Scribd
Upload
3 views3 pages

Project2_Aumiller

The document presents the results of three statistical tests: a Chi-Squared Goodness of Fit Test, ANOVA, and a T-test. The Chi-Squared test found no significant relationship between a student's sex and failures, while ANOVA indicated that both absences and failures significantly affect G3 grades independently. The T-test revealed that students who failed spent significantly less time studying compared to those who did not, leading to the rejection of the null hypothesis regarding study time differences.

Uploaded by

Emily Aumiller
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views3 pages

Project2_Aumiller

The document presents the results of three statistical tests: a Chi-Squared Goodness of Fit Test, ANOVA, and a T-test. The Chi-Squared test found no significant relationship between a student's sex and failures, while ANOVA indicated that both absences and failures significantly affect G3 grades independently. The T-test revealed that students who failed spent significantly less time studying compared to those who did not, leading to the rejection of the null hypothesis regarding study time differences.

Uploaded by

Emily Aumiller
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Emily Aumiller

INST314

1.​ Chi-Squared Goodness of Fit Test

Research Hypothesis: There is a relationship between a student’s sex and the number of failures
they have.
Null hypothesis: There is no relationship between a student’s sex and the number of failures.

Alpha level: 0.05

Rejection: We will reject the null hypothesis if the p-value is less than 0.05

Assumptions: The variables sex and failures are assumed to be independent.

Conclusion:

●​ The chi-squared test gave a test statistic of 2.0934 , 2 degrees of freedom, and a p-value
of 0.3511.
●​ The p-value is greater than 0.05, so we fail to reject the null hypothesis. We cannot
conclude there’s a meaningful relationship between sex and failures.

The differences in failure rates between male and female students could just be random. There
isn't any significant data that says otherwise.

2.​ ANOVA

Research Hypothesis: Absences and failures significantly affect G3 grades independently.

Null hypothesis: Absences and failures do not significantly affect G3 grades independently.

Alpha level: 0.05

Rejection: We will reject the null hypothesis if the p-value is less than 0.05 for any term.

Assumptions: Observations should be independent of each other. One student’s grades should not
influence another’s. The G3 grades should have similar levels of variation across the different
groups for absences and failures.

Conclusion:
Since both p values were less than .05 we reject the null and say that absences and failures
significantly affect grades.

Additive model results:


There is a significant effect of both absences and failures on grades. Both independently impact
grades. Students with higher absences or failures tend to have lower grades.
-​ Absences: F(31,347)=2.156, p=0.000493 (significant).
-​ Failures: F(3,347)=23.624, p=6.17×10−14 (significant)

Interactive Model results:


While absences and failures affect grades individually, their interaction is not significant. This
means the way absences affect grades doesn’t depend on one another.
-​ Absences: F(31,324)=2.165, p=0.00049 (significant).
-​ Failures: F(3,324)=23.715, p=6.69×10−14 (significant).
-​ Interaction: F(23,324)=1.059F, p=0.39119p = 0.39119p=0.39119 (not significant).

Post Hoc Analysis:

Absences: Notable differences between specific levels of absences. Students with no absences
vs. those with 2, 4, or 6 absences had significantly different grades.

Failures: Notable differences between failure levels. Students with no failures vs. those with 1,
2, or 3 failures had significantly different grades.

3.​ T-test

Research Hypothesis: The mean study time is the same for students with and without failures.

Null hypothesis: The mean study time is different between the two groups.

Alpha level: 0.05

Rejection results: We will reject the null hypothesis if the p-value from the test is less than 0.05

Assumption: The data in each group should be normally distributed. The two groups, failures and
no failures, must be independent, one student’s study time does not influence another student’s
study time.
Conclusion:
-​ Students who failed classes spent significantly less time studying compared to students
who didn’t fail. The average study time for students who failed was 1.7 hours, while for
those who didn’t fail, it was 2.1 hours.
-​ The t-test showed that this difference is statistically significant (t=−4.122, p<0.05),
meaning the difference is unlikely to be due to random chance. The confidence interval
for the difference in study time is between -0.60 and -0.21 hours, confirming the results.
-​ We reject the null hypothesis. The p-value is very small which means there is strong
evidence supporting that the average study time is different between the groups failing
and not failing.

You might also like