1 Executive Summary

  • The aim of this report was to investigate the relationship between the academic performance and healthy habits of university students.
  • Data collected shows statistically weak correlation between the health habits students partake in and their academic performance.
  • Student who exhibited less healthy habits did not greatly differ from those who exhibited more.



2 Full Report

2.1 Initial Data Analysis (IDA)

2.1.1 Source

Data was collected from university students, excluding first year students as Weighted Average Mark (WAM) was required. A Google Forms survey was used to compile the data : (https://forms.gle/f61y3m4gWvit2cte8)

2.1.2 Structure

36 responses were received from the survey. The responses are represented by the rows while the variables are represented by the columns. Variables received include quantitative data (exercise,WAM,leisure) and qualitative (content engagement, attendance, stress levels, workload, recorded/live lecture).

2.1.3 Limitations

Individuals may have potentially lied reagrding their WAM, despite the survey anonymity. Individuals may have also under or overestimated their stress levels since it is objective on how stressed they are. Individuals may not be able to objectively assess where they are on each gradient from 1-10.

2.1.4 Assumptions

It was assumed that respondents would be able to accurately self-assess their participation in the questionnaire. The questionnaire included questions about study habits, time and frequency of exercise, and feelings of stress. It is assumed that all respondents will have the same perception and evaluation of the scales 1-10.

2.1.5 Data Cleaning

Absurd responses were excluded as well as any outliers in order to preserve the assumption that the data is normally distributed.

Project2 <- read_csv("Survey Responses.csv",show_col_types = FALSE) 


Data = rename(Project2, 
       Engagement = `From 1-10 with 1 being never and 10 being everyday, how often do you engage with your weekly readings and course content outside of lectures?`, 
       Exercise = `Do you engage in exercise weekly? If so, how many hours per week do you exercise?  (please answer in whole numbers)`, 
       WAM = `What is your most recent WAM?`, 
       LModes = `Do you prefer attending live lectures or watching the recorded lectures?`,
       Attendance = `From 1-10 with 1 being never and 10 being everyday, how often do you attend your lectures?`, 
       Stress = `From 1 - 10 with 1 being not stressed at all to 10 being extremely stressed, what are your daily stress levels at this point in the semester?`, 
       Leisure = `How many hours do you usually have of free time per week? (please answer in whole numbers)` )

Data$Exercise = as.integer(as.character(Data$Exercise))
data = filter(Data, Leisure <= 0)
Final = drop_na(Data)

2.2 Research Theme

An Investigation in the Relationship between Academic Performance and Healthy Habits of University Students.

2.2.1 Research Question 1

Does a student’s Weighted Average Mark (WAM) correlate to their stress levels?

Final <- read_csv("FilteredData.csv")
scatter = ggplot(Final, aes(x=Stress, y=WAM)) + geom_point(size=2) + geom_smooth(method=lm , color="blue", se=FALSE) + labs(title = "Relationship between Individual Stress Levels and Academic Performance", x = "Level of Stress", y = "WAM") + scale_x_continuous(n.breaks=8) 

corr = round(cor(Final$Stress, Final$WAM),3)

scatter + geom_text(x=3.7, y =86, label = paste0('r = ', corr), colour="red") 

model <- lm(WAM~Stress, Final)
ggplot(model, aes(x = .fitted, y = .resid)) +
  geom_point() +
  geom_hline(yintercept = 0) + labs(title = "The Residual Graph for WAM vs Stress Levels", y = "Residuals (Marks)", x = "Level of Stress" )

Correlation: r= -0.198 ( Weak Negative Correlation)

Although a weak correlation is present, it can still be considered that students who have a higher WAM tend to have lower stress levels. This reflects the ability of students who achieved higher WAMs to manage their stress well. Additionally, the residue plot appears to be random and supports the idea that the linear fit is appropriate for this data.

2.2.2 Research Question 2

What is the relationship between the students’ current WAM and their degree of content engagement?

scatter = ggplot(Final, aes(x=Engagement, y=WAM)) + geom_point(size=2) + geom_smooth(method=lm , color="blue", se=FALSE) + labs(title = "Relationship between Content Engagement and Academic Performance", x = "Level of Content Engagement", y = "WAM") + scale_x_continuous(n.breaks=8) 

corr = round(cor(Final$Engagement, Final$WAM),3)


scatter + geom_text(x=2, y =87, label = paste0('r = ', corr), colour="red") 

model <- lm(WAM~Engagement, Final)
ggplot(model, aes(x = .fitted, y = .resid)) +
  geom_point() +
  geom_hline(yintercept = 0) + labs(title = "The Residual Graph for WAM vs Content Engagement", y = "Residuals (Marks)", x = "Level of Content Engagement")

Correlation: r= 0.284 (Weak Positive Correlation)

Although a weak correlation is present, it can still be considered that students who have a higher WAM tend to have higher levels of content engagement. This reflects the student’s commitment to revision of course material outside of academic hours. Additionally, the residue plot appears to be random and supports the idea that the linear fit is appropriate for this data. Additionally, the residue plot appears to be random and supports the idea that the linear fit is appropriate for this data.

ggplot(Final, aes(x = LModes, y =WAM, fill = LModes)) + labs(title = "Lecture Mode of Attendance vs WAM", y = "WAM", x = "Modes of Attendance") + geom_boxplot(outlier.colour="red", alpha=0.7) + stat_boxplot(geom ='errorbar', width = 0.6) +
    stat_summary(fun.y=mean, geom="point", shape=20, size=4, color="black", fill="black") +
    theme(legend.position="none" ) + coord_flip()

We can observe from the comparative box plot that students attending live-lectures, although only slightly, are seen to perform better academically.

2.2.3 Articles

Studies exploring the relationship between healthy habits and academic performance seem to compliment the data collected; with the Kötter Et al.(2017) study, higher perceived stress levels were associated with poorer academic performance, while the Edwards Et al. (2019) study has shown that students who attend live lectures capture a greater amount of content as opposed to students who watch the lecture recordings.

3 References

Style: APA

Edwards, M.R., Clinton, M.E. A study exploring the impact of lecture capture availability and lecture capture usage on student attendance and attainment. High Educ 77, 403–421 (2019). https://doi.org/10.1007/s10734-018-0275-9

Kötter, T., Wagner, J., Brüheim, L. et al. Perceived Medical School stress of undergraduate medical students predicts academic performance: an observational study. BMC Med Educ 17, 256 (2017). https://doi.org/10.1186/s12909-017-1091-0

4 Acknowledgements

List of meetings and contributions

01/03/24 - Zoom meeting, Alicia, Marjan, Nicole (Brainstorm Research Themes and Questions)

06/03/24 - Fisher Library, Marjan, Alicia, Lyra, Nicole, (Further Developed Ideas for Research Themes and Questions)

08/03/24 - Fisher and online, Marjan, Alicia,Nicole, (Reviewed Research Theme, Questions and Variables)

16/04/24 - Zoom Meeting, Marjan, Alicia, Lyra, Nicole, Roy, Reviewing the data collected from the survey and assigned roles for Project 2

18/04/24 - During lab, all in attendance, graphs were finalized and everyone worked on their assigned parts to put together the final product