Related papers: Visualizing Progress in Broadening Participation in Computing: The Value of Context

Visualizing Progress in Broadening Participation in Computing: The Value of Context

URL: http://arxiv.org/abs/2403.14708v1
Date: Mon, 18 Mar 2024 01:12:02 GMT
Title: Visualizing Progress in Broadening Participation in Computing: The Value of Context
Authors: Valerie Barr, Carla E. Brodley, Manuel A. Pérez-Quiñones,
Abstract summary: Concerns about representation in computing within the U.S. have driven numerous activities to broaden participation. Majority of literature on broadening participation in computing reports data on gender or on race/ethnicity, omitting data on students' intersectional identities.
Score: 2.5749138817029835
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Abstract: Concerns about representation in computing within the U.S. have driven numerous activities to broaden participation. Assessment of the impact of these efforts and, indeed, a clear assessment of the actual "problem" being addressed are limited by the nature of the most common data analysis which looks at the representation of each population as a percentage of the number of students graduating with a degree in computing. This use of a single metric cannot adequately assess the impact of broadening participation efforts. First, this approach fails to account for changing demographics of the undergraduate population in terms of overall numbers and relative proportion of the Federally designated gender, race, and ethnicity groupings. A second issue is that the majority of literature on broadening participation in computing (BPC) reports data on gender or on race/ethnicity, omitting data on students' intersectional identities. This leads to an incorrect understanding of both the data and the challenges we face as a field. In this paper we present several different approaches to tracking the impact of BPC efforts. We make three recommendations: 1) cohort-based analysis should be used to accurately show student engagement in computing; 2) the field as a whole needs to adopt the norm of always reporting intersectional data; 3) university demographic context matters when looking at how well a CS department is doing to broaden participation in computing, including longitudinal analysis of university demographic shifts that impact the local demographics of computing.

Related papers

GenderBias-\emph{VL}: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing [72.0343083866144]
This paper introduces the GenderBias-emphVL benchmark to evaluate occupation-related gender bias in Large Vision-Language Models. Using our benchmark, we extensively evaluate 15 commonly used open-source LVLMs and state-of-the-art commercial APIs. Our findings reveal widespread gender biases in existing LVLMs.
arXiv Detail & Related papers (2024-06-30T05:55:15Z)
Leveraging Prototypical Representations for Mitigating Social Bias without Demographic Information [50.29934517930506]
DAFair is a novel approach to address social bias in language models. We leverage prototypical demographic texts and incorporate a regularization term during the fine-tuning process to mitigate bias.
arXiv Detail & Related papers (2024-03-14T15:58:36Z)
Assessing Generalization for Subpopulation Representative Modeling via In-Context Learning [5.439020425819001]
This study evaluates the ability of Large Language Model (LLM)-based Subpopulation Representative Models (SRMs) to generalize from empirical data. We explore generalization across response variables and demographic subgroups.
arXiv Detail & Related papers (2024-02-12T01:55:51Z)
"It Can Relate to Real Lives": Attitudes and Expectations in Justice-Centered Data Structures & Algorithms for Non-Majors [12.812284290481916]
We examine how postsecondary students of diverse gender and racial identities experience a justice-centered Data Structures and Algorithms course. Across the class, we found a significant increase in the following attitudes: computing confidence and sense of belonging. Black, Latinx, Middle Eastern and North African, Native American, and Pacific Islander students had no significant differences compared to white and Asian students.
arXiv Detail & Related papers (2023-12-19T21:56:24Z)
Simpson's Paradox and Lagging Progress in Completion Trends of Underrepresented Students in Computer Science [0.09831489366502298]
It is imperative for the Computer Science (CS) community to ensure active participation and success of students from diverse backgrounds. This work compares CS to other areas of study with respect to success of students from three underrepresented groups: Women, Black and Hispanic or Latino. Using a data-driven approach, we show that trends of success over the years for underrepresented groups in CS are lagging behind other disciplines.
arXiv Detail & Related papers (2023-11-25T01:00:23Z)
Towards affective computing that works for everyone [0.1450405446885067]
Missing diversity, equity, and inclusion elements in affective computing datasets directly affect the accuracy and fairness of emotion recognition algorithms across different groups. Our work analyzes existing affective computing datasets and highlights a disconcerting lack of diversity in current affective computing datasets regarding race, sex/gender, age, and (mental) health representation.
arXiv Detail & Related papers (2023-09-19T17:31:29Z)
Evaluating and Incentivizing Diverse Data Contributions in Collaborative Learning [89.21177894013225]
For a federated learning model to perform well, it is crucial to have a diverse and representative dataset. We show that the statistical criterion used to quantify the diversity of the data, as well as the choice of the federated learning algorithm used, has a significant effect on the resulting equilibrium. We leverage this to design simple optimal federated learning mechanisms that encourage data collectors to contribute data representative of the global population.
arXiv Detail & Related papers (2023-06-08T23:38:25Z)
Fairness meets Cross-Domain Learning: a new perspective on Models and Metrics [80.07271410743806]
We study the relationship between cross-domain learning (CD) and model fairness. We introduce a benchmark on face and medical images spanning several demographic groups as well as classification and localization tasks. Our study covers 14 CD approaches alongside three state-of-the-art fairness algorithms and shows how the former can outperform the latter.
arXiv Detail & Related papers (2023-03-25T09:34:05Z)
Investigating Participation Mechanisms in EU Code Week [68.8204255655161]
Digital competence (DC) is a broad set of skills, attitudes, and knowledge for confident, critical and use of digital technologies. The aim of the manuscript is to offer a detailed and comprehensive statistical description of Code Week's participation in the EU Member States.
arXiv Detail & Related papers (2022-05-29T19:16:03Z)
Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation [23.661509482014058]
We grapple with questions that arise along three stages of the machine learning pipeline when incorporating intersectionality as multiple demographic attributes. We advocate for supplementing domain knowledge with empirical validation when choosing which demographic attribute labels to train on. We warn against using data imbalance techniques without considering their normative implications.
arXiv Detail & Related papers (2022-05-10T01:00:52Z)
Affect Analysis in-the-wild: Valence-Arousal, Expressions, Action Units and a Unified Framework [83.21732533130846]
The paper focuses on large in-the-wild databases, i.e., Aff-Wild and Aff-Wild2. It presents the design of two classes of deep neural networks trained with these databases. A novel multi-task and holistic framework is presented which is able to jointly learn and effectively generalize and perform affect recognition.
arXiv Detail & Related papers (2021-03-29T17:36:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.