Pfl Wp2

Gatin 1

Arthur Gatin Professor Fedorova Writing 2 22 March 2019 Writing Conventions That Have Shaped Computer Science Education Education is often the first discipline to emerge in a new field, and computer science is no exception. One of the institutions that pioneered teaching computer science was Purdue University. Their computer science department, according to the university’s website, was established in 1962, but it was not until the 1970s that the faculty reached ten members. Other universities across the nation share a similar timeline for the formation of computer science departments, which marks the beginning of education in computer science. Ever since then, faculty members across the globe have been collaborating on creating publications on increasing effectiveness of their teaching and sharing those publications within their discourse community. This long-lasting process has since changed the field and created certain writing conventions used today. In the current net of ideas, computer science educators build on existing work, acknowledge the context of their studies, and reflect on student feedback to make their publications more effective and make a significant difference in the discipline. One defining feature of publications in CS education is their reliance on other works. Although it is a relatively common writing practice among academics in various fields, it has greater importance in Computer Science education. One of the main reasons for it is the fact that CS education has a closed set of topics, and almost 50 years of research have made it increasingly difficult to find a paper that would unravel something previously unseen. As a

Gatin 2

result, according to Diba Mirza, Ph.D., professor of computer science at UCSB and a CS education researcher, modern publications in computer science education are divided into three major groups. They either summarize teaching experience with a particular program or feature, review and make suggestions to existing teaching strategies, or provide an overview of the current state of research on a topic (personal communication, February 6, 2019). References to previous research or experiments are essential for papers from any of the said groups to secure their place in the net of knowledge. Professor Ivona Bezakova’s publication on board games in introductory computer science courses presents an excellent example of how references may be used in such publications. Early in the paper, Bezakova establishes why such research was relevant – “Previous research has demonstrated that learning to program in context increases student motivation [23, 24]” (Bezakova 17). This allows the audience to get an idea of what has motivated her research, provides related work on the matter, and illustrates how her results may be applied in the classroom. Other references serve a different purpose, providing examples of existing games that were used by Bezakova as models for creating her own: “Summet, et al. [27] proposed an interesting context for CS1 centered around programming robot strategies: their students implemented various robot behaviors, using physical robot devices” (Bezakova 18). Presenting the groundwork of her colleagues allows Bezakova to then expand on that knowledge and present her games as an improvement to an already existing teaching strategy. Overall, references add significant value to the paper by demonstrating that it is not simply an experience report, but a careful study and valuable addition to existing knowledge. Diba Mirza considers citations to be one of the primary characteristics of a successful paper. According to her, a good publication would have a well-defined place in the existing

Gatin 3

space of ideas around computer science education. This goal is achieved by connecting a paper to previously conducted research or other related works. Mirza adds that the use of references and citations is one of the currently used criteria to select papers for computer science conferences (personal communication, February 6, 2019). Citing previous works allows scholars to view a paper not as a standalone, but as an extension to existing knowledge, presenting groundwork on the topic and what was done to deepen that knowledge. References also benefit researchers in finding the most influential works on the topic and discovering the ways in which those works influenced succeeding research. However, references would not make experiments successful on their own. Observation shows that successful papers are also more likely to demonstrate the context of their experiments and results. The reason for this case is that in Computer Science education, one cannot just apply results from a study to a different environment without adjustments, says Mirza (personal communication, February 6, 2019). The context of an experiment matters, and its acknowledgment helps determine whether results of a publication could be repeated or applied in a different classroom. A mention of the context an experiment was run in and analysis of variables that could have influenced the results have also become a convention among publications in the discipline and could be found in works of a wide range of topics. Paul Dickson’s paper on the use of undergraduate teaching assistants is a good example of how one could acknowledge the context of a study and use that to predict the results of a similar future experiment. Dickson, professor at Hampshire College, compares the results of using undergraduate tutors in computer science classrooms among very different schools, including Brown, Stanford, and UC Irvine. Each of those institutions is unique in terms of student body, the number of students, and size of computer science departments, so similar

Gatin 4

outcomes of using the tutoring program in all of those schools allow Dickson to conclude that neither of those parameters determines the result. He ends his essay by mentioning that according to data from different schools in the nation and their successes with the tutoring program, any university can benefit from adding the program to its curriculum (Dickson 79). Here, Dickson acknowledged the context of his experiments by stating how diverse his pool of institutions was, providing ground to suggest that the outcome is systematic. A discussion about the validity of the results and potential variables that might have influenced the outcome is required for the audience to determine whether their own community would be able to benefit from a research and its results. Such attention to context, according to Mirza, significantly improves the quality of a paper, and, with conferences becoming more and more selective in their choice of publications, has become another convention among published works in computer science education (personal communication, February 6, 2019). However, even educational theories that do recognize their contextuality are worthless if they don’t have a clear positive effect in the classroom. Another convention that emerged from a recent stagnation in computer science reminds that CS education is based around students. In the publication titled “Reversing the Landslide in Computer-Related Degree Programs”, Irma Becerra-Fernandez explored the decline in interest in computer science as a specialty. Her data shows that the number of Bachelor’s degrees awarded was down more than 15% in 2006, following a 13% decrease in 2005. More alarming is the fact that the decline in enrollment in Computer Science was not as steep – students were simply losing interest in the major. (Becerra-Fernandez 128) This trend resulted not only in rapid developments in education designed to offset this change, but also in a new popular writing convention emerging in CS education. With data showing stagnation in enrollment and decrease in retention rates for

Gatin 5

Computer Science students, educators sought to reflect on student satisfaction in their publications to increase their validity. Even today, in a new wave of interest towards the field, this writing feature continues to exist to show how a program or teaching strategy will increase students’ satisfaction with a course. Joseph Cottam’s paper on undergraduate teaching assistants provides an example of how such feedback might be collected and used in a publication: “Peer tutoring has been observed to improve both retention of individuals in the computer science program and student performance in individual classes” (Cottam 218). Here, student retention, a form of student feedback, was placed at the beginning of the results section, demonstrating its importance. Such placement in a paper from 2011 provides evidence to suggest that, at the time, a question of student retention in Computer Science was still somewhat urgent among academics. However, student feedback is widely used in papers even after the “retention crisis”. For example, Adrienne Decker’s paper from 2019 contains student feedback collected on a much deeper level: “note that the median line for the UTA semesters appears near the top of the box indicating a higher trend of satisfaction with the UTA performance” (Decker 50). Here, Decker has used quantitative data from surveys to determine students’ level of satisfaction with a particular program in her CS department. These results play a big role in Decker’s argument in favor of using the said program, suggesting that overwhelmingly positive student feedback is expected to substantiate her argument. Diba Mirza also agreed that student feedback is an important part of a publication and confirmed that educators look at that feedback to decide whether a program is worth implementing in their own departments (personal communication, February 6, 2019). Computer science education is a vital part of computer science. Despite the discipline not being centered around writing, CS education produces countless publications on how to be a

Gatin 6

better teacher, influencing future generations of scientists. Members of this academic community have, over the years, created writing conventions that make it easier to convey information in computer science education during seminars, conferences, or forums. Referencing previous research, acknowledging the context of a particular study, and paying attention to student feedback have become such conventions and can be found in almost any modern paper on CS education. Together with many more, these strategies are designed to help educators share knowledge more effectively and prepare new generations of computer scientists for their exciting futures.

Gatin 7

Works Cited

Becerra-Fernandez, Irma, et al. “Reversing the Landslide in Computer-Related Degree Programs.” Communications of the ACM, vol. 53, no. 2, 2010, p. 127., doi:10.1145/1646353.1646387. Bezakova, Ivona, et al. “Board Game Strategies in Introductory Computer Science.” Proceeding of the 44th ACM Technical Symposium on Computer Science Education - SIGCSE '13, 2013, doi:10.1145/2445196.2445210. Cottam, Joseph A., et al. “Tutoring for Retention.” Proceedings of the 42nd ACM Technical Symposium on Computer Science Education - SIGCSE '11, 2011, doi:10.1145/1953163.1953227. Decker, Adrienne, et al. “Through the Looking Glass.” ACM SIGCSE Bulletin, vol. 38, no. 1, 2006, p. 46., doi:10.1145/1124706.1121358. Dickson, Paul E. “Using Undergraduate Teaching Assistants in a Small College Environment.” Proceedings of the 42nd ACM Technical Symposium on Computer Science Education SIGCSE '11, 2011, doi:10.1145/1953163.1953187.

Mirza, Diba. Personal Interview. 6 Feb. 2019.