CASCOR: Correlate Pairs of Cases

Last updated 06 Jan 99

CASCOR will read a Case Data file and correlate each specified pair of cases. The mean and standard deviation of the raw ratings is reported, as well as the number of tasks performed. An overall average Fisher Z is calculated. This program is designed for studies which have Time 1 / Time 2 data, i.e., incumbents (or raters) who completed the survey twice.

Program Invocation

The specific syntax for program invocation varies between operating systems. This document diplays the code for the AIX version.


@codap cascor <opt> <ssss> <CDf>

opt:  R   Roster the control cards.  Useful for adding 
          identifying information about the pair.
ssss:     The study number 
CDf:      Input Case Data file

Request Cards

These cards define each pair of cases to be correlated. The cases are defined by the Case Control number assigned by INPSTD (C0001), not the booklet ID or KPATH number. These control cards are terminated by an "@EOF" card or an end-of-file condition.


<ccn1> <ccn2> <info>

<ccn1>   cc 01-04  Case control number for this case at time 1
<ccn2>   cc 06-09  Case control number for this case at time 2
<info>   cc 11-70  Optional identifying information, such as name

Example

@codap cascor r ssss cd100
 469 1569 Herbert
 303 1604 Burgoss
 810 1730 Hooper
 738 1464 Otero
 228 1361 Smith
@eof

Correlations will be calculated for five pairs of cases. An overall Fisher Z will also be computed. The control cards are rostered because of the 'R' option. This can help the user associate the correlation with an individual.

Back to document index