alternatives to oneway

[Alan Jeddi]

Hello,
I have a variable go_score, and I would like to test whether there is a trend (ie, whether it is higher or lower with increasing age, when age is split into intervals of 5 years).

I conducted a oneway (oneway goscore age_5int) but I believe this test assesses the null hypothesis that there is no difference between groups.

I would like to instead test for trend. I don't want to do a linear regression at this point; are there any other ways of going about doing this? Some type of chi square test of trend? I dont know the test for this; can anybody help?

Thanks

[Igor Paploski]

There is a good text about it here.

[Carlo Lazzaro]

Alan:
by categorizing age and avoiding -regress- you miss the chance of investigating whether a quadratic relationship exists between -c.age##c.age- and -goscore- when adjusted for the remaining predictors.
Please also note that, in general, dichotomizing a continuos predictor in a regression model is not recommended (https://www.ncbi.nlm.nih.gov/pubmed/16217841).