Investigating customer perceptions of sustainable design features to drive purchasing decisions for sustainable products
Abstract/Contents
- Abstract
- Fierce competition on e-commerce platforms challenges designers to create products that appeal to customers. In particular, this occurs with sustainable products where an apparent demand for sustainable products fails to translate into real purchasing decisions. When creating sustainable products, designers tend to prioritize engineered sustainability features while neglecting perceived sustainability features. Engineered sustainable features are often hidden, for example energy usage or manufacturing methods of a product. Customers therefore rely on visual and descriptive features that align with what they perceive is sustainable, although these features may not contribute to real engineered sustainability. For a sustainable product to be successful, it therefore needs to meet both engineered requirements and perceived requirements. To study the role of perceived sustainability on driving purchasing decisions, the work presented in this dissertation takes a multidisciplinary approach borrowing techniques from computer science, design, and marketing. First, a data-driven approach was used to extract features perceived as sustainable from online reviews using crowdsourced annotations and a natural language processing machine learning algorithm. Second, a novel collage design approach was developed to test the extracted features from online reviews in terms of how users identify the features as sustainable. Third, a shopping simulation was developed to validate how features perceived as sustainable can influence purchasing decisions of products when compared to dummy features. Chapter 2 presents a method for designers to extract features perceived as sustainable from online reviews. Annotators identified phrases in product reviews that were relevant to one of the three sustainability pillars -- social, environmental, and economic -- and rated the positive and negative sentiment in the phrases. A logistic classifier was then used to extract salient features perceived as sustainable from the annotations. The method was tested on 1500 reviews of French presses and the extracted features were compared to a life cycle analysis output. The findings demonstrated that a gap exists between perceived and engineered sustainability, highlighting the importance of understanding features perceived as sustainable and the value of the proposed method. Chapter 3 investigates validity metrics of highly qualitative text annotations. While external validity metrics, for example, precision, recall, and F1, are commonly used in computer science, internal validity metrics such as inter-rater reliability, are commonly used in design. The study tested four variations of Krippendorff's U-alpha using the annotations from Chapter 2 to compare internal validity metrics with external validity metrics. The results found that external validity metrics are more robust in the case of highly qualitative text annotations, providing insight for designers on best practices for assessing validity of highly qualitative annotations. Chapter 4 presents a novel design method using a collage to test extracted features perceived as sustainable. The collage consisted of two axes, sustainability, and likeability, where participants placed products and selected features from a dropdown menu according to how they perceived the products. Participants evaluated six French press on the three sustainability pillars -- social, environmental, and economic -- and on how much they like the products. In the dropdown menu, participants selected between features perceived as sustainable and features perceived as not sustainable. The results suggested that participants more often selected features perceived as sustainable for products they placed higher on the sustainability axis, validating that they identified those features as sustainable. Moreover, a significant but low correlation was measured between the placement of products on the sustainability and likeability axis, demonstrating the value of the collage tool to measure both dimensions separately. The findings confirm that the collage is an effective method for testing features perceived as sustainable with users. Chapter 5 investigates the generalizability of the proposed methods by recreating them using electric scooters and baby glass bottles. Features perceived as sustainable were extracted for both products using the method outlined in Chapter 2. External validity metrics for electric scooters were comparable to that of the French presses, while they were much lower for baby glass bottles. It was identified that an imbalance of positive and negative reviews for baby glass bottles led to the weak performance in the machine learning models. The remainder of the study focused on electric scooters, testing the features using the collage approach outlined in Chapter 4. The findings were comparable to those in Chapter 4 with French presses. The study demonstrated that the proposed methods generalize with limitations, mainly that the selected products should have a balanced set of positive and negative reviews. Finally, Chapter 6 presents a shopping simulation to test how extracted features perceived as sustainable can influence purchasing decisions. A variety of features, images and descriptions were tested using a within-subject fractional factorial experiment. Participants navigated mockups of Amazon shopping pages and selected a product to purchase. They also rated products in terms of willingness to pay and sustainability. The results showed that participants were more likely to select to purchase products with features perceived as sustainable than dummy features. Moreover, participants were willing to pay more for products with perceived sustainability features and rated them as more sustainable, despite none of the features contributing to engineered sustainability. The findings validated that features perceived as sustainable can drive purchasing decisions and highlighted the importance of including both engineered and perceived features in sustainable design. This dissertation demonstrates the value of features perceived as sustainable in sustainable design. While they may not contribute to engineered sustainability, they align with what the customer expects is sustainable. The results underscore the importance of designing for both perceived and engineered sustainability requirements to drive growth for sustainable products.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | El Dehaibi, Nasreddine |
|---|---|
| Degree supervisor | MacDonald, Erin E |
| Thesis advisor | MacDonald, Erin E |
| Thesis advisor | Goodman, Noah (Noah D.) |
| Thesis advisor | Tucker, Conrad |
| Degree committee member | Goodman, Noah (Noah D.) |
| Degree committee member | Tucker, Conrad |
| Associated with | Stanford University, Department of Mechanical Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Nasreddine El Dehaibi. |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering. |
| Thesis | Thesis Ph.D. Stanford University 2021. |
| Location | https://purl.stanford.edu/sk676zf3930 |
Access conditions
- Copyright
- © 2021 by Nasreddine El Dehaibi
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...