This study tested a self-generated utility value intervention aimed at increasing undergraduate statistics students’ motivation and achievement. The intervention was based on Situated Expectancy-Value Theory (Eccles & Wigfield, 2020) and encouraged students to make relevant connections between statistics learning content and their lives, primarily emphasising the content’s usefulness to the student, or utility value. In testing a self-generated utility value intervention within the domain of undergraduate statistics, the study extended research previously conducted in high school and undergraduate sciences (psychology and biology) and replicated Hulleman et al. (2017) which tested the role of frequency of students’ connections between the learning content and their lives in their motivation and achievement. In addition to transferring a self-generated utility value intervention to the domain of statistics, the study’s main contribution was made by investigating the role of connection quality—the quality of utility value connections undergraduate statistics students made between the learning content and their lives in their motivation and achievement. The study used collected data from a blindly randomised longitudinal field experiment conducted with undergraduate business school students from a research-intensive university located in the north-eastern USA. The students were of two differing sections of the same 15-week introductory statistics course. The self-generated utility value intervention consisted of prompts, twice during the semester, which instructed stud¬¬ents to write 2-3 paragraphs in response to. Data collected was comprised of students’ gender, first-generation status, initial/final achievement assessments, pre/post self-reports on motivation (expectancy, cost, intrinsic value, utility value) and connection frequency, and researcher scaled ratings coding on student intervention responses for connection quality. Part I Results from this study suggest that the intervention significantly increased students’ achievement (d = .42)—an approximately 7-percentage point difference between intervention and control group conditions. Furthermore, the intervention was found to be especially effective at increasing at-risk, low initial achievement, students’ motivation (expectancy, d = .54) and achievement (d = .87)—an approximately 14.5-percentage point difference between group conditions. Study results also suggest that the intervention’s impact on at-risk students’ achievement was mediated via motivation increases—through students’ expectancy for success, though, not through students’ utility value. The Part I results were confirmatory of Hulleman et al.’s (2017) findings—the intervention effected students’ achievement, but the pathway of indirect effects traversed through students’ expectancy, not their utility value which Hulleman et al. (2017) and this study both hypothesised it would do instead. Part II Results attempted to explain the intervention’s pathways of effects through expectancy to achievement by creating new measures, connection quality measures. Connection quality measures were constructed to capture students’ utility value more effectively than the self-reported utility value survey measure. This study’s Part II Results suggest that the intervention was found, again, to significantly increase students’ achievement (d = 1.46), but the indirect intervention effects traversed a pathway to affecting students’ achievement, not through their expectancy, but through their utility value (as captured via the newly minted connection quality measures), to their motivation (cost and interest), and then to their achievement. The new connection quality measures, exploratorily, were found to capture students’ utility value more effectively than the self-reported utility value survey measure, enabling the self-generated utility value intervention’s effects on students’ achievement and motivation to be further explained. / Math & Science Education