Read the summary and the most important questions on Principles and practice of structural equation modeling | 9781606238769 | Rex B. Kline.

• 1 Introduction

  This is a preview. There are 1 more flashcards available for chapter 1
  Show more cards here

• 1.3 explicit distinction between observed and latent variables

  This is a preview. There are 3 more flashcards available for chapter 1.3
  Show more cards here

• What is the residual error of indicator? 

  A residual represents variance unexplained by the factor that the corresponding indicator is supposed to measure.

• What is the unexplained variance of the residual error of the indicator? 

  The unexplained variance of the residual error could be due to random measurement error or score unreliability 

• Does SEM takes into account residual error?

  Yes, SEM does take into account explicity the residual error. This makes the model more realistic than for example Multiple Regression that assumes that the residual error is zero. 

• 1.4 Covariances always, but Means Can Be analyzed, too

  This is a preview. There are 3 more flashcards available for chapter 1.4
  Show more cards here

• What are the two main goals concerning the use of covariances in SEM?

  1. Analyse patterns of covariance among a set of observed variables. 

  2. Explain as much of the variance possible with the researcher's model. 

• What is the covariance structure? 

  The part of SEM that represents hypothesis about variance and covariance. 

• What is the added value to have a mean structure possibility in SEM?

  The added value of mean structure is that this makes it possible to test whether latent variables differ for groups, for example boys and girls, within the model. 

• 1.5 SEM requires large samples

  This is a preview. There are 2 more flashcards available for chapter 1.5
  Show more cards here

• What is the most common rule for sample size when using maximum likelihood estimation?

  The N:q rule of Jackson (2003). Jackson (2003) suggest that researchers should think about the sample size in terms of the ratio of number of cases (N) to the number of model parameters that require statisitical estimates (q). The ideal ratio is 20:1

• When is the N:q ratio a problem?

  When the N:q ratio is lower than 10:1 the trustworthiness of the results decreases a lot!

• 1.7 SEM and the general linear Model

• What's the relationship between SEM and GLM?

  GLM is a more restricted case of SEM. As SEM gives also the possibility for more causal relationships and rejecting the whole model (or not) and therefore to study a whole model level. 

• 1.8 Widespread enthusIasm, but with a cautionary tale

  This is a preview. There are 2 more flashcards available for chapter 1.8
  Show more cards here

• What is it called when you ignore the existence of an alternative / equivalent model?

  It is called confirmation bias, as overly positive evaluation of one specific model exists.