Hello,

What are levels of measurement of your variables? It may to high extend predetermine analyses.

Regards

You then have six different univariate models. Please be aware that the more equations you examine the greater your chance of finding a significant outcome even if the null hypothesis is true. In this case there is a 0.735 chance of finding no significant outcomes if the null hypothesis is true. One significant outcome will happen about 20% of the time, two significant outcomes about 3% and three significant outcomes are 0.2%. This assumes alpha=0.05, and the 6 dependent variables are independent of each other. The probability of finding no significant outcomes given that the null hypothesis is true will diminish as the number of tests increases. 

Download my papers from RG.

I analyze all posible combination models by eight LDFs.