0
$\begingroup$

An example to illustrate the question I have:

Suppose Strength is a function of gender (male/female) and, AMONG MALES, testosterone level (low, medium, or high). I don't care about testosterone levels in females, and those cells are blank for females in the data set.

If I were doing this through dummy coding, I would code this using four dummy variables: female, MaleLow, MaleMedium, MaleHigh.

strength = B1(female) + B2(MaleLow) + B3(MaleMed) + B4(MaleHigh)

To ask if sex matters, I would test mean strength of females vs weighted average of the 3 male categories.

To test if testosterone level matters, I want to test the null hypothesis that the mean strength in the 3 male categories is equal.

Can I do this using GLM without having to manually dummy code everything? if I put the categorical variable testosterone in the fixed factors, I am not sure how SPSS will handle it because this variable is only relevant to the male subset, and females all have missing values. How do I ask SPSS to test the specific hypothesis I am asking (if I'm correct, does B2=B3=B4, and does B1 = weighted average of B2,B3,B4)

The actual problem I have is a little more complex, in that the "females" have their own subcategories that are irrelevant to males, and there are additional categorical and continuous variables.

$\endgroup$
  • $\begingroup$ For clarity, I don't think I can use the built-in ANOVA function with its contrasts in SPSS, because the actual model I'm fitting has multiple categorical variables and multiple continuous ones. In addition, one of the categoricals, which can take on two values, I would like to break into further subcategories. That's why I am asking how to test these hypotheses using GLM. $\endgroup$ – Sadashiv Santosh Feb 6 '16 at 5:05
0
$\begingroup$

It seems that the data for females has no information about your hypotheses, so why not just limit the dataset to males? Since you have other unspecified variables, you might get something from pooling, but you would have to assume that gender does not matter for them or in effect you are separating the genders into different samples anyway.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.