Aggregated A/B Experiment Data
Data Science and Analytics
Tags and Keywords
Trusted By




"No reviews yet"
Free
About
This dataset contains simulated user-aggregated data from an A/B experiment, designed for interview preparation and the practice of statistical tests. It captures webpage views and button clicks, offering insights into user behaviour within an e-commerce service context. The data was generated using specific parameters, including success rate, uplift, beta, and skew, making it ideal for analysing A/B test results.
Columns
- user_id: A unique identifier for each user.
- group: Represents the experiment group to which the user belongs (e.g., 'control' or 'experiment').
- views: The total number of webpage views recorded for a user.
- clicks: The total number of experimental feature clicks by a user.
Distribution
This dataset is provided in a CSV file format, specifically named
ab_test_results_aggregated_views_clicks_2.csv
, with a file size of 2.48 MB. It contains 120,000 valid records, with no mismatched or missing entries for any of its four columns. The user_id
values range up to 120,000, with 2,400 user IDs per defined range. The 'group' column has two unique values, with 'control' being the most common, accounting for 50% of the entries. The 'views' column has a mean of 4.99 and a maximum of 205 views, while the 'clicks' column has a mean of 0.18, with a maximum of 9 clicks per user. Notably, 101,477 records show zero clicks.Usage
This dataset is well-suited for a variety of applications, including:
- Interview preparation: Excellent for practicing data analysis and statistical testing scenarios commonly encountered in data science and analytics interviews.
- Statistical analysis: Ideal for applying various statistical tests to determine the significance of A/B test results.
- Exploratory data analysis: Allows for in-depth exploration of user interaction data, such as webpage views and clicks, in an e-commerce setting.
- Data analytics exercises: Can be used to understand user behaviour patterns and evaluate the impact of experimental features.
Coverage
The data represents simulated user-aggregated information from an A/B test. As such, it does not have a specific geographic or demographic scope, nor a defined real-world time range. All data points are valid and there are no missing entries.
License
CC0: Public Domain
Who Can Use It
- Aspiring data analysts and scientists: To hone their skills in A/B test evaluation and statistical inference.
- Students and educators: For learning and teaching concepts related to experimental design and data analysis.
- Researchers: For studying user behaviour modelling in a controlled, simulated environment.
Dataset Name Suggestions
- A/B Test User Behaviour Data
- Simulated Webpage Experiment Results
- E-commerce A/B Test Analytics
- User Views and Clicks Data
- Aggregated A/B Experiment Data
Attributes
Original Data Source: Aggregated A/B Experiment Data