MTH 231: Exploratory Data Analysis - Project 1
Understanding Student Performance
Group Members:
Name:
Name:
Name:
Name:
Due Date: Wednesday, October 9 - IN CLASS
1
Background
This project involves conducting univariate analysis on a dataset that pro-
vides a comprehensive overview of various factors affecting student perfor-
mance in exams. This analysis will help us understand individual variables
in isolation and explore how they contribute to academic success.
About the Dataset
The dataset includes a wide range of variables from study habits and socio-
economic factors to personal characteristics and environmental influences.
These variables are a mix of numerical and categorical types, offering a
rich field for analysis and visualization.
• Hours Studied - Number of hours spent studying per week.
• Attendance - Percentage of classes attended.
• Parental Involvement - Level of parental involvement in the stu-
dent’s education (Low, Medium, High).
• Access to Resources - Availability of educational resources (Low,
Medium, High).
• Extracurricular Activities - Participation in extracurricular activ-
ities (Yes, No).
• Sleep Hours - Average number of hours of sleep per night.
• Previous Scores - Scores from previous exams.
• Motivation Level - Student’s level of motivation (Low, Medium,
High).
• Internet Access - Availability of internet access (Yes, No).
• Tutoring Sessions - Number of tutoring sessions attended per month.
• Family Income - Family income level (Low, Medium, High).
• Teacher Quality - Quality of the teachers (Low, Medium, High).
• School Type - Type of school attended (Public, Private).
• Peer Influence - Influence of peers on academic performance (Posi-
tive, Neutral, Negative).
2
• Physical Activity - Average number of hours of physical activity
per week.
• Learning Disabilities - Presence of learning disabilities (Yes, No).
• Parental Education Level - Highest education level of parents (High
School, College, Postgraduate).
• Distance from Home - Distance from home to school (Near, Mod-
erate, Far).
• Gender - Gender of the student (Male, Female).
• Exam Score - Final exam score.
These variables cover a broad range of factors that could influence a stu-
dent’s academic performance, providing a rich dataset for exploratory data
analysis. The dataset contains a total of 6,607 records, which represent
individual student entries.
General Notes
To effectively communicate your findings through graphs, ensure the fol-
lowing in your graphical displays:
• Label your graphs with descriptive titles to clearly indicate what is
being shown.
• Axes labels should be clear and precise, stating what each axis repre-
sents.
• Scales on axes should be set to meaningful and useful ranges to en-
hance readability and interpretation.
• Consistency in visual elements like color, scale intervals, and legend
placement should be maintained across similar types of graphs for
easy comparison.
Additionally, all projects should be typed using LaTeX or Word. Include
all graphs generated from R within your document. This will not only
help in presenting your findings more professionally but will also ensure
that all visual data supports your textual analysis cohesively.
3
Study Proposal: Investigating Individual Factors Af-
fecting Academic Performance
Study Objective
The objective of this study is to examine how individual lifestyle and de-
mographic factors influence academic performance. Students will use uni-
variate analysis techniques to explore and visualize data, gaining insights
into how single variables may correlate with students’ exam scores.
Research Question
Which individual factors correlate with students’ academic performance?
This question encourages students to investigate how specific, single vari-
ables (such as hours studied or sleep hours) relate individually to academic
outcomes. The goal is to understand the unique contribution of each fac-
tor to student success, without comparing or combining them with other
variables. This aligns with the capabilities of univariate analysis, where
the focus is on one variable at a time.
Suggested Variables for Analysis
Quantitative Variables
Students can choose to analyze variables such as:
• Hours Studied
• Sleep Hours
• Previous Scores
• Physical Activity (transform into a numerical variable if initially cat-
egorical, e.g., hours per week)
Categorical Variables
Students can analyze variables like:
• Parental Involvement
• Internet Access
• School Type
4
• Gender
• Extracurricular Activities
Data Analysis Tasks
Histograms
Task: Create histograms to visualize the distribution of selected quanti-
tative variables like Hours Studied and Sleep Hours.
Description: Analyze the shape, center, and spread of the distributions.
Discuss potential implications of findings for student performance.
Comparative Histograms
Task: Compare histograms for a quantitative variable across two cate-
gories of a demographic variable (e.g., Hours Studied for Male vs. Female
students).
Description: Evaluate differences in the distributions and infer potential
disparities in study habits or resources.
Bar Charts
Task: Use bar charts to visualize the frequency of categorical variables
such as School Type and Internet Access.
Description: Discuss how access to resources like the internet might
correlate with academic outcomes.
Comparative Bar Charts
Task: Create comparative bar charts to show differences in a categorical
variable like Extracurricular Activities across two categories of another
variable (e.g., comparing participation in activities between Public and
Private school students).
Description: Infer how extracurricular engagement could impact aca-
demic performance.
5
Expected Outcomes
• Insights: Students should aim to identify patterns and trends that
suggest correlations between individual factors and academic perfor-
mance.
• Understanding Data: Students will deepen their understanding of
how univariate analysis can provide initial insights into data sets and
prepare them for more complex analyses in the future.
Rubric
Introduction (10%)
• Objective: Begin with a clear statement of the research objective or
question.
• Variable Selection: Describe which variables were chosen for analy-
sis and explain the rationale behind these choices. Discuss why these
variables are relevant to the study and what you hypothesize their
impact might be on academic performance.
Analysis (40%)
• Objective: Conduct detailed univariate analysis using the chosen
variables.
• Required Elements:
– Histograms: Create histograms for each chosen quantitative
variable to visualize the distribution.
– Comparative Histograms: Use comparative histograms to show
differences in distributions across categories (e.g., by gender or
school type).
– Bar Charts: Construct bar charts for categorical data to display
the frequency of each category.
– Comparative Bar Charts: Develop comparative bar charts to
compare categorical data across different groups.
– Descriptive Statistics: Calculate and interpret mean, median,
mode, range, and standard deviation for each chosen quantitative
variable.
6
Synthesizing Findings (15%)
• Objective: Summarize the key findings from the analysis.
• Details: Review and highlight significant patterns, trends, and anoma-
lies observed in the data. Provide a concise synthesis that ties these
observations back to the study’s initial hypotheses or questions.
Reflecting on Implications (15%)
• Objective: Discuss the broader implications of your findings.
• Details: Consider how the results might apply in real-world contexts
or influence educational practices. Reflect on how the data might
inform student behaviors, teaching strategies, or policy decisions.
Considering Limitations (15%)
• Objective: Critically evaluate the limitations of your study.
• Details: Acknowledge any constraints related to sample size, data
collection methods, or potential biases. Discuss how these limitations
could impact the reliability or generalizability of your conclusions.
Future Directions (5%)
• Objective: Suggest areas for further research.
• Details: Based on your findings and the limitations of the current
study, propose potential future studies. These could explore deeper
into causal relationships, different demographic settings, or interven-
tions that might address issues identified in the study.
7