Uppsats

Modeling sustainable retail adoption among Indonesian Generation Z consumers : A logistic regression approach on zero waste stores

Kandidat-uppsats

KTH/Sannolikhetsteori, matematisk fysik och statistik

Publicerad: 2026

Språk: Engelska

Sammanfattning

Plastic packaging waste is one of Indonesia's largest environmental problems, with around 4.9 million tons mismanaged each year. Zero waste stores (ZWS), which sell goods in bulk and rely on customers bringing their own containers, have emerged as a possible solution to reduce single use packaging. Their long term viability depends on whether consumers are willing to shop there. Generation Z (Gen-Z) is of particular interest, as it is both a large share of the Indonesian consuming population and is commonly described as the most environmentally aware generation. This thesis investigates how consumer behavior influences the success of zero waste retail in Indonesia, focusing on Gen-Z. The aim is to identify which behavioral, attitudinal and demographic factors best predict whether a consumer has shopped at a ZWS. The study is based on a survey of 491 Indonesian Gen-Z respondents and uses logistic regression as the main statistical method. Three model specifications were fitted, covering attitudinal and behavioral variables, demographic variables, and the two combined. Best subset selection based on AIC was used to obtain reduced models, and the final model was evaluated through residual diagnostics, bootstrap confidence intervals, ten fold cross validation and ROC analysis. The findings were validated against a survey of ZWS owners. The results show that prior knowledge of ZWS, gender, occupation and willingness to accept less protective packaging are the strongest predictors of having shopped at a ZWS. General environmental awareness was not significant. The final model indicates that the main barriers to scaling ZWS are demographic targeting and perceived trade offs rather than awareness.

Utforska vidare

Liknande uppsatser

Uppsatser med liknande ämnen och nyckelord.