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

Concepts de base

Assessing the impact of broadening participation efforts requires a multifaceted approach beyond traditional metrics.

Résumé

The content delves into the limitations of using standard analysis methods to evaluate the impact of broadening participation efforts in computing. It emphasizes the importance of cohort-based analysis, intersectional data reporting, and considering university demographic context. Recommendations are made for more comprehensive quantitative data analysis to truly understand the challenges and successes of broadening participation activities.
1. Introduction:

Concerns about representation in computing.
Limitations of common data analysis methods.
Need for a new framework to evaluate longitudinal change.
2. Cohort Analysis:

Examining diversity based on degrees earned by subgroups.
Standard analysis vs. longitudinal trends.
Importance of cohort-based data for accurate assessment.
3. University Context:

Influence of computing departments on student demographics.
Opportunity gap assessment within universities.
Importance of reporting intersectional data.
4. Entropy-Based Diversity Metrics:

Use of entropy-based metrics for measuring diversity.
Comparison across institutions using Jensen-Shannon divergence.
Strengths and weaknesses of entropy-based measures.
5. Conclusion:

Recommendations for quantitative data analysis.
Tracking intersectional demographics for better inclusivity.
Importance of qualitative analysis alongside quantitative data.

Stats

"In 2020, 21% of CS degrees were awarded to women."
"In contrast, 6% of graduates from university identified as Hispanic men and 9% as Hispanic Women."

Citations

"We argue that, in order to truly assess the effectiveness...we should use multiple data analysis methods."
"Graduation data is not sufficient for monitoring the impact...particularly on the introductory sequence."

Idées clés tirées de

Visualizing Progress in Broadening Participation in Computing

by Vale... à arxiv.org 03-25-2024

https://arxiv.org/pdf/2403.14708.pdf

Questions plus approfondies

How can universities improve access to demographic data for computing departments?

One way universities can enhance access to demographic data for computing departments is by establishing centralized systems that collect and organize this information. By creating a standardized process for gathering and sharing demographic data, universities can ensure that computing departments have easy access to the necessary information. Additionally, implementing protocols for regular reporting and updates on student demographics can help keep the data current and relevant.
Furthermore, universities could invest in technology solutions that streamline the collection and analysis of demographic data. Utilizing advanced analytics tools or software specifically designed for tracking student demographics can simplify the process and provide more comprehensive insights into trends and patterns within computing programs.
Collaboration between different university departments such as Institutional Research, IT Services, and Computing Departments is also crucial. By working together, these entities can establish efficient channels for sharing demographic data while ensuring compliance with privacy regulations and ethical standards.

How are potential drawbacks or biases when using entropy-based metrics for diversity evaluation?

While entropy-based metrics like Shannon Equitability Index (EH) offer valuable insights into diversity by measuring evenness in population distribution, they come with certain limitations. One drawback is their tendency to overlook contextual factors such as institutional mission or specific program goals when assessing diversity. This oversight may lead to misinterpretations of results, especially when comparing institutions with distinct objectives or student populations.
Another potential bias arises from treating all sub-populations equally without considering historical disparities or systemic inequalities present in society. EH calculations may not account for underlying issues affecting certain groups' representation in computing programs due to past discrimination or lack of opportunities.
Moreover, relying solely on entropy-based metrics might oversimplify complex diversity dynamics within computing departments. These measures often fail to capture intersectional identities adequately, disregarding nuances related to gender identity, race/ethnicity combinations, socioeconomic backgrounds, among others.

How can qualitative analysis complement quantitative assessments in understanding student experiences?

Qualitative analysis plays a vital role in providing depth and context to quantitative assessments of student experiences within computing programs. While quantitative data offers numerical insights into demographics and outcomes like graduation rates or enrollment figures; qualitative analysis delves deeper into understanding students' perceptions, motivations, challenges faced during their academic journey.
By conducting interviews, focus groups discussions surveys; researchers gain rich narratives about students' lived experiences which quantitative methods alone cannot capture fully.
Qualitative research helps uncover nuanced aspects such as feelings of belongingness/inclusion/exclusion among underrepresented groups; barriers encountered accessing resources/support networks; impact of curriculum design/teaching methodologies on learning outcomes - shedding light on areas needing improvement/enhancement.
Combining both approaches provides a holistic view enabling educators/administrators make informed decisions tailored towards fostering an inclusive environment promoting success across diverse student body segments.