Ifyou want to do this by hand you can find the formula for a Chow test in statistics books. To do it in SPSS, create a dummy variable for the group (0 or 1), then compute new variables as the product of the dummy and each of the independent variables. The significance tests for the coefficients of these product variables gives you precisely the tests you want. - David Greenberg, Sociology Department, New York University, New York, NY On Mon, 30 Nov 1998, David W. Wright wrote:

> Running multiple regression with a saturated model of x factors; the sample > is broken into two groups & the saturated model ran again separately for > each group. What is the formula for determining whether a statitsically > significant difference exist between the unstandardized betas across the > two groups on factor x? For example, grp1 x=.123 (stderr.456) & grp2 > x=.789 (stderr.012). The formula needs to take into account the different > sample sizes across the two groups. > > > > David W. Wright, Ph.D.. > Graduate Coordinator > Wichita State University > Dept. of Sociology, Box 25 > Wichita, Kansas 67260-0025 > > Office phone: 316-978-3280 > Fax: 316-978-3234 > Internet: DWRIGHT@TWSUVM.UC.TWSU.EDU >