Thanks so much for providing this useful information. It is very difficult to find any scoring information on the VIA on the web so your hard work is appreciated by many of us. However we are still struggling to find a scoring key for the VIA 120 – that is the “short form”. Have spent hours looking on the web for a scoring key but our search has been fruitless. Can only find the original 240 and the VIA-Youth. Any direction you can provide would be greatly appreciated! My email address has been provided.

]]>https://startpage.com/do/mypage.pl?prf=c13344f516a5c0925e6339d5a8dedc26&query=

]]>Is there a way to transport search settings (E.g. Use anonymous picture search, do not filter video and results)?

e.g. this link:

https://startpage.com/do/mypage.pl?prf=c13344f516a5c0925e6339d5a8dedc26

Thanks again,

Andi

There is a far more important error in my actual post; one that I will correct shortly. The error relates to the rescaling of the variables after they have been generated. The rescaling will mess up the covariances! Instead of using the Cholesky decomposition of the correlation matrix, one should use the decomposition of the covariance matrix.

]]>I have added some trivial R-code below which is easily modified. I was interested in data in which variables V1 and V2 are highly correlated (r=.9) and V3 and V4 are highly correlated (r=.9), and the remaining correlations are low (r=0)

############### START

r <- matrix( # Correlation matrix

c(1.0, 0.9, 0.0, 0.0,

0.9, 1.0, 0.0, 0.0,

0.0, 0.0, 1.0, 0.9,

0.0 ,0.0, 0.9, 1.0)

, nrow=4)

c <- chol(r) # Choleski decomposition of the correlation matrix

w <- matrix(rnorm(800), ncol=4) # Create a datamatrix of random normal deviates with the appropriate number of cases

correlated.data <- w %*% c

############### END

]]>