Package lm.beta misscalculations?

General
1

Hi there,

I'm using package lm.beta to calculate the standardized beta regression coefficents for a multiple regression. For a interaction term I get a beta of -1.79587 (last line), which feels odd, shouldn't beta coeffs all be below 1?

I'm using this model:

hpo.modeldef4a <- 'HPO_Gesamt ~ HPR_Gesamt_z*SD.german + SD.math + SR3.theta'

This is the complete output:


Call:
lm(formula = hpo.modeldef4a, data = dta2)

Residuals:
     Min       1Q   Median       3Q      Max 
-17.8455  -3.6578   0.7294   4.0519  11.7470 

Coefficients:
                       Estimate Standardized Std. Error t value Pr(>|t|)   
(Intercept)            19.94715      0.00000    7.06051   2.825  0.00592 **
HPR_Gesamt_z            2.27562      2.27136    0.69137   3.291  0.00146 **
SD.german              -0.51768     -0.03753    1.52376  -0.340  0.73491   
SD.math                 1.34316      0.13193    1.18694   1.132  0.26105   
SR3.theta               1.90916      0.21542    0.90289   2.114  0.03747 * 
HPR_Gesamt_z:SD.german -0.37094     -1.79587    0.13996  -2.650  0.00963 **
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 5.593 on 83 degrees of freedom
Multiple R-squared:  0.537,	Adjusted R-squared:  0.5091 
F-statistic: 19.25 on 5 and 83 DF,  p-value: 1.131e-12

Thanks for your help!

Regards, Martin

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.