Executive Summary

This report investigated whether study time differs between DATA1001 and DATA1901 students and whether commute time is associated with weekly study hours. DATA1901 students studied slightly more per week, while commute time had a very weak negative association with study hours. These findings suggest that course level may be more relevant to study time than commute time, although other factors may contribute.

EDA (Max 200 words)

Structure
Data cleaning

Overview

Source:

This analysis draws on a comprehensive dataset containing 29 variables, collected from a survey of 5,057 DATA1X01 students. To address our specific research goals, we isolated the following three metrics:

mainstream_advanced – Qualitative (nominal) hours_studying – Quantitative (continuous) commute – Quantitative (continuous)

Data cleaning

Code

library(tidyverse)
survey_data <- read.csv("DATA1001_Project_1.csv")
#df2 <- df[ , c("mainstream_advanced", "hours_studying", "commute") ]
#rite.csv(df2, "filtered_data1001_project_1.csv", row.names = FALSE)

Limitations

Participation was voluntary, meaning the dataset represents a sample, not the entire population. The self-reported survey introduces response bias, as students may over-report study time or inaccurately calculate commute time. Commute time also varies daily due to traffic, transport mode, and class hours, meaning reported figures may not reflect a consistent average. Lastly, confounding variables like part-time work and course load were not considered.

Assumptions

We assumed respondents provided honest and reasonable estimates for both study and commute times. We also assumed “hours studying” excluded mandatory contact hours such as lectures and tutorials.

Research Question 1

Does the distribution of weekly study hours differ between mainstream_advanced

Code

library(tidyverse)
library(ggplot2)
ggplot(survey_data, aes(x = mainstream_advanced, y = hours_studying, fill = mainstream_advanced)) +
  geom_boxplot(width = 0.6, alpha = 0.8) +
  labs(x = "Unit", y = "Hours Studying (hours/week)", fill = "Unit") +
  theme_minimal() +
  theme(axis.text.x = element_text(size = 12),
        legend.position = "none")

The boxplot shows that advance course-DATA1901 students in tend to spend slightly more time studying each week than DATA1001 students, with a higher median and a slightly wider IQR. One possible reason is that advanced courses usually involve more complexity concepts and problem need to solve, so students may need to spend more time studying.

However, course difficulty is not the only factor. the advance course students may have stronger prior knowledge, higher confidence, or greater interest in data science, which could make them more motivated to study. Research in STEM education found that autonomous motivation was linked to more study time and higher final grades (Botnaru et al. 2021).

Overall, the difference may come from a mix of course demands, confidence, and motivation. However, the boxplot alone cannot show that course level directly causes students to study more.

Research Question 2

Is there a linear correlation between commute time and weekly study hours among DATA1001/1901 students?

Linear Model Plot

Code

ggplot(survey_data, aes(x = commute, y = hours_studying)) +
  geom_point(color = "blue", size = 0.5) +
  geom_smooth(
    method = "lm",
    se = FALSE,
    formula = y ~ x,
    color = "red"
  ) +
  labs(
    title = "Linear correlation between commute time and weekly study hours among DATA1X01 students.",
    x = "Commute Time (minutes)",
    y = "Hours Studying (Per week)"
  )

Residual Plot

Code

model <- lm(hours_studying ~ commute, data = survey_data)

r <- cor(survey_data$commute, survey_data$hours_studying)
r2 <- summary(model)$r.squared

ggplot(model, aes(x = .fitted, y = .resid)) +
  geom_point(alpha = 0.6, color = "blue", size = 0.8) +
  geom_hline(
    yintercept = 0,
    linetype = "dashed",
    color = "red"
  ) +
  labs(
    title = "Residual Plot",
    x = "Fitted Values",
    y = "Residuals"
  )

Warning: `fortify(<lm>)` was deprecated in ggplot2 4.0.0.
ℹ Please use `broom::augment(<lm>)` instead.
ℹ The deprecated feature was likely used in the ggplot2 package.
  Please report the issue at <https://github.com/tidyverse/ggplot2/issues>.

Our analysis demonstrates a very weak negative correlation between commute time and weekly study hours among DATA1X01 students. The scatter plot shows no clear linear pattern, with study hours remaining widely dispersed as commute time increases. The correlation coefficient is approximately (r = -0.057), confirming that the linear relationship between the two variables is extremely weak.

The residual plot shows clustering and uneven variance rather than a random distribution around zero, suggesting that the assumptions of the linear model are not well satisfied. The model has an (R^2) of approximately 0.0032, meaning that commute time explains only about 0.32% of the variation in weekly study hours. Therefore, commute time alone does not appear to be a useful predictor of students’ intended weekly study hours.

Articles (Max 50 words)

RQ1: (Botnaru et al. 2021) Predicting final grades in STEM courses: A path analysis of academic motivation and course-related behavior using self-determination theory https://www.sciencedirect.com/science/article/abs/pii/S0023969021000175

Acknowledgements

Meetings and Workshop Minutes

The Acknowledgment section includes a list of group meetings (date and time and attendance), the contribution of each group member, and all resources used (e.g. url of stack overflow, url of Ed post, date and details of drop-in session with tutor).

AI Usage Statement

An AI usage statement including: the name and version of the AI tools, the publisher, the URL, a brief description of the context in which the tool was used. If no AI was used, this must be clearly stated.

Professional Standard of Report

A short statement explaining how 1 of the Shared Values and 1 of the Ethical Principles has been adhered to.