This portfolio highlights hands-on data projects using tools like R, SQL, and Excel. Each section includes a real-world problem, the tools used, and key insights visualized for impact.
Student Performance Analysis: A Google Data Analytics Capstone Project.
Introduction
The purpose of this capstone project is to analyze student performance data in order to identify patterns and factors that influence academic success. By uncovering key insights related to variables such as study time, absences, parental education, and other personal or school-related factors, this analysis aims to help educators and policymakers make informed decisions to improve student outcomes.
Business Task
The goal is to uncover insights that can help educators understand key drivers of academic success and identify opportunities for targeted support.
About the Author
Francis Pekpe is a certified data analyst and published author of Power Within and Unbreakable. With a strong foundation in storytelling and communication, they combine their analytical skills and creative background to deliver insights that are both meaningful and easy to understand. After completing the Google Data Analytics Certificate, they now apply their expertise in R, SQL, and data visualization to real-world problems, as shown in this capstone project on student performance.
Data Source
To ensure the accuracy and reliability of the analysis, the dataset was cleaned using R. This process included checking for and removing any missing or null values, standardizing variable formats, and selecting only the most relevant columns for analysis—such as age, gender, study time, absences, and final grades. Duplicate entries were also checked and none were found. The cleaned dataset provides a solid foundation for meaningful exploration and visualization.
Data Cleaning
The original dataset contained 649 student records and 33 variables. Before analysis, several data preparation steps were performed to ensure the quality and relevance of the data.
| age sex |
| Min. :15.00 Length:649 1st Qu.:16.00 Class :character Median :17.00 Mode :character Mean :16.74 3rd Qu.:18.00 Max. :22.00 |
| studytime failures |
| Min. :1.000 Min. :0.0000 1st Qu.:1.000 1st Qu.:0.0000 Median :2.000 Median :0.0000 Mean :1.931 Mean :0.2219 3rd Qu.:2.000 3rd Qu.:0.0000 Max. :4.000 Max. :3.0000 |
| absences G1 |
| Min. : 0.000 Min. : 0.0 1st Qu.: 0.000 1st Qu.:10.0 Median : 2.000 Median :11.0 Mean : 3.659 Mean :11.4 3rd Qu.: 6.000 3rd Qu.:13.0 Max. :32.000 Max. :19.0 |
| G2 G3 |
| Min. : 0.00 Min. : 0.00 1st Qu.:10.00 1st Qu.:10.00 Median :11.00 Median :12.00 Mean :11.57 Mean :11.91 3rd Qu.:13.00 3rd Qu.:14.00 Max. :19.00 Max. :19.00 |
Insights and Recommendations
– Students who report studying more hours (study time level 3 or 4) tend to achieve higher final grades.
– There is a noticeable difference in grade distributions between male and female students, with females slightly outperforming on average.
– High absenteeism and past class failures appear to correlate with lower grades.
Recommendations
– Provide additional support to students with high absenteeism and previous failures, as they are more at risk of underperformance.
– Encourage consistent study habits, especially for those in the lower study time categories.
– Consider gender-based engagement strategies to ensure equitable support across the student body.
These findings can inform school policies or interventions aimed at improving academic performance.
dashboard – tableau
Introduction:
This project demonstrates how structured data from a fictional retail dataset was analyzed using SQL to uncover business insights. The focus was on identifying top-performing products, analyzing profit margins across categories, and spotting regional trends in sales performance. Through SQL queries, I efficiently filtered, grouped, and joined data tables to produce actionable insights that can support strategic business decisions. This project showcases my ability to work with relational databases, extract meaningful metrics, and communicate findings effectively.
View Dashboard on Tableau Public
Summary of Data & Insights
This project visualizes sales and profit data across regions and product categories. The goal was to surface actionable insights for optimizing sales strategy, inventory decisions, and marketing focus.
Key Insights:
· Western region leads in sales, Southern region leads in profitability.
· Technology is the top-selling category, but Office Supplies offers higher margins.
· Sales spikes during specific quarters suggest opportunities for seasonal campaigns.
Challenges and Value Delivered
Challenges:
- Cleaning raw data for Tableau compatibility
- Designing a layout that communicates both sales trends and profit breakdowns clearly
- Creating filters that allow stakeholder-level interactivity
Value Delivered:
- Created a visually intuitive dashboard that allows sales and marketing teams to explore performance by region and category
- Delivered actionable insights that help stakeholders plan promotions and inventory strategy
Marketing Performance Analysis (Email, Social & Ads)
This analysis presents a clear, visual representation of a multi-channel marketing campaign spanning five months. The goal was to monitor user engagement and conversion across three primary digital channels: Email, Social Media, and Online Ads.
Objective:
To track the effectiveness of each marketing channel in driving user actions, specifically clicks (interest) and conversions (action), and evaluate trends that could guide future strategy.
Key Insights:
- Consistent Growth: All three channels saw a steady increase in both clicks and conversions from January to May.
- Highest Engagement: Ads consistently drove the highest number of clicks and conversions, making it the top-performing channel.
- Email Efficiency: While Email had fewer clicks, it maintained a healthy conversion rate—suggesting more qualified leads.
- Social Media’s Role: Strong mid-tier performer with balanced growth, indicating successful audience targeting and engagement.
Visual Interpretation:
- Solid Lines represent Clicks for each channel.
- Dashed Lines represent Conversions, tracking how effectively interest translated into action.
- The chart makes it easy to detect dips (like in March for Email) and surges (like Ads in April/May), which are useful for performance reviews and campaign timing.
Conclusion:
This dashboard illustrates how data visualization helps in evaluating the return on marketing investments. By aligning efforts with months of higher engagement, teams can optimize campaigns, allocate budgets wisely, and drive meaningful growth.