Hypothesis: Those who trust politicians are more likely to vote.

weight off.
fre vote.
RECODE vote (1=1) (2=0) INTO vote_dummy.
VARIABLE LABELS vote_dummy ‘voted last election dummy’.
VALUE LABELS vote_dummy 0’no’ 1’yes’.
fre vote vote_dummy.

fre trstplt.
RECODE trstplt (0 thru 5=0) (6 thru 10=1) INTO trstplt_dummy.
VARIABLE LABELS trstplt_dummy ‘Trust in politicians dummy variable’.
VALUE LABELS trstplt_dummy 1’trust’ 0’no trust’.
fre trstplt trstplt_dummy.

LOGISTIC REGRESSION vote_dummy WITH trstplt_dummy.

*If you want to create the graphs then you can get the data with the following command:.
CROSSTABS vote_dummy BY trstplt_dummy /CELLS=COLUMN.

1-voted
0-did not vote
1-trust
0-no trust

b1: The log odds of voting among those who trust politicians are by 1.599 higher than among those who do not trust politicians.
b0: The log odds for voting among those who do not trust politicians are 0.854.
b0+b1: The log odds of voting among those who trust politicians is (0.854+1.599).
Exp(b1): The odds of voting among those who trust politicians are 4.947 times as high as among those who do not trust politicians.
Exp(b0): The odds for voting among those who do not trust politicians are 2.349.
Exp(b0)*Exp(b1): The odds of voting among those who trust politicians are (4.947*2.349).

Conclusion: The results support the hypothesis stating that “Those who trust politicians are more likely to vote” since the p-value is smaller than 0,05 and the odds for voting among those who trust politicians are higher than among those who do not trust politicians.