PRODCLIN - (MacKinnon, D. P., Fritz, M. S., Williams, J., & Lockwood, C. M. Distribution of the product confidence limits for the indirect effect program PRODCLIN. In Press, Behavioral Research Methods).

Prodclin is a program that uses the distribution of the product of two normally distributed variables to compute asymmetric confidence intervals for the mediated effect.  Prodclin consists of two parts: 1) The Fortran program prodclin.exe that computes the critical values for the distribution of the product and 2) a statistical package specific component that allows the user to input the values of the a and b paths, and then outputs the final confidence limits.  Prodclin is available for SAS (prodclin.sas), SPSS (prodclin.sps), and R (prodclin.r), each of which contain package specific directions for using the program.  To download the files, right click the Fortran ‘prodclin.exe' link and select “Save Target As…”, then repeat with the desired statistical package file.  Note that both files need to be saved to the same location on your local computer.

Fortran: Prodclin.exe                 

SAS: Prodclin.sas

SPSS: Prodclin.sps

R: Prodclin.r

The program ‘prodclin2.exe' is another version of the Fortran program ‘prodclin.exe' that allows the user to directly input the values for a and b, their respective standard errors, the correlation between a and b, and the Type I error rate.  This program then returns the asymmetric confidence limits.  To download the file, right click the Fortran ‘prodclin2.exe' link and select “Save Target As…”.

Fortran: prodclin2.exe

NOTE:   Due to the method the Fortran programs use to find the critical values of the distribution of the product, there are certain parameter combinations for which the program is unable to converge.  If this happens, try using fewer decimal places.”

The Prodclin program is an edited version of a Fortran program written by by Alan Miller (1997) called FNPROD. Given specific mean values of zα and zβ , along with their correlation (equal to zero for the indirect effect from Equations 2 and 3), and a value for zαzβ , FNPROD returns the cumulative percentile for that value of zαzβ using numerical integration based on an algorithm by Meeker and Escobar (1994) and work by Morris (1992). Rather than using trial and error to find the value of zαzβ for confidence limits, the FNPROD program was edited so that given values for α, ß, σα , σβ , ρ (the correlation between α and ß), and a Type I error rate, the program iteratively finds the corresponding critical values. The program was further edited so that the values of α, ß, σα , and σβ could be read in from various statistical packages and the critical values returned to the statistical software so that the asymmetric confidence limits could then be computed.

 

 

 

 

 

 

 

 

 

 

 

September 20, 2006 6:00 PM