📊 Faragir Sanat Mehrbin — Customer Satisfaction Analysis (SPSS)

📚 Table of Contents

• 🔖 Project Overview
• ✅ Step 1. Descriptive Statistics
• 🧪 Step 2. Reliability Test (Cronbach’s Alpha)
• 🔗 Step 3. Correlation Analysis
• 📈 Step 4. Regression Analysis
• 🚀 Recommendations to Lift Satisfaction

🔖 Project Overview

This repo outlines a focused customer satisfaction study using SPSS. The aim is to map how key experience drivers—Delivery, Product Quality, Price Fairness, Support, Ease of Ordering—relate to Overall Satisfaction, and package the work for decision-makers.

Inputs. Likert (1–5) survey covering the factors above .
Methods (SPSS). Descriptives → Reliability (Cronbach’s α) → Correlations → Linear Regression.
Outputs. Clean, GitHub-ready tables & visuals; simple explanations; an action framework (impact × effort).

✅ Step 1. Descriptive Statistics

🔍 What this step does

Summarises each factor’s central tendency (Mean) and dispersion (Std. Dev.).
Scale: 1 = Very dissatisfied · 3 = Neutral · 5 = Very satisfied

📊 Results (SPSS)

Factor	N	Mean	Std. Dev.
Overall Satisfaction	549	3.58	1.051
Delivery	549	3.49	0.941
Product Quality	549	3.72	0.712
Price Fairness	549	3.43	0.631
Support Satisfaction	549	3.59	0.635
Ease of Ordering	549	3.82	0.524
Valid N (listwise)	549

📝 Reading guide:
Higher Mean = better perceived experience.
Std. Dev. ≤ ~1.0 ⇒ customers broadly agree; >1.2 ⇒ views are polarised.

💡 Interpretation (business)

• Strong baselines: Ease of Ordering (3.82), Product Quality (3.72).
• Improvement candidates: Price Fairness (3.43), Delivery (3.49).
• Consistency: All SDs ≤ ~1.05 → perceptions are stable; improvements should shift the overall average.

🔑 Key Takeaway (simple)

People say ordering is easy and quality is good. They are less happy with price fairness and delivery.
Focus next on clearer pricing and on-time delivery.

🧪 Step 2. Reliability Test (Cronbach’s Alpha)

Purpose. Check whether the five driver items (Delivery, Product Quality, Price Fairness, Support, Ease of Ordering) behave consistently as a scale. Overall Satisfaction is excluded—it’s the outcome.

📊 Result (SPSS)

Metric	Value
Cronbach’s Alpha	0.868
N of Items	5

💡 Interpretation (reporting)

An alpha of 0.868 indicates good internal consistency: these drivers are measuring the same underlying experience without being redundant. This means the scale is stable and reliable, and each item still contributes useful, distinct information.

🧭 What this enables

We can safely use these items together in the next steps—examining relationships with Overall Satisfaction (correlations) and estimating unique driver impact (regression)—while keeping Overall Satisfaction as the dependent variable.

---

🔗 Step 3. Correlation Analysis

Purpose. Identify which experience factors move with Overall Satisfaction before controlling for overlap.

📊 Correlation Matrix (Pearson r)

	Overall Satisfaction	Delivery	Product Quality	Price Fairness	Support Satisfaction	Ease Of Ordering
Overall Satisfaction	1.000	.699**	.632**	.677**	.629**	.619**
Delivery	.699**	1.000	.610**	.655**	.629**	.575**
Product Quality	.632**	.610**	1.000	.579**	.598**	.549**
Price Fairness	.677**	.655**	.579**	1.000	.607**	.531**
Support Satisfaction	.629**	.629**	.598**	.607**	1.000	.579**
Ease Of Ordering	.619**	.575**	.549**	.531**	.579**	1.000

Notes: ** indicates p < .01 (2-tailed). N = 549 for all pairs.

💡 Interpretation (reporting)

All five drivers are strong, positive companions of Overall Satisfaction. The tight cluster of high correlations shows a coherent customer experience, with Delivery and Price Fairness showing the strongest raw links. Because drivers also correlate with each other, we proceed to regression to isolate unique impact.

🔑 Key takeaway

Each driver is meaningfully tied to satisfaction; Delivery and Price Fairness stand out prior to controls.

📈 Step 4. Regression Analysis

Purpose. Estimate each driver’s unique contribution to Overall Satisfaction while controlling for overlap among drivers.

📊 Model Summary

Model	R	R Square	Adjusted R Square	Std. Error of Estimate
1	0.798	0.636	0.633	0.637

🧪 ANOVA

Source	Sum of Squares	df	Mean Square	F	Sig.
Regression	385.256	5	77.051	189.983	.000
Residual	220.223	543	0.406
Total	605.479	548

🔩 Coefficients (DV = Overall Satisfaction)

Term	B	Std. Error	Beta	t	Sig.
(Constant)	-1.803	0.218	—	-8.268	.000
Delivery	0.295	0.044	0.264	6.738	.000
Product Quality	0.238	0.054	0.161	4.442	.000
Price Fairness	0.407	0.062	0.244	6.536	.000
Support Satisfaction	0.185	0.063	0.112	2.958	.003
Ease Of Ordering	0.370	0.070	0.184	5.314	.000

🧾 Reporting Note

The model explains 63.6% of variance in satisfaction (R² = 0.636, F(5,543) = 189.98, p < .001). All predictors are statistically significant. By standardized effects (Beta), Delivery shows the largest unique association, followed by Price Fairness, Ease, Product Quality, and Support.

🔑 Key takeaway

Satisfaction is well-explained by the five drivers; Delivery and Price Fairness provide the strongest unique lift when other factors are held constant.

🚀 Recommendations to Lift Satisfaction

🎯 Main Focus: Fair Prices

Why: lowest mean 3.43, strong link to satisfaction (r = .677), solid unique impact (β = .244, B = .407/pt).
Do next: Show the full price early (include shipping/tax), offer bundle/volume discounts, and explain the value (warranty, calibration, reviews).
Check weekly: “Price fairness” rating in the survey, and how many people quit at the payment page.

📦 Next: Reliable Delivery

Why: second lowest mean 3.49, strongest driver (r = .699, β = .264).
Do next: Show the delivery day before paying, give a tracking link, and message customers if there’s a delay.
Check: % of orders that arrive on the promised day, average days late, and damaged items.