Designing and Testing a Technology Development Model Based on Environmental Ethics Using a Mixed-Methods Approach

Abstract

The present study was conducted with the aim of designing and testing a technology development model based on environmental ethics using a mixed-methods approach. In terms of purpose, this study is applied in nature, and in terms of data, it employed an exploratory mixed design combining qualitative and quantitative approaches. In the qualitative phase, the statistical population consisted of key informants in the field of research at the Department of Environment of Tehran Province. Using snowball sampling, interviews were conducted with 11 participants until theoretical saturation was achieved. In the quantitative phase, the population comprised all male and female managers and employees of the Department of Environment of Tehran Province (N = 1350). According to Krejcie and Morgan’s sample size table, a sample size of 298 was determined, and stratified random sampling was used. With an additional 10% to compensate for potential attrition, 302 questionnaires were collected. For data collection, library and field methods were used in the qualitative phase, while a researcher-made questionnaire was used as the field method in the quantitative phase. To assess the credibility of the qualitative instrument, three triangulation strategies were used: methodological triangulation (71%), investigator triangulation (77%), and participant triangulation (83%). In the quantitative phase, face and content validity were assessed, and reliability was determined using Cronbach's alpha, which indicated acceptable validity and reliability. Data analysis in the qualitative phase was performed using theoretical coding, while in the quantitative phase, confirmatory factor analysis was employed. The results showed that the technology development model based on environmental ethics consisted of 133 open codes (indicators), 26 axial codes (components), and 6 factors (dimensions). The causal factors included 4 axial codes (components) and 20 open codes (indicators); the contextual factors included 5 axial codes (components) and 30 open codes (indicators); the intervening factors included 5 axial codes (components) and 24 open codes (indicators); the strategies factor included 7 axial codes (components) and 34 open codes (indicators); and the consequences factor included 5 axial codes (components) and 25 open codes (indicators). Ultimately, the model fit indices, along with factor loadings and significance coefficients, indicated an acceptable model fit.