-------------------------------------------------------------------------------
help for prepar                                                   (Feb 4, 2010)
-------------------------------------------------------------------------------

Description

Writes the code and data file needed to process variables in PARSCALE, with the option to run PARSCALE and merge in the scores and standard errors.

prepar varlist , ID(subject id) RUnname(string) [MINsize(integer 20)) SCALE(string 1.7) NQpt(string 11) CYcles(string 1000) SCORE(string EAP) PARameterfile(string) SKIP(varlist} ALL(integer 0)) CLeanup(integer 0)) ]

where:

varlist is the list of variables (items, blocks) to be included in the PARSCALE run.

ID is the subject id variable.

runname is the name of the PARSCALE run to be generated. code_runname.psl will be the PARSCALE input file, data_runname.txt will be the PARSCALE data file.

Options

MINsize is the minimum category frequency (number of observations/category). The default is 20. Categories will be combined, if possible, to meet the minsize. Any variable that does not have enough observations in at least 2 categories will be dropped. If you do not want any categories combined, specify minsize(1).

SCALE changes the scale parameter from the default of 1.7.

nqpt - changes the number of quadrature points from a default of 11.

cycles - changes the maximun number of iterations in PARSCALE stage 2 from the default of 1000.

SCORE Default is EAP. You can specify MLE or WML instead.

PARameterfile calculates scores using an existing PARSCALE parameter file.

SKIP uses the existing PARSCALE parameter file only for the variables specified here.

ALL ALL(1) runs the PARSCALE code, and merges the scores and their standard errors into the original data set.

CLeanup CLeanup(1) deletes the PARSCALE files when the merge is completed. CLeanup(2) deletes all but the PAR file. Only for use with the ALL option.

Remarks

Drops any variable that does not have enough observations for at least 2 categories, and displays a warning message.

Warning message for variables with more than 15 levels (PARSCALE will reject). Letter codes automatically made for 2-digit values.

Allows missing values.

Allows character or numeric ID's up to 9 digits.

Creates a numeric, 10-digit ID code for PARSCALE by adding 1,000,000,000 to the ID, first converting string IDs to a unique number if needed.

Default PARSCALE CALIB instructions are:

>CALIB GRADED, LOGISTIC, SCALE=1.7, NQPT=11, CYCLES=1000, CRIT=.001

If you want other choices, you can edit the .PSL file before running it. [Except for SCALE, CYCLES, and NQPT, which you can choose on the command line.]

The maximum number of variables is 230.

Written for Stata 8.0.

Examples

prepar item1 - item13, id(id) ru(gender0)

prepar item1 - item13, id(id) ru(gender0) par(PRO) minsize(1) scale(1) score(MLE)

prepar item1 - item13, id(id) ru(gender0) par(PRO) minsize(1) skip(item1-item5) cleanup(2)

prepar item1 - item13, id(id) ru(gender0) all(1) cleanup(1)

Authors

Laura Gibbons and Paul Crane. University of Washington, Copyright 2005-2010. Email: gibbonsl@u.washington.edu

We appreciate the assistance of Tom Koepsell and Rich Jones.

Also see

difwithpar and runparscale.