Hypothesis: Women are more likely to be discriminated based on nationality than men.

weights off.
FREQUENCIES gndr.
RECODE gndr (1=0) (2=1) INTO gender_dummy.
VARIABLE LABELS gender_dummy ‘gender dummy’.
VALUE LABELS gender_dummy 0 ‘male’ 1 ‘female’.
FREQUENCIES gndr gender_dummy.

LOGISTIC REGRESSION dscrntn WITH gender_dummy.

1-marked
0-not marked
0-male
1-female


b0+b1: The log odds of being discriminated based on nationality for females is (b0+b1=-3,853-(-0,121)=-3,974).

b0: The log odds of being discriminated based on nationality for males is -3.853. (The effect of being a woman on discrimination based on nationality is negative because Exp(b1)=0.886.)

Exp(b1): The odds of being discriminated based on nationality among women are 0,886 times as low as among men.

Exp(b0): The odds of being discriminated based on nationality among men are 0,021.

Conclusion: Since the significance is 0.731 which is higher than 0.05 the difference that we see is statistically not significant. Thus, we did not find any evidence that would support our hypothesis.

Note: In your own research once you see that the results are statistically not significant you do not have to analyze the odds ratio or other measures because all the difference that these measures show is appearing probably just by chance.

UP