl (2000a). The theory subsumes and unifies current approaches to causation, including graphical, potential outcome, probabilistic, decision analytical, and structural equation models, and provides both a mathematical foundation and a friendly calculus for the analysis of causes and counterfactuals. In particular, the chapter demonstrates how the theory engenders a coherent methodology for inferring (from a combination of data and assumptions) answers to three types of causal queries: (1) queries about the effects of potential interventions, (2) queries about probabilities of counterfactuals, and (3) queries about direct and indirect effects.Less
This chapter presents a general theory of causation based on the Structural Causal Model (SCM) described by Pearl (2000a). The theory subsumes and unifies current approaches to causation, including graphical, potential outcome, probabilistic, decision analytical, and structural equation models, and provides both a mathematical foundation and a friendly calculus for the analysis of causes and counterfactuals. In particular, the chapter demonstrates how the theory engenders a coherent methodology for inferring (from a combination of data and assumptions) answers to three types of causal queries: (1) queries about the effects of potential interventions, (2) queries about probabilities of counterfactuals, and (3) queries about direct and indirect effects.
Jon Williamson
- Published in print:
- 2004
- Published Online:
- September 2007
- ISBN:
- 9780198530794
- eISBN:
- 9780191712982
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780198530794.003.0010
- Subject:
- Mathematics, Logic / Computer Science / Mathematical Philosophy
The framework developed thus far in the book is extended to cope with the situation in which causal relations themselves are causes or effects. Examples of this phenomenon are presented and the ...
More
The framework developed thus far in the book is extended to cope with the situation in which causal relations themselves are causes or effects. Examples of this phenomenon are presented and the formalism of recursive Bayesian nets is developed to handle these cases.Less
The framework developed thus far in the book is extended to cope with the situation in which causal relations themselves are causes or effects. Examples of this phenomenon are presented and the formalism of recursive Bayesian nets is developed to handle these cases.
Patrick Dattalo
- Published in print:
- 2008
- Published Online:
- January 2009
- ISBN:
- 9780195315493
- eISBN:
- 9780199865475
- Item type:
- book
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195315493.001.0001
- Subject:
- Social Work, Research and Evaluation
Sample size determination is an important and often difficult step in planning an empirical study. From a statistical perspective, sample size depends on the following factors: type of analysis to be ...
More
Sample size determination is an important and often difficult step in planning an empirical study. From a statistical perspective, sample size depends on the following factors: type of analysis to be performed, desired precision of estimates, kind and number of comparisons to be made, number of variables to be examined, and heterogeneity of the population to be sampled. Other important considerations include feasibility, such as ethical limitations on access to a population of interest and the availability of time and money. The primary assumption of the book is that, within the context of ethical and practical limitations, efforts to obtain samples of appropriate size and quality remain an important and viable component of social science research. This text describes the following available approaches for estimating sample size in social work research and discusses their strengths and weaknesses: power analysis; heuristics or rules-of-thumb; confidence intervals; computer-intensive strategies; and ethical and cost considerations. In addition, strategies for mitigating pressures to increase sample size, such as emphasis on model parsimony (e.g., fewer dependent and independent variables), simpler study designs, an emphasis on replication, and careful planning of analyses are discussed. The book covers sample-size determination for advanced and emerging statistical strategies, such as structural equation modeling, multi-level analysis, repeated measures MANOVA, and repeated measures ANOVA which are not discussed in other texts.Less
Sample size determination is an important and often difficult step in planning an empirical study. From a statistical perspective, sample size depends on the following factors: type of analysis to be performed, desired precision of estimates, kind and number of comparisons to be made, number of variables to be examined, and heterogeneity of the population to be sampled. Other important considerations include feasibility, such as ethical limitations on access to a population of interest and the availability of time and money. The primary assumption of the book is that, within the context of ethical and practical limitations, efforts to obtain samples of appropriate size and quality remain an important and viable component of social science research. This text describes the following available approaches for estimating sample size in social work research and discusses their strengths and weaknesses: power analysis; heuristics or rules-of-thumb; confidence intervals; computer-intensive strategies; and ethical and cost considerations. In addition, strategies for mitigating pressures to increase sample size, such as emphasis on model parsimony (e.g., fewer dependent and independent variables), simpler study designs, an emphasis on replication, and careful planning of analyses are discussed. The book covers sample-size determination for advanced and emerging statistical strategies, such as structural equation modeling, multi-level analysis, repeated measures MANOVA, and repeated measures ANOVA which are not discussed in other texts.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0004
- Subject:
- Social Work, Research and Evaluation
This chapter describes when and how to conduct a confirmatory factor analysis (CFA). CFA is a step in the scale development process, and it is also the first step in testing structural models. ...
More
This chapter describes when and how to conduct a confirmatory factor analysis (CFA). CFA is a step in the scale development process, and it is also the first step in testing structural models. Therefore, all researchers using a latent variable analysis approach must have an understanding of CFA, whether or not they are developing and testing a new scale. CFA is also compared to exploratory factor analysis (EFA).Less
This chapter describes when and how to conduct a confirmatory factor analysis (CFA). CFA is a step in the scale development process, and it is also the first step in testing structural models. Therefore, all researchers using a latent variable analysis approach must have an understanding of CFA, whether or not they are developing and testing a new scale. CFA is also compared to exploratory factor analysis (EFA).
Peter Miksza and Kenneth Elpus
- Published in print:
- 2018
- Published Online:
- March 2018
- ISBN:
- 9780199391905
- eISBN:
- 9780199391943
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/oso/9780199391905.003.0014
- Subject:
- Music, Theory, Analysis, Composition, Performing Practice/Studies
This chapter presents structural equation modeling as a tool for conducting research regarding how collections of variables may be related to each other as well as to a particular outcome or even ...
More
This chapter presents structural equation modeling as a tool for conducting research regarding how collections of variables may be related to each other as well as to a particular outcome or even multiple outcomes. Structural equation modeling refers to a collection of analytical techniques that can be used to model complex patterns of predictive relationships among a collection of both measured and latent variables. As a statistical tool, structural equation modeling combines the features of regression and factor analysis. The chapter offers conceptual illustrations and practical steps for carrying out structural equation modeling by describing mediation and moderation analyses in the context of music education research.Less
This chapter presents structural equation modeling as a tool for conducting research regarding how collections of variables may be related to each other as well as to a particular outcome or even multiple outcomes. Structural equation modeling refers to a collection of analytical techniques that can be used to model complex patterns of predictive relationships among a collection of both measured and latent variables. As a statistical tool, structural equation modeling combines the features of regression and factor analysis. The chapter offers conceptual illustrations and practical steps for carrying out structural equation modeling by describing mediation and moderation analyses in the context of music education research.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0006
- Subject:
- Social Work, Research and Evaluation
Sometimes instead of getting results when they run an SEM analysis, researchers are confronted with discouraging messages about programming errors, data problems, or other causes of estimation ...
More
Sometimes instead of getting results when they run an SEM analysis, researchers are confronted with discouraging messages about programming errors, data problems, or other causes of estimation failures. This chapter first summarizes possible causes of estimation failures. It then provides guidelines for interpreting the results of successful estimation procedures both statistically and substantively. Finally, it discusses strategies for improving fit when model test results are valid (i.e., the model ran and converged, all parameters estimates are within valid ranges, and no errors are reported by the program) but unsatisfactory (i.e., fit criteria are not met).Less
Sometimes instead of getting results when they run an SEM analysis, researchers are confronted with discouraging messages about programming errors, data problems, or other causes of estimation failures. This chapter first summarizes possible causes of estimation failures. It then provides guidelines for interpreting the results of successful estimation procedures both statistically and substantively. Finally, it discusses strategies for improving fit when model test results are valid (i.e., the model ran and converged, all parameters estimates are within valid ranges, and no errors are reported by the program) but unsatisfactory (i.e., fit criteria are not met).
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0001
- Subject:
- Social Work, Research and Evaluation
This introductory chapter first sets out the purpose of the book, which is to serve as a concise practical guide for the informed and responsible use of structural equation modeling (SEM). It is ...
More
This introductory chapter first sets out the purpose of the book, which is to serve as a concise practical guide for the informed and responsible use of structural equation modeling (SEM). It is designed for social work faculty, researchers, and doctoral students who view themselves more as substantive experts than statistical experts, but who need to use SEM in their research. It is designed for social workers who desire a degree of analytical skill but have neither the time for coursework nor the patience to glean from the immense SEM literature the specifics needed to carry out an SEM analysis. The chapter then discusses what is SEM, the role of theory in SEM, the kinds of data that can or should be analyzed with SEM, and the research questions best answered by SEM.Less
This introductory chapter first sets out the purpose of the book, which is to serve as a concise practical guide for the informed and responsible use of structural equation modeling (SEM). It is designed for social work faculty, researchers, and doctoral students who view themselves more as substantive experts than statistical experts, but who need to use SEM in their research. It is designed for social workers who desire a degree of analytical skill but have neither the time for coursework nor the patience to glean from the immense SEM literature the specifics needed to carry out an SEM analysis. The chapter then discusses what is SEM, the role of theory in SEM, the kinds of data that can or should be analyzed with SEM, and the research questions best answered by SEM.
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/9780195339888.003.0001
- Subject:
- Social Work, Research and Evaluation
This chapter discusses the major uses of confirmatory factor analysis including measurement development, psychometric evaluation of measures, construct validation, testing method effects, and testing ...
More
This chapter discusses the major uses of confirmatory factor analysis including measurement development, psychometric evaluation of measures, construct validation, testing method effects, and testing measurement invariance, and presents examples of these uses in the social work literature. Different types of validity are briefly defined and discussed. Confirmatory factor analysis is compared with three other common data analysis approaches: exploratory factor analysis, principal components analysis, and structural equation modeling. Software for conducting confirmatory factor analysis is briefly discussed.Less
This chapter discusses the major uses of confirmatory factor analysis including measurement development, psychometric evaluation of measures, construct validation, testing method effects, and testing measurement invariance, and presents examples of these uses in the social work literature. Different types of validity are briefly defined and discussed. Confirmatory factor analysis is compared with three other common data analysis approaches: exploratory factor analysis, principal components analysis, and structural equation modeling. Software for conducting confirmatory factor analysis is briefly discussed.
Christian Büchel and Karl Friston
- Published in print:
- 2001
- Published Online:
- March 2012
- ISBN:
- 9780192630711
- eISBN:
- 9780191724770
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780192630711.003.0016
- Subject:
- Neuroscience, Techniques
This chapter shows that new methods for measuring effective connectivity allow us to characterize the interactions between brain regions which underlie the complex interactions among different ...
More
This chapter shows that new methods for measuring effective connectivity allow us to characterize the interactions between brain regions which underlie the complex interactions among different processing stages of functional architectures. It reviews the basic concepts of effective connectivity in neuroimaging. The methods introduced to assess effective connectivity are multiple linear regression, covariance structural equation modelling and variable parameter regression. The first example demonstrates that non-linear interactions can be characterized using simple extensions of linear models, while in the second, structural equation modelling is introduced as a device that allows one to combine observed changes in cortical activity and anatomical models. Finally, the chapter concludes that the approach to neuroimaging data and regional interactions is an exciting endeavour, which is starting to attract more attention.Less
This chapter shows that new methods for measuring effective connectivity allow us to characterize the interactions between brain regions which underlie the complex interactions among different processing stages of functional architectures. It reviews the basic concepts of effective connectivity in neuroimaging. The methods introduced to assess effective connectivity are multiple linear regression, covariance structural equation modelling and variable parameter regression. The first example demonstrates that non-linear interactions can be characterized using simple extensions of linear models, while in the second, structural equation modelling is introduced as a device that allows one to combine observed changes in cortical activity and anatomical models. Finally, the chapter concludes that the approach to neuroimaging data and regional interactions is an exciting endeavour, which is starting to attract more attention.
James B. Grace, Samuel M. Scheiner, and Donald R. Schoolmaster, Jr.
- Published in print:
- 2015
- Published Online:
- April 2015
- ISBN:
- 9780199672547
- eISBN:
- 9780191796487
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780199672547.003.0009
- Subject:
- Biology, Biomathematics / Statistics and Data Analysis / Complexity Studies, Ecology
This chapter describes structural equation modeling (SEM), which represents a probabilistic modeling framework for studying causal hypotheses about systems. SEM relies on interconnected series of ...
More
This chapter describes structural equation modeling (SEM), which represents a probabilistic modeling framework for studying causal hypotheses about systems. SEM relies on interconnected series of equations to represent networks as complex hypotheses. As a general modeling methodology, SEM potentially permits any type of functional response (error distributions) and linkage form (linear, non-linear, or non-parametric). The methodology also includes procedures for evaluating proposed models against data, permitting the discovery of unsuspected mechanisms leading to an understanding of how multiple processes collectively control systems. A key element of SEM is the use of graphical models to represent the causal logic implied by the equations. In this treatment the chapter concisely describes SEM fundamentals, incorporating the latest advances in the core methodology. Worked examples are used to illustrate procedures.Less
This chapter describes structural equation modeling (SEM), which represents a probabilistic modeling framework for studying causal hypotheses about systems. SEM relies on interconnected series of equations to represent networks as complex hypotheses. As a general modeling methodology, SEM potentially permits any type of functional response (error distributions) and linkage form (linear, non-linear, or non-parametric). The methodology also includes procedures for evaluating proposed models against data, permitting the discovery of unsuspected mechanisms leading to an understanding of how multiple processes collectively control systems. A key element of SEM is the use of graphical models to represent the causal logic implied by the equations. In this treatment the chapter concisely describes SEM fundamentals, incorporating the latest advances in the core methodology. Worked examples are used to illustrate procedures.
Rami Benbenishty and Ron Avi Astor
- Published in print:
- 2005
- Published Online:
- January 2009
- ISBN:
- 9780195157802
- eISBN:
- 9780199864393
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195157802.003.0008
- Subject:
- Social Work, Children and Families, Crime and Justice
This chapter presents structural equation models of how school context variables (risky peer behaviors, school policies, teachers' support and student participation) combined with victimization lead ...
More
This chapter presents structural equation models of how school context variables (risky peer behaviors, school policies, teachers' support and student participation) combined with victimization lead to different kinds of students' subjective interpretations about their school: assessment of the severity of school violence, missing school due to fear, and a sense of safety. Key findings and implications include: (1) the amount of variance explained for the subjective assessment by within school variables alone is quite high; (2) different subjective interpretive outcomes are influenced by different types of victimization and school variables; (3) students' views of their school violence problem are influenced mainly by the risky peer behaviors and the response of the school staff to violent events; (4) non-attendance due to fear is influenced mainly by personal experiences with severe events of peer-violence; (5) a sense of safety is influenced mainly by a positive school climate; and (6) these patterns are true for all school levels studied and across ethnicity and gender.Less
This chapter presents structural equation models of how school context variables (risky peer behaviors, school policies, teachers' support and student participation) combined with victimization lead to different kinds of students' subjective interpretations about their school: assessment of the severity of school violence, missing school due to fear, and a sense of safety. Key findings and implications include: (1) the amount of variance explained for the subjective assessment by within school variables alone is quite high; (2) different subjective interpretive outcomes are influenced by different types of victimization and school variables; (3) students' views of their school violence problem are influenced mainly by the risky peer behaviors and the response of the school staff to violent events; (4) non-attendance due to fear is influenced mainly by personal experiences with severe events of peer-violence; (5) a sense of safety is influenced mainly by a positive school climate; and (6) these patterns are true for all school levels studied and across ethnicity and gender.
Patrick Dattalo
- Published in print:
- 2008
- Published Online:
- January 2009
- ISBN:
- 9780195315493
- eISBN:
- 9780199865475
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195315493.003.0002
- Subject:
- Social Work, Research and Evaluation
This chapter provides a brief description of the rationale and limitations of statistical power analysis, and presents important issues related to determining sample size for both commonly used and ...
More
This chapter provides a brief description of the rationale and limitations of statistical power analysis, and presents important issues related to determining sample size for both commonly used and emerging statistical procedures in social work research. Procedures include difference between two means, difference between two proportions, odds ratio, relative risk, chi square, ANOVA, correlation, regression, logistic regression, structural equation modeling, and multi-level analysis.Less
This chapter provides a brief description of the rationale and limitations of statistical power analysis, and presents important issues related to determining sample size for both commonly used and emerging statistical procedures in social work research. Procedures include difference between two means, difference between two proportions, odds ratio, relative risk, chi square, ANOVA, correlation, regression, logistic regression, structural equation modeling, and multi-level analysis.
Natasha K. Bowen and Shenyang Guo
- Published in print:
- 2011
- Published Online:
- January 2012
- ISBN:
- 9780195367621
- eISBN:
- 9780199918256
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780195367621.003.0007
- Subject:
- Social Work, Research and Evaluation
This chapter discusses three advanced structural equation modeling topics: how to conduct a power analysis for SEM; how to prevent and solve problems of underidentification; and how to conduct a ...
More
This chapter discusses three advanced structural equation modeling topics: how to conduct a power analysis for SEM; how to prevent and solve problems of underidentification; and how to conduct a multiple-group analysis.Less
This chapter discusses three advanced structural equation modeling topics: how to conduct a power analysis for SEM; how to prevent and solve problems of underidentification; and how to conduct a multiple-group analysis.
Ricardo Silva
- Published in print:
- 2011
- Published Online:
- September 2011
- ISBN:
- 9780199574131
- eISBN:
- 9780191728921
- Item type:
- chapter
- Publisher:
- Oxford University Press
- DOI:
- 10.1093/acprof:oso/9780199574131.003.0032
- Subject:
- Mathematics, Logic / Computer Science / Mathematical Philosophy
The presence of latent variables makes the task of estimating causal effects difficult. In particular, it might not even be possible to record important variables without measurement error, a common ...
More
The presence of latent variables makes the task of estimating causal effects difficult. In particular, it might not even be possible to record important variables without measurement error, a common fact in fields such as psychology and social sciences. A fair amount of theory is often used to design instruments to indirectly measure such latent variables, such that one obtains estimates of measurement error. If the measurement error is known, then causal effects can be identified in a variety of scenarios. Unfortunately, a strictly theoretical approach for formalizing a measurement model is error prone and does not provide alternative models that could equally or better explain the data. The chapter introduces an algorithmic approach that, given a set of observed indicators of latent phenomena of interest and common assumptions about the causal structure of the world, provides a set of measurement models compatible with the observed data. This approach extends previous results in the literature which would select an observed variable only if it measured a single latent variable. The extensions cover cases where some variables are allowed to be indicators of more than one hidden common cause.Less
The presence of latent variables makes the task of estimating causal effects difficult. In particular, it might not even be possible to record important variables without measurement error, a common fact in fields such as psychology and social sciences. A fair amount of theory is often used to design instruments to indirectly measure such latent variables, such that one obtains estimates of measurement error. If the measurement error is known, then causal effects can be identified in a variety of scenarios. Unfortunately, a strictly theoretical approach for formalizing a measurement model is error prone and does not provide alternative models that could equally or better explain the data. The chapter introduces an algorithmic approach that, given a set of observed indicators of latent phenomena of interest and common assumptions about the causal structure of the world, provides a set of measurement models compatible with the observed data. This approach extends previous results in the literature which would select an observed variable only if it measured a single latent variable. The extensions cover cases where some variables are allowed to be indicators of more than one hidden common cause.
Horace A. Bartilow
- Published in print:
- 2019
- Published Online:
- September 2020
- ISBN:
- 9781469652559
- eISBN:
- 9781469652573
- Item type:
- chapter
- Publisher:
- University of North Carolina Press
- DOI:
- 10.5149/northcarolina/9781469652559.003.0006
- Subject:
- Public Health and Epidemiology, Public Health
To test the theoretical components of the argument presented in chapter 5, this chapter develops an empirical model of how U.S. transnational corporations and paramilitary death squads mediate the ...
More
To test the theoretical components of the argument presented in chapter 5, this chapter develops an empirical model of how U.S. transnational corporations and paramilitary death squads mediate the U.S.-sponsored drug war’s effect on human rights repression in Latin America. In outlining this empirical model, this chapter is organized as follows: It first juxtapose the theoretical arguments of dependency and neoclassical liberal theories regarding the human rights effects of transnational capital by highlighting the theoretical and empirical limitations of neoclassical liberal claims. This is followed by a discussion of the empirical model, which draws on the extant human rights literature to identify important control variables that are important predictors of state repression. It then discusses important theoretical modifications that are incorporated into the overall empirical model. This is followed by a discussion of the limitations of the indicators used to measure the model’s mediating variables. structural equation modeling is used to analyze cross-national data for thirty-one countries from the Latin American region covering the period 1980 to 2012. All the components of the theoretical argument found strong statistical support.Less
To test the theoretical components of the argument presented in chapter 5, this chapter develops an empirical model of how U.S. transnational corporations and paramilitary death squads mediate the U.S.-sponsored drug war’s effect on human rights repression in Latin America. In outlining this empirical model, this chapter is organized as follows: It first juxtapose the theoretical arguments of dependency and neoclassical liberal theories regarding the human rights effects of transnational capital by highlighting the theoretical and empirical limitations of neoclassical liberal claims. This is followed by a discussion of the empirical model, which draws on the extant human rights literature to identify important control variables that are important predictors of state repression. It then discusses important theoretical modifications that are incorporated into the overall empirical model. This is followed by a discussion of the limitations of the indicators used to measure the model’s mediating variables. structural equation modeling is used to analyze cross-national data for thirty-one countries from the Latin American region covering the period 1980 to 2012. All the components of the theoretical argument found strong statistical support.
