Profiling Conversational Programmers at University: Insights into their Motivations and Goals from a Broad Sample of Non-Majors

Background and Context. Instruction in most introductory computing courses is typically focused on how to program. However, non-majors who take computing courses have a diverse set of desired endpoints. One group of non-majors are the conversational programmers, who do not want to program in their career but enroll in computing courses to improve their ability to communicate about technical topics and their competitiveness in the job market. Research suggests that these learners need an alternate instructional approach, but so far, conversational programmers in higher educational contexts have only been studied in a limited number of small-scale studies. Objectives. To inform curriculum design for conversational programmers at the university level, we (a) examine the prevalence of conversational programmers among non-majors and their characteristics, (b) understand conversational programmers' desired learning goals and classroom activities, and (c) investigate factors associated with these learners' motivation to learn computing. Methods. We designed a survey based on Expectancy-Value Theory and prior work about conversational programmers. We collected responses from randomly sampled non-major students at a large public university, and we analyzed the survey data with descriptive and inferential statistics. Findings. We found that conversational programmers are the largest proportion of non-majors in our sample, both overall and across historically underrepresented groups in CS. We replicated prior findings of low self-efficacy for programming of conversational programmers. We found that conversational programmers' motivation for taking more computing courses is paradoxically driven more by their interest in computing than its utility, despite their general lack of enjoyment in computing. We validate a previously proposed set of conversational programmers' learning goals and show that they value employment-oriented learning goals over those focused on conversations. Implications. Our results suggest that addressing the needs of conversational programmers can contribute to broadening participation in computing. Our study motivates a learner-centered curriculum design that could address conversational programmers' learning needs by enhancing their self-efficacy and interests prior to focusing on conversational goals.