List at least 1 external reference other than your data source.
Future directions are good, but your critique of the analysis needs to be more extensive - what is the problem with being based only on confidence intervals and p-values? You can tie this in to the future directions for improvement but make sure it's clear what the limitations are of the current analysis before you talk about how you'll improve it!
Specific points:
For the purpose of our analysis, we only made use of 4 columns, a binary variable for whether a student is in a romantic relationship and 3 columns for grades throughout the year (on a scale of 0-20) which were used as totals from both data sets (for Maths and Portugese).
This is confusing - are you saying you have grades for Math, Portuguese, and Total (Math + Portuguese)?
We notice that there is only a slight overlap between the CIs for both the cases, which led us to suspect that there may not be significant difference between the average total grades of students in a relationship and those who are not.
Otherwise this suggests that a larger overlap would mean a more significant difference
The p-values for both the tests are less than 0.05, our criteria for level of significance, which implies that the we have enough evidence to reject the null hypothesis and that the results are statiscally significant.
List at least 1 external reference other than your data source.
Future directions are good, but your critique of the analysis needs to be more extensive - what is the problem with being based only on confidence intervals and p-values? You can tie this in to the future directions for improvement but make sure it's clear what the limitations are of the current analysis before you talk about how you'll improve it!
Specific points:
This is confusing - are you saying you have grades for Math, Portuguese, and Total (Math + Portuguese)?
Otherwise this suggests that a larger overlap would mean a more significant difference
Tie this back to the research question.