A framework for power analysis using a structural equation modelling procedure

BackgroundThis paper demonstrates how structural equation modelling (SEM) can be used as a tool to aid in carrying out power analyses. For many complex multivariate designs that are increasingly being employed, power analyses can be difficult to carry out, because the software available lacks sufficient flexibility.Satorra and Saris developed a method for estimating the power of the likelihood ratio test for structural equation models. Whilst the Satorra and Saris approach is familiar to researchers who use the structural equation modelling approach, it is less well known amongst other researchers. The SEM approach can be equivalent to other multivariate statistical tests, and therefore the Satorra and Saris approach to power analysis can be used.MethodsThe covariance matrix, along with a vector of means, relating to the alternative hypothesis is generated. This represents the hypothesised population effects. A model (representing the null hypothesis) is then tested in a structural equation model, using the population parameters as input. An analysis based on the chi-square of this model can provide estimates of the sample size required for different levels of power to reject the null hypothesis.ConclusionsThe SEM based power analysis approach may prove useful for researchers designing research in the health and medical spheres.

[1]  Matthew J. Taylor,et al.  Why Covariance? A Rationale for Using Analysis of Covariance Procedures in Randomized Studies , 1993 .

[2]  Susanne Hempel,et al.  Perceived Parenting Styles, Depersonalisation, Anxiety and Coping Behaviour in Adolescents , 2003 .

[3]  R. Hoyle.  Structural equation modeling: concepts, issues, and applications , 1997 .

[4]  K. Jöreskog.  A general approach to confirmatory maximum likelihood factor analysis , 1969 .

[5]  R. MacCallum,et al.  Power analysis and determination of sample size for covariance structure modeling. , 1996 .

[6]  K. Bollen.  Latent variables in psychology and the social sciences. , 2002, Annual review of psychology.

[7]  D. Moher,et al.  The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials , 2001, The Lancet.

[8]  James H. Steiger,et al.  Driving Fast in Reverse , 2001 .

[9]  T. Wansbeek.  Measurement error and latent variables in econometrics , 2000 .

[10]  Andrew Rutherford,et al.  Introducing Anova and Ancova: A Glm Approach , 2000 .

[11]  A. Satorra,et al.  Power of the likelihood ratio test in covariance structure analysis , 1985 .

[12]  Edgar Erdfelder,et al.  GPOWER: a priori, post-hoc, and compromise power analyses for MS-DOS [Computer Program] , 2004 .

[13]  David Machin,et al.  Sample Size Tables for Clinical Studies , 1997 .

[14]  Scott E. Maxwell,et al.  How the power of MANOVA can both increase and decrease as a function of the intercorrelations among the dependent variables. , 1994 .

[15]  G. Keren,et al.  Between- or Within-Subjects Design: A Methodological Dilemma , 1993 .

[16]  D Kaplan,et al.  Asymptomatic Independence and Separability in Convariance Structure Models: Implications for Specification Error, Power, and Model Modification. , 1993, Multivariate behavioral research.

[17]  Jacob Cohen.  Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[18]  Stephen Dubin.  How many subjects? Statistical power analysis in research , 1990 .

[19]  K. Murphy,et al.  Statistical Power Analysis: A Simple and General Model for Traditional and Modern Hypothesis Tests, Second Ediction , 1998 .

[20]  James H. Steiger,et al.  Driving Fast In Reverse The Relationship Between Software Development, Theory, and Education in Structural Equation Modeling , 2001 .