An empirical research project applying core statistical methodologies to investigate, clean, and model consumer data within the baby stroller industry. This study evaluates the explicit relationship between commercial pricing structures, mechanical properties, and real-world consumer satisfaction.
Baby strollers represent a major financial investment for families, with prices spanning from budget tier options to high-end luxury models. This project bridges data science and market analysis to discover whether premium pricing inherently scales with consumer satisfaction, or if specific categories (such as Jogging or Travel Systems) offer superior utility per dollar.
-
Data Harvesting: Compiling structural product variables across a sample size of
$n \ge 30$ unique stroller models. - Exploratory Data Analysis (EDA): Visualizing features, identifying outliers, and analyzing distribution normality.
- Inferential Modeling: Using Simple Linear Regression and Hypothesis testing to determine structural value patterns.
The data framework captures both quantitative and qualitative variables to construct a balanced model:
| Variable Profile | Data Type | Operational Definition |
|---|---|---|
| Price ( |
Quantitative (Continuous) | Independent variable tracking retail MSRP in USD. |
| Customer Rating ( |
Quantitative (Continuous) | Dependent variable tracking aggregate consumer score (1.0–5.0 stars). |
| Weight | Quantitative (Continuous) | Physical net weight of the stroller chassis measured in lbs/kg. |
| Stroller Type | Qualitative (Categorical) | Classification groupings: Standard, Jogging, Lightweight, Double. |
| Brand Tier | Qualitative (Categorical) | Grouped economic classifications: Economy, Mid-Range, Premium. |
To prevent extreme luxury entries or anomalous data rows from skewing the descriptive metrics, data points are filtered using the strict Interquartile Range criteria:
The primary hypothesis tests if a product's price predictably determines its consumer evaluation metrics using an Ordinary Least Squares (OLS) model:
-
Null Hypothesis (
$H_0$ ): Price has no predictive power over user rating ($\beta_1 = 0$ ). -
Alternative Hypothesis (
$H_a$ ): Price significantly impacts user rating trends ($\beta_1 \neq 0$ ). -
Alpha Threshold: Tested strictly at
$\alpha = 0.05$ .
Ensure your local analysis system has the following core runtime environments configured:
- Next Js or R Environment v4.0+
- Analysis Libraries:
pandas,numpy,scipy,matplotlib,seaborn
- Clone the repository locally:
git clone [https://github.com/ruslanleker1663/stroller-statistic-rullers.git](https://github.com/ruslanleker1663/stroller-statistic-rullers.git) cd stroller-statistic-rullers