regression interpretation

#1
Which results will I use to determine the difference in males and females when all other predictors are held constant. I have 4 predictors (status, income, verbal score and spending).
I did a lm as: fit<-lm(spending ~ sex, data=spending) and produce a result
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 29.775 5.498 5.415 2.28e-06 ***

sex -25.909 8.648 -2.996 0.00444 **

However, other students used the original LM and were correct that females spend 22.12 less than men.

Call: lm(formula = spending ~ sex + status + income + verbal, data =spending)

Residuals: Min 1Q Median 3Q Max -51.082 -11.320 -1.451 9.452 94.252 Coefficients: Estimate Std. Error t value Pr(>|t|)
(Intercept) 22.55565 17.19680 1.312 0.1968
sex -22.11833 8.21111 -2.694 0.0101 *
status 0.05223 0.28111 0.186 0.8535
income 4.96198 1.02539 4.839 1.79e-05 * verbal -2.95949 2.17215 -1.362 0.1803

Please help in figuring out why my model did not work?
 

trinker

ggplot2orBust
#2
You say:

the difference in males and females when all other predictors are held constant [and]
Code:
Call: lm(formula = spending ~ sex + status + income + verbal, data =spending)
I would think then that you're model should include all the predictors before the sex one.
Code:
Call: lm(formula = spending ~ status + income + verbal + sex, data =spending)
.

Also, if spending is your outcome variable it can't be your predictor too (at least I'm assuming you have a model predicting spending; but you're trying to predict sex, though it doesn't seem so, perhaps binary logistic regression is a better approach)