bootstrapping with moderator

[d.jung]

dear all,

i am currently testing the moderating effect of a dummy (company size 1=large, 0=small) on the relationships between 10 independent and 4 dependent variables. unfortunately, when i run the bootstrapping procedure i always get the "singular matrix error" after a couple of samples.

i set the # of cases to 67 (= n) and the # of samples to 500 as suggested in other posts.

does anybody have an idea how to fix this? i searched the whole forum already but still do not know how to fix it.. do i perhaps have to choose sth with regard to "sign changes"?

best regards,
david

[jmbecker]

This has possibly the same reason as described in here: viewtopic.php?t=664