This study aims to compare two different learning environments designed online in terms of their contributions to the development of mathematical modelling competencies. In the research, which was designed with the triangulation method, among the mixed research methods in which the qualitative and quantitative research methods are used together. The participants, determined with the purposeful sampling technique, consisted of 53 fourth class students who were studying at the mathematical teaching department of a faculty of education. In the research, while the quantitative data were analyzed with the paired samples t-test and Wilcoxon test, the qualitative data were analyzed with the content analysis. As a result of the analyses, while a significant difference was found in favour of the post-test between the modelling test questions of the pre-service teachers who attended the distance education in groups, the modelling test results of the pre-service teachers who attended individually did not differ. In addition, while it was observed that the participants who attended individually had difficulty in expressing themselves more throughout the process, it was determined that those who worked with groups tried to verify the results they reached more during the modelling process compared with those who participated individually.