Building a Student Performance Analysis Application in R

Preparing and Cleaning the Dataset

In this project, you will clean a messy file of 205 rows. First, run 00_make_data.R, which creates the raw file. It holds 200 students plus five repeated rows, and it has missing values in maths, attendance, and gender.

Save your own code as 02_intermediate_project.R.

The code below removes repeats, fills gaps, and adds an average mark and an attendance band. The check at the end stops the script if the cleaning did not work.

Writing the cleaning steps in code means anyone can repeat them on a new file.

Analyzing Student Performance and Attendance

Now use group_by() to compare sections and attendance bands.

• List the top five students.
• Measure correlations.
• Run a t-test and an ANOVA to check for real group differences.

Study the groups side by side, and ask whether any difference is large enough to matter.

Visualizing and Interpreting the Results

Three charts show the story:

1. A histogram of marks.
2. A scatter plot of attendance against marks.
3. A box plot by section.

The code saves them as images, and it also saves the cleaned file for the next level.

Always match each chart to one question, so the reader never has to guess its purpose.

Key Findings from the Analysis

Cleaning removed five repeated rows and filled all gaps, leaving 200 records.

• Study hours and marks: A strong link, with a correlation of 0.861.
• Attendance and marks: A weaker link of 0.285.
• Attendance group averages: Students with high attendance average 65.4 marks, compared with 57.8 for students with low attendance.
• Gender and section: No real difference was found.

Remember that the data is made up for practice, and filling gaps with the median hides some real variation.

Try it yourself: Drop the incomplete rows instead and compare.

By the end of the Intermediate Level, you can:

• Select, filter, group, sort, and combine data with dplyr.
• Clean missing values and remove duplicate rows.
• Import data from CSV and Excel files.
• Create clear charts with ggplot2 and Base R.
• Describe distributions, calculate summary statistics, and identify outliers.
• Explore relationships using correlation.
• Test group differences using a t-test and ANOVA.
• Build a student performance analysis application and interpret its results.

What Comes Next?

In the Advanced Level, you will learn to build prediction models and complete a more advanced data science project.