From Sustainability Claims to Certified Sustainability: A Descriptive Data Analytics Approach
582.54 KB
avoin
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
Lataukset13
Pysyvä osoite
Verkkojulkaisu
DOI
Tiivistelmä
Sustainability and environmental impact are issues of extreme importance, that should affect consumer behaviour and purchase decisions. Information considering sustainability and environmental impact of a product is limited and dependent on the willingness of brands to display said information, apart from eco-labels required by legislation. A major problem in assessing a product’s sustainability is greenwashing: a misleading and dishonest form of green marketing, that has several negative effects on the efforts towards a more sustainable market economy.
Inspired by the EU directive 2024/825 for empowering consumers for the green transition through better protection against unfair practices and through better information, this thesis aims to apply descriptive data analytics to real-life data in order to describe the relationship between sustainability claims in product descriptions and the real sustainability of a product depicted by the product having at least one certification, and to see if sustainability related terms are used with unsustainable products in order to mislead the customer. The research question is: "How strongly is a sustainability claim of a product description associated with the product’s certification status?".
To research the relationship between generic sustainability claims and product certifications, logistic regression models were made using GreenDB data. Due to extreme sparsity of the model features in the data, meaning the lack of generic environmental claims in the product descriptions, the logistic regression models were not able to provide any reliable results and detecting greenwashing from the false positives was not possible. As previous research shows, greenwashing is very hard to identify and assess without substantial proofs of sustainability or a labeled dataset. However, statistical tests with odds ratios did show a statistically significant relationship between at least one generic sustainability claim in the product description and the product having at least one certification, meaning that a sustainability claim in the product description increases the odds of the product having a certification by 64 % in the dataset used.