Description Changes in version 1.1.4-4 Changes in version 1.1.4-3 Changes in version 1.1.4-2 Changes in version 1.1.4-1 Changes in version 1.1.4 Changes in version 1.1.3.6 Changes in version 1.1.3.5 Changes in version 1.1.3.5 Changes in version 1.1.3 Changes in version 1.1.2 Changes in version 1.1.1

The function performs a parallel analysis using simulated polychoric correlation matrices. The function will extract the eigenvalues from each random generated polychoric correlation matrix and from the polychoric correlation matrix of real data. A plot comparing eigenvalues extracted from the specified real data with simulated data will help determine which of real eigenvalue outperform random data. A series of matrices comparing MAP vs PA-Polychoric vs PA-Pearson correlations methods, FA vs PCA solutions are finally presented. Random data sets are simulated assuming or a uniform or a multinomial distribution or via the bootstrap method of resampling (i.e., random permutations of cases). Also Multigroup Parallel analysis is made available for random (uniform and multinomial distribution and with or without difficulty factor) and bootstrap methods. An option to choose between default or full output is also available as well as a parameter to print Fit Statistics (Chi-squared, TLI, RMSEA, RMR and BIC) for the factor solutions indicated by the Parallel Analysis.

Version 1.1.4-4 the following parameters were added: 1) `weights`

allows the reader to pass a vector of weights to compute a weighted random polychor (or pearson) correlation matrix and simulate weighted samples. If also `bootstrap`

is selected then weighted samples will be bootstrapped. Finally if `multisample`

is also selected then the `random.polychor.pa`

function will simulate weighted random samples for each sub-sample.
Version 1.1.4-4 fixed the following issues: 1) if a value is passed for continuity check now the correct values is displayed in the output, and not the fixed values of .5 as was in previous version; 2) in previous versions if `fit.pa`

parameter was selected it was not infrequent that the `random.polychor.pa()`

function unexpectedly stopped after having run part of the simulations. The problem was due to the fact that for some simulated data, the function found non-valid value for RMSEA and/or for BIC indices. In the present version a series of checks were added to prevent the function to stop unxepectedly for these problems linked to fit indices estimation.

Version 1.1.4-3 fixed two problems: in example 1 we use bfi data from psych package as data example. However the datafile has been moved to `psychTools`

so the example is modified accordingly. The second problem concerns an unexpected crash of `random.polychoric.pa`

when calling `polychoric`

function (from `psych`

package) due to the correction for continuity that is set by default to 0.5 (i.e. `correct=TRUE`

in `polychoric`

function. Correction for continuity is used for replacing cells with zero counts. This correction for continuity in some situations determines NAs that causes the `polychoric`

function to stop. To handle such problem we added a parameter to the `random.polychor.pa`

function to set to 0.0 the correction for continuity. Users are warned that polychoric correlation matrices with and without correction for continuity differ.

Version 1.1.4-2 fixed minor bugs when running the example 1, and when displaying the time needed to complete the simulations.

Version 1.1.4-1 a problem with the `psych`

dependency was fixed:

the option (

`polycor=TRUE`

) in the`polychoric`

function was removed and consequently it is also no more possible to call the polycor function in running the`random.polychor.pa`

function

Version 1.1.4 added a number of cahnges:

a parameter

`distr`

allows to shift the simulation from uniform distribtuion to multinomial distributionBootstrap method (with random permutations of cases) was added

Multi-Group for random and bootstrap version of the Parallel Analysis is made available

Fit statistics (Chi-squared, TLI, RMSEA, RMR, BIC) for all factor solution indicated by Parallel Analysis

option to print a default output (number of factors indicated by Parallel Analysis) or a full output (adds: matrices of simulated and empirical eigenvalues for random, bootstrap, and multigroup)

In version 1.1.3.6 a check for the range of quantile (between 0 and 1) was added.

The search for zeroes within the provided datafile was removed, so data with zeroes are now accepted.

In version 1.1.3.5 a paramether was added,

`diff.fact`

, in order to simulate random dataset with the same probability of observing each category for each variable as that observed in the provided (empirical) dataset.

Version 1.1.3 tackles two problems signalled by users:

the possibility to make available the results of simulation for plotting them in other software. Now the

`random.polychor.pa()`

will show, upon request, all the data used in the scree-plot.The function

`polichoric()`

of the`psych()`

package does not handle data matrices that include 0 as possible category and will cause the function to stop with error. So a check for the detection of the 0 code within the provided data.matrix is now added and will cause the random.polychor.pa function to stop with a warning message.

Version 1.1.2 simply has updated the function that calculates the polychoric correlation matrix due to changes in the

`psych()`

package.

Version 1.1.1, fixed a minor bug in the regarding the estimated time needed to complete the simulation.

Also in this version, the function is now able to manage supplied data.matrix in which variables representing factors (i.e., variables with ordered categories) are present and may cause an error when the Pearson correlation matrix is calculated.

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.