Back to The Numbers

Data Analytics Portfolio

Regression Analysis: Predicting Consumer Purchase Intent

A multiple regression analysis examining the key drivers of consumer purchase intent in the Ghanaian retail market, using survey data from 320 respondents.

2026
regression
Share
Regression Analysis: Predicting Consumer Purchase Intent

Tools Used

RSPSSExcelggplot2

Key Outcome

"Identified 3 statistically significant predictors of purchase intent (brand trust, perceived value, social influence), explaining 67% of variance (R² = 0.67, p < 0.001)."

Case Study

Overview

This project investigated the factors that drive consumer purchase intent in the Ghanaian retail sector. Using a structured survey instrument, data was collected from 320 respondents across Accra and Kumasi.

Methodology

A multiple linear regression model was built using the following independent variables:

  • Brand Trust Score (1–10 scale)
  • Perceived Value (price-quality ratio perception)
  • Social Influence (peer recommendations and social media)
  • Convenience (proximity and ease of purchase)
  • Previous Purchase Experience

The dependent variable was Purchase Intent, measured on a 5-point Likert scale.

Key Findings

The final model explained 67% of the variance in purchase intent (R² = 0.67, adjusted R² = 0.65, F(5, 314) = 127.3, p < 0.001).

The three most significant predictors were:

  1. Brand Trust (β = 0.42, p < 0.001) — the strongest predictor by far
  2. Perceived Value (β = 0.31, p < 0.001)
  3. Social Influence (β = 0.19, p = 0.003)

Implications

For Ghanaian retailers, these findings suggest that brand-building and trust signals (reviews, testimonials, consistency) should be prioritised over price competition alone. Social proof mechanisms — particularly peer recommendations — have a significant and measurable impact on purchase decisions.

Share

Explore More

More from The Numbers

View All Projects →