STD and UnSTD betas

31 views

Skip to first unread message

Kape

unread,

Oct 31, 2018, 4:50:31 AM10/31/18

to lavaan

Hi!

I have conducted a CLPM and a RI-CLPM models in R using Lavaan. The models seem to work fine (goodness of fit indices, results) although I noticed that the unstandardized and standardized estimates look pretty similar (both in the CLPM and RI). I am using mean scales in the model (scaling 1-3), and to me, the data looks fine.

I have pasted a copy of the results from the CLPM below. Basically, what I'm wondering is whether the similarity in the betas could be a sign of a problem or not?

Best regards,

Kape

Regressions:
                   Estimate Std.Err z-value P(>|z|)   Std.lv Std.all
T2Dis ~
    T1Dis      (a)    0.424    0.022   19.244    0.000    0.424    0.460
T3Dis ~
    T2Dis      (a)    0.424    0.022   19.244    0.000    0.424    0.445
T2Sol ~
    T1Sol      (b)    0.405    0.023   17.814    0.000    0.405    0.425
T3Sol ~
    T2Sol      (b)    0.405    0.023   17.814    0.000    0.405    0.426
T2Con ~
    T1Con      (c)    0.456    0.022   20.485    0.000    0.456    0.435
T3Con ~
    T2Con      (c)    0.456    0.022   20.485    0.000    0.456    0.460
T2Sol ~
    T1Dis      (d)    0.157    0.023    6.827    0.000    0.157    0.160
T3Sol ~
    T2Dis      (d)    0.157    0.023    6.827    0.000    0.157    0.155
T2Con ~
    T1Dis      (e)    0.055    0.033    1.661    0.097    0.055    0.045
T3Con ~
    T2Dis      (e)    0.055    0.033    1.661    0.097    0.055    0.042
T2Del ~
    T1Dis      (f)   -0.048    0.012   -4.213    0.000   -0.048   -0.124
T3Del ~
    T2Dis      (f)   -0.048    0.012   -4.213    0.000   -0.048   -0.110
T2Del ~
    T1Del      (g)    0.614    0.054   11.300    0.000    0.614    0.439

Terrence Jorgensen

unread,

Nov 2, 2018, 9:28:35 AM11/2/18

to lavaan

Basically, what I'm wondering is whether the similarity in the betas could be a sign of a problem or not?

Standardized estimates are what you would get if the variables were z scores. If your observed variances are close to 1, then you can expect standardized and unstandardized estimates to be similar.

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam