Search for: All records

Background and Context.  Computing is considered a fundamental skill for civic engagement, self-expression, and employment opportunity. Despite this, there exist significant equity gaps in post-secondary computing enrollment and retention. Specifically, in the California State University (CSU) system, which serves close to half a million undergraduate students, students identifying as Hispanic/Latino make up a smaller percentage of CS majors than expected from the state’s overall population; and, once enrolled, tend to leave the CS major at higher rates than other students. Purpose.  We report on the impacts of a curricular intervention aimed at strengthening the sense of belonging of Hispanic/Latino students in computing, with the eventual goal of improving retention in computing majors for those students. Methods.  Working in an alliance of six universities within the CSU (five of which are designated as Hispanic-Serving Institutions), we have incorporated socially responsible computing across early CS courses. We aim for alignment between our curriculum and students’ communal goal orientations, and for coursework that attends to students’ interests, values, and cultural assets. Over a two-year-long study, we collected survey data to learn about the impact of our curricular intervention on students’ sense of belonging and perceived learning and agency. Findings.  We found that students generally reported high communal goal orientations and, at the campuseswithoutcompetitive enrollment policies, our intervention had a significant positive impact on students’ senses of belonging. This effect was observed between control and treatment terms as well as within treatment terms. We also note that Hispanic/Latino students were more likely than other students to report that non-curricular factors like work and family obligations interfered with their learning, and appeared to experience slightly stronger benefits from the intervention. Implications.  Our data suggest positive outcomes for integrating socially responsible computing into early CS courses, especially for Hispanic/Latino students at certain Primarily Undergraduate Institutions (PUIs). Unlike much prior research, we found that conducting studies outside of Primarily White Institutions (PWIs) can provide new insights into the impact of curricular interventions on student experience and retention. Our varying results by campus suggest that factors such as campus population, acceptance rate, and departmental enrollment policies ought to also be taken into account in studies that aim to broaden participation in computing. Would results from prior research on recruitment and retention of Hispanic/Latino students or other underrepresented students look different if such studies were replicated at institutions with different demographics and enrollment policies? 

Abstract Background Alignment-free methods for sequence comparisons have become popular in many bioinformatics applications, specifically in the estimation of sequence similarity measures to construct phylogenetic trees. Recently, the average common substring measure, ACS , and its k -mismatch counterpart, ACS k , have been shown to produce results as effective as multiple-sequence alignment based methods for reconstruction of phylogeny trees. Since computing ACS k takes O ( n log k n ) time and hence impractical for large datasets, multiple heuristics that can approximate ACS k have been introduced. Results In this paper, we present a novel linear-time heuristic to approximate ACS k , which is faster than computing the exact ACS k while being closer to the exact ACS k values compared to previously published linear-time greedy heuristics. Using four real datasets, containing both DNA and protein sequences, we evaluate our algorithm in terms of accuracy, runtime and demonstrate its applicability for phylogeny reconstruction. Our algorithm provides better accuracy than previously published heuristic methods, while being comparable in its applications to phylogeny reconstruction. Conclusions Our method produces a better approximation for ACS k and is applicable for the alignment-free comparison of biological sequences at highly competitive speed. The algorithm is implemented in Rust programming language and the source code is available at https://github.com/srirampc/adyar-rs .