Running Simple Logistic Regression:
Research question: What is the relationship between pretest score and one’s passing on post-test?
- From the menu, click on Analyze -> Regression -> Binary Logistic
- In the appearance window, move DV (pass) to Dependent… -> IV (Pre) to Covariates:
- Hit Options… -> Check CI for exp(B) -> At last step -> Continue
- Hit Continue and then hit Paste.
Write-up (APA format):
Logistic regression model was performed to see whether pretest score predicts the odds of an individual’s passing on posttest. The overall model was found to be statistically significant (Chi-squared value (1) = 21.83, p < .05), with Nagelkerke R-squared value of .68. Pretest was found to be statistically significant in predicting one’s odds of passing (chi-squared value (1) = 9.50, p = .002). In particular, the odds of one’s passing will be increased by 10% for every additional increase in pretest score (OR = 1.04 – 1.17).
Running Multiple Logistic Regression:
RQ: whether reading (bytxrstd), SES (f1ses), and student morale (f1stumor) are significant predictor of students’ likelihood of passing math exam?
- From the menu, click on Analyze -> Regression -> Binary Logistic…
- In the appearance window, move DV (passmath) to Dependent… -> IV (bytxrsd, f1ses, f1stumor) to Covariates:
- Hit Options… -> Check CI for exp(B) -> At last step -> Continue
- Hit Continue and then hit Paste.
Write-up (APA Format):
Logistic regression model was performed to see whether ses, pretest score, and student moral predict the odds of an individual’s passing on math. The overall model was found to be statistically significant (Chi-squared value (3) = 94.92, p < .05), with Nagelkerke R-squared value of .44. SES (chi-squared (1) = 8.74, p = .003) and pretest score (chi-squared (1) = 34.04, p < .05) were found to be statistically significant in predicting the odds of an individual’s passing on math. Specifically, the odds of an individual’s passing on math will be on average increased by 129% for every additional increase in SES, after controlling for student moral and pretest score (OR = 2.29, 95% CI = 1.32 – 3.97). And, the odds of an individual’s passing on math will be on average increased by 115% for every additional increase in pretest, after controlling for student moral and pretest score (OR = 1.15, 95% CI = 1.10 – 1.21).