Student Experience Analytics

Student Experience Analytics is a data-driven project aimed at evaluating and improving students' learning experiences at an EdTech company.

The project involves collecting real-world data through a Google Form survey, which gathers insights on key aspects such as course quality, instructor expertise, support services, the Learning Management System (LMS), and the overall learning environment.

The collected data is extracted from Google Sheets, transformed into a structured format (CSV) using Python, and automated using GitHub Actions for continuous updates. Through analytical techniques, the project identifies trends, pain points, and areas for improvement, enabling data-driven decision-making to enhance student satisfaction and learning outcomes.

Table of Contents

1. Project Overview
2. Key Features
3. Tech Stack
4. Installation & Setup
5. Directory Structure
6. ELT Architecture Overview
7. Data Sources
8. Usage
   • Exploratory Data Analysis (EDA)
     • Demography Analysis
     • Course Experience Analysis
     • Instrutor Evaluation
     • Overall Satisfaction and NPS Analysis
     • Correlation Analysis
   • Sentiment Analysis of Open-Ended Responses
9. Contributing
10. License
11. Contact

Project Overview

This project leverages data analytics and machine learning to evaluate student experiences, including:

• Course Experience Analysis
• Learning Environment Analysis
• Overall Satisfaction and NPS Analysis
• Predictive modeling (dropout risk, performance forecasting)
• Institutional insights for educators & administrators

Goal: Enhance student experience through actionable data insights.

Key Features

• Automated Data Processing – Clean, transform, and analyze student datasets.
• Interactive Dashboards – Visualize trends using Streamlit.
• Predictive Analytics – ML models to forecast performance.
• Sentiment Analysis – Evaluate student feedback from surveys.
• Scalable Architecture – Modular code for different educational datasets.

Tech Stack

• Languages: Python
• Libraries: Pandas, NumPy, Scikit-learn, Matplotlib/Seaborn, Plotly
• Tools: Jupyter Notebook, VS Code, Git
• Deployment: Streamlit (for demo dashboards)

Installation & Setup

Prerequisites

• Python 3.8+
• Pip package manager

Steps

1. Clone the repository:

   git clone https://github.com/intellisenseCodez/student-experience-analytics.git
   cd student-experience-analytics
   

2. Set up a virtual environment (recommended):

   python3 -m venv venv
   source venv/bin/activate  # Linux/Mac
   venv\Scripts\activate     # Windows
   

3. Install dependencies::

   pip install -r requirements.txt
   

4. Run Code to analysis

   python ./src/extract_raw_data.py               # extract new data from API
   jupyter execute notebooks/01-transform.ipynb   # perform data transformation
   jupyter execute notebooks/02-analytics.ipynb   # perform EDA
   
   

Directory Structure

project-root/
├── .github/
│   └── workflows/
│       └── elt_pipeline.yml
├── data/
│   ├── cleaned/                    # Transformed data
│   └── raw/                        # Raw extracted data
├── images/
├── notebooks/
│   ├── 01-transform.ipynb          # Transformation script
│   └── 02-analytics.ipynb          # Analysis script
├── reports/                        # visualizations
├── src/
│   ├── __init__.py     
│   ├── dashboard.py                # streamlit demo dsahboard
│   ├── extract_raw_data.py         # extract script
│   ├── google_service.py           # google API service
│   ├── logger.py                   # logger setup
├── .env
├── .gitignore                      # 🔐 Added to .gitignore
├── README.md                       # README.md
└── requirements.txt                # requirements

ELT Architecture Overview

Data Sources

• Real-world data was collected through a Google Form survey to assess students' experiences at an EdTech company. The survey focused on key aspects such as course quality, instructor expertise, support services, the learning management system (LMS), and the overall learning environment.

The survey form includes questions such as:

• How would you rate your overall learning experience.
• Is the course meeting your expectations?
• Rate the effectiveness of the teaching method.
• Would you recommend us to others? etc.

Security Setup

To use the Google Sheets API, you need a Google Cloud Platform Project with the API enabled, as well as authorization credentials

• Google Sheets API Setup:

  • Create a New Project in Google Cloud Console

  • Enable API and Services

  • Create a service account

  • Create New Key, downloaded JSON file, and move it into your project folder

• GitHub Secrets:

  Settings > Secrets > New repository secret
  
  

Usage

Exploratory Data Analysis (EDA)

This contains an exploratory data analysis of student experience survey data for an Edutech company. The analysis aims to identify key insights, strengths, and areas for improvement to help enhance educational services and better support students.

Demographic Analysis

1. Gender Distribution

The survey shows a gender imbalance, with male respondents (64.8%) significantly outnumbering females (34.3%). The minimal representation of non-disclosed genders suggests the need for more inclusive options.

2. Current Status distribution

96.3% of respondents are current students, while only 3.7% are alumni. This suggests the feedback primarily reflects the experience of active learners, with minimal input from graduates.

Course Experience Analysis

1. Overall learning experience rating

• Positive Skew with Room for Improvement

  • 60.2% of ratings are 4-5 stars (32.4% + 27.8% = 65/108 responses), indicating generally satisfied learners.

  • However, 28% gave 3 stars (neutral), suggesting a sizable group finds the experience "average" with unmet potential.

  • Only 1.9% gave the lowest rating (1-star), which is favorable but warrants investigation into their pain points.

2. Instrutor Evaluation

• Strong Performance with Room for Excellence

  • High Ratings Dominate:

  • 34.3% of students gave 5/5 ("Excellent"), reflecting strong confidence in instructors’ expertise.

  • Combined 4-5 star ratings cover ~70%+ of responses (exact % depends on missing 4-star data), indicating most learners view instructors as highly knowledgeable.

  • Average: 4.1/5—a robust score, but the skew toward 5 stars suggests outliers may be pulling the average up.

• Critical Minority Dissatisfaction

  • 14.3% rated instructors 1-2/5 ("Poor"), a concerning gap. Potential issues:

  • Knowledge gaps: Instructors may lack depth in niche topics.

    • Delivery: Expertise ≠ effective teaching (e.g., poor explanations, unengaging style).

    • "3-star" ratings (if any): Neutral scores (not shown) could indicate adequate but uninspiring expertise.

Overall Satisfaction and NPS Analysis

Net promoter score (NPS), also known as net promoter, is a metric that assesses the willingness of customers to recommend a company’s products or services to other people. The metric aims to identify customers who are less satisfied with the customer experience or product and transform them into the company’s promoters.

The net promoter score’s calculation is based on a customer survey. The survey asks only one question: “On a scale from 1 to 5, how likely would you recommend our product/service to other people?”

Based on their responses, all respondents are broken down into three categories:

• Detractors: Unhappy customers who can spread negative reviews of a company.
• Passive: Satisfied but unenthusiastic customers who can be taken by competitors.
• Promoters: Loyal customers who can spread positive reviews of a company.

Net promoter score is determined as the percentage difference between promoters and detractors:

Net Promoter Score = Promoters (%) – Detractors (%)

Correlation Analysis

• Strongest Positive Correlations:

  • Overall Learning Rating & Overall Satisfaction (0.77): This suggests that a positive learning experience significantly impacts overall satisfaction.

  • Learning Environment Comfort & Overall Satisfaction (0.6): A comfortable learning environment contributes to student satisfaction.

  • Instructor Expertise & Instructor Approachability (0.53): More knowledgeable instructors tend to be perceived as more approachable.

• Moderate Positive Correlations:

  • Instructor Expertise & Overall Learning Rating (0.47): High-rated instructors contribute to a better learning experience.

  • Instructor Approachability & Overall Learning Rating (0.48): When students feel their instructors are approachable, they rate their learning experience higher.

  • Customer Support Satisfaction & Overall Satisfaction (0.38): A well-supported learning experience leads to increased satisfaction.

• Negative Correlations:

  • LMS Platform Rating & Overall Learning Rating (-0.34): A surprising inverse relationship, possibly indicating dissatisfaction with the learning platform despite good learning outcomes.

  • LMS Platform Rating & Overall Satisfaction (-0.34): A poorly rated LMS could negatively impact overall student satisfaction.

• Insights & Recommendations:

  • Improving the LMS platform experience: Since it correlates negatively with satisfaction, identifying and resolving platform issues (usability, responsiveness, features) can enhance the student experience.

  • Focus on Instructor Quality: Both expertise and approachability significantly impact the learning rating and satisfaction. Investing in instructor training and mentorship can improve these scores.

  • Enhancing Learning Environment: Since it correlates well with satisfaction, ensuring a conducive environment (both physically and virtually) should be a priority.

  • Customer Support Optimization: Although it has a moderate effect, good support can increase satisfaction.

Sentiment Analysis of Open-Ended Responses

Individual Sentiment Distributions:

• Course Improvement Suggestions (Blue Histogram):

  • The distribution appears somewhat skewed towards the positive side, with a noticeable peak around the neutral to slightly positive range (around 0 to +0.25).
  • The mean sentiment polarity is 0.23, and the median is 0.22, both indicating a slightly positive overall sentiment towards course improvement suggestions.
  • There is a spread of sentiment, with some negative suggestions as well, but they are less frequent.

• Instructor Improvement Suggestions (Green Histogram):

  • This distribution shows a strong positive skew, with a prominent peak in the moderately positive range (around +0.25 to +0.75).
  • The mean sentiment polarity is 0.50, and the median is 0.51, both indicating a generally positive sentiment in the suggestions for instructor improvement. This might suggest that while students have ideas for improvement, their feedback is often constructive.

• Learning Environment Improvement Comments (Orange Histogram):

  • The distribution is more centered around the neutral point (0), with a wider spread compared to the other categories.
  • The mean sentiment polarity is 0.08, and the median is -0.05, suggesting a slightly negative to neutral overall sentiment regarding the learning environment. This aligns with previous analyses indicating concerns about the learning environment.

• Customer Support Comments (Red Histogram):

  • This distribution is heavily skewed towards the positive end of the spectrum, with a dominant peak in the strongly positive range (around +0.5 to +1.0).
  • The mean sentiment polarity is 0.70, and the median is 0.71, both indicating a highly positive sentiment towards customer support. This reinforces earlier findings that customer support is a strength.

• Course Duration Suggestions (Green Histogram):

  • The distribution appears somewhat bimodal or at least has a broader peak around the neutral to slightly positive range (around 0 to +0.25), with some spread towards both negative and positive sentiments.
  • The mean sentiment polarity is 0.11, and the median is 0.11, both indicating a slightly positive overall sentiment regarding suggestions for course duration. This suggests that while students have opinions on duration, they might not be strongly negative or positive.

• Summary Statistics Table: The table provides numerical confirmation of the observations from the histograms:

  • CourseImprovement_Sentiment: Mean (0.234), Median (0.217) - Slightly positive.
  • Instructor_Sentiment: Mean (0.502), Median (0.506) - Moderately positive.
  • Environment_Sentiment: Mean (0.079), Median (-0.048) - Slightly positive mean, slightly negative median, indicating a more neutral to slightly negative leaning.
  • Support_Sentiment: Mean (0.696), Median (0.706) - Strongly positive.
  • Duration_Sentiment: Mean (0.115), Median (0.110) - Slightly positive.

• Overall Analysis and Implications

  • Customer Support is a Clear Strength: The overwhelmingly positive sentiment towards customer support is a significant asset for the company. This should be maintained and highlighted.

  • Instructor Feedback is Generally Positive: While students offer suggestions for improvement, the overall sentiment towards instructors is positive. This suggests that the instructors are generally well-received, but there's always room for refinement based on the specific suggestions.

  • Course Improvement Suggestions are Mildly Positive: Students have ideas for making the courses better, and their sentiment is generally constructive. Analyzing the specific content of these suggestions is crucial for actionable improvements.

  • Learning Environment is a Key Area for Improvement: The slightly negative to neutral sentiment surrounding the learning environment aligns with previous direct feedback. Addressing the specific issues raised in the open-ended comments related to the learning environment should be a priority to enhance the overall student experience in Lagos.

  • Course Duration Feedback is Mixed but Slightly Positive: Students have varying opinions on course duration. Further analysis of the specific suggestions could reveal whether there's a desire for longer, shorter, or more flexible durations for certain courses.

• Recommendations:

  • Leverage Positive Sentiment: Recognize and appreciate the positive aspects of customer support and instructors.
  • Prioritize Learning Environment Improvements: Focus on addressing the concerns and suggestions related to the learning environment to improve student comfort and effectiveness.
  • Analyze Qualitative Data: Conduct a detailed qualitative analysis of the open-ended feedback within each category to understand the specific suggestions and reasons behind the sentiment scores. This will provide actionable insights for course content, instructor training, learning environment enhancements, customer support protocols, and course duration adjustments.
  • Monitor Sentiment Over Time: Track these sentiment scores over future surveys to measure the impact of any changes implemented based on this feedback.

  This sentiment analysis provides a valuable layer of understanding to the direct feedback, allowing the company to prioritize areas for improvement based on the overall emotional tone of the student comments.