Introduction

Background:

• The topic inspired by my current workplace, an Australian owned active wear brand commited to responsible sourcing and sustainable packaging, and my personal interest in the fashion industry.
• In the last 15 years, clothing accounts for 60% of total textile use; production doubled, while clothing use dropped by 40% due to the fast fashion trend (Ellen MacArthur, 2019)
• Fashion manufacturing is the second-largest polluting industry, after oil with: water pollution from untreated water, high water consumption stressing freshwater sources, and rising textile waste from discarded clohing. (Vikas, 2021)
• 100 billion garments produced annually with over 92 million tonnes end up in landfills each year and its projected to reach 134 million tonnes/year by 2030 (Igini, 2023)
• Less than 1% of clothing materials are recycled, leading to over $100B material loss annually (Ellen MacArthur, 2019)

The Cause and Rationale

• Growing awareness of the environmental impact of the fashion industry.
• Clear cost gap between sustainable and non-sustainable fashion products (Blackstormco, 2024).
• Despite rising interest in sustainability, many consumers are still price-sensitive.
• 49% of consumers want to make eco-friendly choices but are unwilling to pay extra (Loughlin, 2023).
• The study aims to analyse price differences between eco-friendly and non-eco-friendly brands.
• Also examines the association between sustainable brands and recycling initiatives.

Rationale:

By examining the price difference between eco-friendly and non-eco-friendly manufactured brands and exploring the association between sustainability practices and recycling programs, this study seeks to provide data-driven insights into how sustainable production influences both pricing strategies and environmental responsibility. The result aim to help consumers better understand the cost-value relationship of eco-friendly fashion and evaluate whether sustainable brands actively contribute to waste reduction through recycling programs.

Problem Statement

Overall Problem/ Research Question:

1. Is the average price significantly different bertween eco-friendly and non-eco-friendly manufactured brands?
2. Is there a significant association between the type of manufacturing process (eco-friendly and non-eco-friendly) and the presences of recycling programs among fashion brands?
   

How Statistics Will Be Used:

1. Price Comparison:
   Apply independent two-sample t-test to compare the average between two manufacturing types. This will determine whether the difference is statistically significant.
   
   
2. Association Analysis:
   Use a Chi-Square test of association to examine the relationship between manufacturing type and recycling programs. This will determine whether there is statistically significant association between sustainability process and recycling programs.

Data Source and Description

• The data was obtained from Kaggle, an open-sourced datasets : https://www.kaggle.com/datasets/waqi786/sustainable-fashion-eco-friendly-trends?

• The dataset contains 15 variables.

• For the purpose of analysis, we focus on 3 Variables:
  

1. Manufacturing Type: Categorical (Eco / Non Eco-Friendly Manufacturing)
   
2. Recycling Program: Categorical (Yes/No)
   
3. Average Price (USD): Numeric (Continuous)
   
   

Data Pre-Processing

Before the analysis, we will start with checking the data types of the selected variables: Average Price (USD), Manufacturing Type and Recycling Program.

## [1] "numeric"

## [1] "character"

## [1] "character"

• As for the manufacturing type and recycling progam variables, we will convert the data type from character to factors to enable statistical analysis

## [1] "numeric"

## [1] "factor"

## [1] "factor"

• Checked for missing values in the selected variables: None found

## [1] 0

## [1] 0

## [1] 0

Check for the distributions and Outliers

To check the distributions and outliers , we use histogram and boxplot.

Based on the histogram, the distribution of Average Prices (USD) appears relatively uniform, suggesting that prices accross brands are evenly spread without a visible concentration in a particular range which indicates that overall there is no lower price segments or strong skew toward higher among the brands in the dataset.

Boxplot

The boxplot shows that there are no noticable outliers and the median prices for eco-friendly and non-eco friendly brands apepar to be in a similar range. This visual observation suggests that there may not be significant difference in average prices in average prices between two manufacturing types.

Descriptive Statistics and Visualisation

Based on the previous visual findings suggest that there are no clear price difference between eco-friendly and non-eco-friendly brands. To confirm this, we will do a statistical analysis comparing their average prices(USD).

Eco_Friendly_Manufacturing	min	max	mean	median	sd	n
Yes	20.02	499.94	257.7332	254.250	138.0998	2520
No	20.23	499.85	261.0014	262.735	137.3634	2480

Based on the descriptive statistics, the value for eco-friendly and non-eco-friendly manufactured brands are quite similar with:

• eco-friendly : mean = 258, median = 254, SD = 138
• non-eco-freindly : mean = 261, median = 263, SD = 137

These numbers suggest that the average prices between the two manufacturing types are very close, indicating no strong difference in central tendency or variability. However, with this descriptive statistics alone can not confirm whether the difference is statistically significant. To validate these observation, an independent two-sample hypothesis t-test and Association test will be done in the next section.

Hypothesis Testing:

Question: Is the average price significantly different bertween eco-friendly and non-eco-friendly manufactured brands?

Test Selection : A two-sample t-test was used to compare the average prices between two manufacturing groups with level of significance = 0.05.

Assumption Checks:

1. Normality: Q-Q Plots for both eco-friendly and non-eco-friendly price distributions show deviations from perfect normality. However, since both groups have sample sizes that greater than 30, Central Limit Theorem (CLT) justifies uses the t-test because the sampling distribution of the mean is approximately normal.

2. Homogeneity of Variance: A Levene’s test was conducted to test equality of variance:

• H0: Variances are equal
• H1: Variances are different

## Levene's Test for Homogeneity of Variance (center = median)
##         Df F value Pr(>F)
## group    1  0.6637 0.4153
##       4998

Based on the result: p = 0.4153 (>0.05). This means we Fail to reject the H0, Which indicates that the variance are equal between two groups so we proceed to the two-sample t-test assiming variances.

\[ H_0: \mu_1 = \mu_2 \quad \text{(eco-friendly average price= non–eco-friendly average price)} \] \[H_A: \mu_1 \ne \mu_2 \quad \text{(average price differ between two groups)}\]

## [1] 0.4015726

## [1] -10.905784   4.369407
## attr(,"conf.level")
## [1] 0.95

Result, Interpretation, and Discussions

Result:

• p-value: 0.4015726
• 95% Confidence Interval: [-10.905784, 4.369407] Because the p-value > 0.05, we fail to reject the null hypothesis. We are 95% confident that the true mean difference in the price between lies between $10.90 less and $4.36 more for eco-friendly brands compared to non-eco-friendly brands. Since this range includes 0, there is no statistically significant difference in average prices.

Interpretation and Discussion:

The results indicates that eco-friendly manufacturing does not lead to higher average prices which aligns with earlier visual and descriptive findings. This mean that sustainable production practices may not necessarily make products more expensive according to this dataset.

While this study focused on comparing the average prices between eco-friendly and non-eco-friendly brands, on the next analysis, we will analyse the association between sustainability practices by examining whether eco-friendly brands are more likely to implement recycling programs using Chi-Square test of Association. With this, it could provide insight to whether sustainable manufacturing practices are linked to other environmentally responsible initiatives within the fashion industry.

Association Test

Question: Does the type of manufacturing process (eco-friendly vs. non-eco-friendly) have a significant association with the presence of recycling programs among fashion brands?

Test Selection:

• Chi-square test of association was chosen because both variables are categorical: Manufacturing Type (eco / non-eco friendly) and Recycling Program ( Yes / no)
• Significance level of 0.05

Hypothesis: \[ H_0:\text{There is no association between manufacturing types and recycling programs.} \] \[H_A: \text{There is an association between manufacturing types and recycling programs.}\]

Data Preparation: We first create a contingency table of Recycling Programs by Manufacturing Types and calculates counts (n) and proporions (prop) for each combination.

Recycling_Programs	Eco_Friendly_Manufacturing	n	prop
Yes	Yes	1272	0.5031646
Yes	No	1256	0.4968354
No	Yes	1248	0.5048544
No	No	1224	0.4951456

Visuals

To see it visually we will use barplot and from this barplot, we can understand that eco-friendly has higher recycling prorgam than non-eco-friendly although it is not significantly different. However, to confirm whether this difference is statistically significant or not, we conduct a Chi-square test of association.

Assumptions and Tests

Assumption: Expected count has no more than 25% of cells should have expected counts < 5.

##      
##            Yes       No
##   Yes 1274.112 1245.888
##   No  1253.888 1226.112

From the contingency table and the expected counts calculated by chi_test$expected, we confirmed that all expected cell counts were greater than 5, this satisfied the assumptions for the Chi-square test of association.

Conducting the Chi-square test:

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  sustainPrac_table
## X-squared = 0.0083169, df = 1, p-value = 0.9273

Results and Interpretation: The Chi-square test yielded a p-value of 0.9273, which is greater than the significance level of 0.05. With this result, we fail to reject the null hypothesis, indicates that there is no statistically significant association between manufacturing types (eco-friendly and non-eco-friendly) and the recycling program. The data provide no evidence that eco-friendly manufacturing is associated / linked to higher likelihood of having a recycling program.

Discussion and Conclusion:

Major Findings:

1. Price comparison: The two-sample t-test revealed no statistically significant difference in average prices between eco-friendly and non-eco-friendly manufactured brands. With visualisations and descriptive statistics supported this result showing similar median, mean, and standard deviation values across two groups.
2. Association with Recycling Programs: The Chi-Square test indicates no significant association between manufacturing type and the presence of recycling programs. Both eco-friendly and non-eco-friendly brands were equally likely or unlikely to have recycling programs.

With this results, based on this dataset, eco-friendly manufacturing does not necessarily increase product prices, nor is it consistently associated with the adoption of recycling programs.

Strengths:

• Valid assumption checking: Normality and homogeneuity of variance assumptions were assessed properly before conducting the t-test and expected cell counts were verified for the Chi-square test.
• Balanced sample sizes: Both analyses had sufficient and relatively balanced group sizes that supports the validity and strengthness of the statistical results.

Limitations:

• The price analysis considered only one numeric variable and did not account for other brand-levels factos. Such as: target audience, geographic locations, and market segment.
• The findings might not generalise to all brands as the dataset may not represent the entire fashion industry.
• The study only assessed the presence of recycling programs without evaluates the program’s scope, quality, or other practices of sustainability.

Future Directions

For future investigations, it could expand on this study with :

• Examining associations between recycling porgrams, eco-friendly manufacturing, and other sustainability practices such as: waste reductions practices.
• Explore how the sustainability practices could relate to other outcomes, such as : brand popularity, environmental impact metrics, or consumer pricing.

This approach could offer a deeper insights into how sustainability practices are interconnected accross the fashion industry and guide both brands and consumers in making environmentally responsible decisions.

Final Conclusion and Take Home Message: Eco-friendly manufactured brands do not cost significantly more than the non-eco-friendly brands, so consumers can choose more sustainable options without a price penalty. To support both environmental sustainability and consumer preferencfes, brands should continue or start to adopt the eco-friendly manufacturing and implement recycling programs to reduce textile waste.

References

• Ali (2024). Sustainable Fashion: Eco-Friendly Trends. Kaggle. Retrieved 12 Oct 2025. https://www.kaggle.com/datasets/waqi786/sustainable-fashion-eco-friendly-trends?

• BlackstormCo (2024). The Cost Disrepancy Between Non-Environmentally Concious Products and More Green Product. Blackstormco. Retrieved 16 Oct 2025. https://blackstormco.asia/the-cost-discrepancy-between-non-environmentally-conscious-products-and-more-green-products/

• Igini, M (2023). 10 Concerning Fast Fashion Waste Statistics. Earth.Org. Retrieved 17 Oct 2025. https://earth.org/statistics-about-fast-fashion-waste/

• Loughlin, B (2023). Consumer Unwillingness To Pay Extra For Sustainable Products: Impact on Small Business. Retrieved 17 Oct 2025. https://instituteofsustainabilitystudies.com/insights/guides/consumer-unwillingness-to-pay-extra-for-sustainable-products/

• MacArthur, E (2017). A new Textiles Economy: Redesigning Fashion’s Future. Ellen MacArthur Foundation. Retrieved 17 Oct 2025. https://www.ellenmacarthurfoundation.org/a-new-textiles-economy

• Vikas, V (2021). Disastrous Effects of Non-Sustainable Fast Fashion. Outlook Money. Retrieved 17 Oct 2025. https://www.outlookmoney.com/outlookmoney/disastrous-effects-of-non-sustainable-fast-fashion-8352