instructions

How to use this site:

at the top left corner of the page there is a "search" bar. Type in keywords for whatever you're looking for here. If you don't find what you need, leave a comment on the "wish list" post; immediately following this one, or send us a message!
happy hunting!
Y and J

Wish List

WE still need some help finding these ideas in citations:
1. Suggested sample size for SEM models
2. What to do when you can't do RCTs
3. Number of indicators required for Latent Variables
4. HLM power analysis

Multiple Imputation rates

A basic identi ability requirement for the imputa-
tion model is that for each variable in the imputation the number of
observations should be at least as many as the number of variables in
the imputation model. Suppose that there are 50 variables in the im-
putation model and that there are 1000 observations in the data set.
Suppose that the rst variable has 960 missing values and only 40 ob-
served values. Since 40 < 50 the imputation model is not identi ed. That variable should be removed from the imputation or it should be imputed from a smaller data sets, for example with 30 variables. If for example there are 60 observed values and 940 missing values the model will be identi ed but essentially 51 parameters(1 mean parameter, 1 residual variance parameter and 49 regression coe cient parameters) in the imputation model are identi ed with only 60 observations. This may work, but would likely lead to slow convergence and poor mixing. So even though the model is identi ed, this variable would cause slow convergence and if the variable is not important one should consider dropping that variable from the imputation

From Asparouhov, T., & Muthen, B. (2010) Multiple Imputation with Mplus. http://www.statmodel.com/download/Imputations7.pdf

Experimental Linking of Items

When considering linking multiple forms for collecting data, a current rule of thumb is that there needs to be 20% item overlap for the *total* number of form items across all forms. Thus, if there are 3 forms, and each form has a total of 20 items, then 4 items on each form must be shared, and the other 16 items per form can be unique.

Latent Class Analysis: Entropy

Entropy statistic (printed in the Mplus output for Latent Profile and Class analysis)

Entropy values greater than .80 indicate a good separation of the identified groups (Ramaswamy et al., 1993)


Ramaswamy, V., DeSarbo, W. S., Reibstein, D. J., & Robinson, W. T. (1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science, 12(1), 103−124.

Latent Variable Number of Observations

In the context of IRT, some researchers have stated that simple 1PL models may be run with as few as 50 participants (Linacre, 1994), and 2PL models require a minimum of 250 individuals; however, several authors have recommended a sample of 500 for accurate parameter estimates in the 2PL model (Baur & Lukes, 2009; Embretson & Reise, 2000).

Petscher (2010)

examples

trying new examples