Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: eivreg and interactions

From   John Antonakis <>
Subject   Re: st: eivreg and interactions
Date   Thu, 29 Nov 2012 17:25:15 +0100

Hi Alim:

No; it doesn't. If you see the output after running the model, eivreg will assume a perfect reliability for the interaction (1.00) if you do what you suggest below.

You need manually construct the interaction and then manually specify the reliability.

For the formula of the reliability of the x*z interaction and given:

Reliability of x = relx
Reliability of z = relz
Correlation of xz =rxz

Reliability of x*z = ((relx *relz )+rxz^2)/(1+rxz^2)

For background on this formula see:

Busemeyer, J. R. and L. E. Jones (1983). "Analysis of Multiplicative Combination Rules When the Causal Variables Are Measured with Error." Psychological Bulletin 93(3): 549-562.



Prof. John Antonakis
Faculty of Business and Economics
Department of Organizational Behavior
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305

Associate Editor
The Leadership Quarterly

On 29.11.2012 00:53, 'Alim Beveridge wrote:
> hello,
> if i do the following:
> eivreg y c.x1##c.x2 , reliab(x1 .93 x2 .85)
> will stata calculate the reliability of the interaction term and use
it to adjust the estimate for its coefficient, or will it assume that the interaction term has reliability 1.0? anyone know the answer?
> thanks,
> 'Alim
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index