SMHS BigDataBigSci SEM - Revision history

Pineaumi: /* See also */

2016-05-24T15:29:36Z

‎See also

Pineaumi: /* See also */

2016-05-24T14:07:27Z

‎See also

Dinov: /* Notation */

2016-05-21T16:32:52Z

‎Notation

Dinov: /* Model-based Analytics - Structural Equation Modeling (SEM) */

2016-05-21T16:26:44Z

‎Model-based Analytics - Structural Equation Modeling (SEM)

Imoubara: /* Notation */

2016-05-18T19:39:32Z

‎Notation

Imoubara at 19:33, 18 May 2016

2016-05-18T19:33:56Z

Pineaumi: /* Notation */

2016-05-16T16:41:08Z

‎Notation

Pineaumi: /* Notation */

2016-05-16T16:40:50Z

‎Notation

Imoubara at 15:49, 16 May 2016

2016-05-16T15:49:15Z

Pineaumi: /* Notation */

2016-05-13T18:26:23Z

‎Notation

@@ Line 48: / Line 48: @@
 ==See also==
 * [[SMHS_BigDataBigSci| Back to Model-based Analytics]]
 * [[SMHS_BigDataBigSci_GCM| Next Section: Growth Curve Modeling]]
 * [[SMHS_BigDataBigSci_GCM| Next Section: Generalized Estimating Equation (GEE) Modeling]]

@@ Line 50: / Line 50: @@
 * [[SMHS_BigDataBigSci| Back to Model-based Analytics]]
 * [[SMHS_BigDataBigSci_GCM| Next Section: Growth Curve Modeling]]
-* [[SMHS_BigDataBigSci_GEE| Next Section: Generalized Estimating Equation (GEE) Modeling]]
+* [[SMHS_BigDataBigSci_GCM| Next Section: Generalized Estimating Equation (GEE) Modeling]]
 <hr>

@@ Line 28: / Line 28: @@
 <center>[[Image:SMHS_BigDataBigSci1.png|500px]]</center>
-<mark> PLEASE FIX LAST ARROW *'''Relations: Direct effects''' (&rarr;), '''Reciprocal effects''' (&harr; or &#8646;), and '''Correlation or covariance''' ( ARROW ) all have different appearance in SEM models.</mark>
+*'''Relations: Direct effects''' (&rarr;), '''Reciprocal effects''' (&harr; or &#8646;), and '''Correlation or covariance''' (&#x293B; or &#x293A;) all have different appearance in SEM models.
 ===Model Components===

@@ Line 6: / Line 6: @@
 ===SEM Advantages===
-<li>It allows testing models with multiple dependent variables</li>
+* It allows testing models with multiple dependent variables
-<li>Provides mechanisms for modeling mediating variables</li>
+* Provides mechanisms for modeling mediating variables
-<li>Enables modeling of error terms</li>
+* Enables modeling of error terms
-<li>Facilitates modeling of challenging data  (longitudinal with auto-correlated errors, multi-level data, non-normal data, incomplete data)</li>
+* acilitates modeling of challenging data  (longitudinal with auto-correlated errors, multi-level data, non-normal data, incomplete data)
 SEM allows separation of observed and latent variables. Other standard statistical procedures may be viewed as special cases of SEM, where statistical significance less important, than in other techniques, and covariances are the core of structural equation models.
 ===Definitions===
-<li>The <b>disturbance</b>, <i>D</i>, is the variance in Y unexplained by a variable X that is assumed to affect Y. </li>
+*The <b>disturbance</b>, <i>D</i>, is the variance in Y unexplained by a variable X that is assumed to affect Y.
   X    →    Y   ←     D
-<li><b>Measurement error</b>, <i>E</i>, is the variance in X unexplained by A, where X is an observed variable that is presumed to measure a latent variable, <i>A</i>. </li>
+* <b>Measurement error</b>, <i>E</i>, is the variance in X unexplained by A, where X is an observed variable that is presumed to measure a latent variable, <i>A</i>.
   A    →    X   ←     E
-<li>Categorical variables in a model are <b>exogenous</b> (independent) or <b>endogenous</b> (dependent).</li>
+* Categorical variables in a model are <b>exogenous</b> (independent) or <b>endogenous</b> (dependent).
 ===Notation===
-<li>In SEM <b>observed (or manifest) indicators</b> are represented by <b>squares/rectangles</b> whereas latent variables (or factors) represented by circles/ovals.</li>
+* In SEM <b>observed (or manifest) indicators</b> are represented by <b>squares/rectangles</b> whereas latent variables (or factors) represented by circles/ovals.
 <center>[[Image:SMHS_BigDataBigSci1.png|500px]]</center>
@@ Line 38: / Line 38: @@
 ===Notes===
-<li> Sample-size considerations: mostly same as for regression - more is always better</li>
+* Sample-size considerations: mostly same as for regression - more is always better.
-<li> Model assessment strategies: Chi-square test, Comparative Fit Index, Root Mean Square Error, Tucker Lewis Index, Goodness of Fit Index, AIC, and BIC.</li>
+* Model assessment strategies: Chi-square test, Comparative Fit Index, Root Mean Square Error, Tucker Lewis Index, Goodness of Fit Index, AIC, and BIC.>
-<li> Choice for number of Indicator variables: depends on pilot data analyses, a priori concerns, fewer is better.</li>
+* Choice for number of Indicator variables: depends on pilot data analyses, a priori concerns, fewer is better.
 ===[[SMHS_BigDataBigSci_SEM_Ex1|Hands-on Example 1 (School Kids Mental Abilities)]]===
@@ Line 46: / Line 46: @@
 ===[[SMHS_BigDataBigSci_SEM_Ex2|Hands-on Example 2 (Parkinson’s Disease data)]]===
 ==See also==

@@ Line 26: / Line 26: @@
 <li>In SEM <b>observed (or manifest) indicators</b> are represented by <b>squares/rectangles</b> whereas latent variables (or factors) represented by circles/ovals.</li>
-[[Image:SMHS_BigDataBigSci1.png|500px]]
+<center>[[Image:SMHS_BigDataBigSci1.png|500px]]</center>
 <mark> PLEASE FIX LAST ARROW *'''Relations: Direct effects''' (&rarr;), '''Reciprocal effects''' (&harr; or &#8646;), and '''Correlation or covariance''' ( ARROW ) all have different appearance in SEM models.</mark>