Connect with us

SNAP Benefits | Supplemental Nutrition Assistance Program

Online SNAP Ads Algorithm Fails to Reach Spanish Speakers Effectively



Online SNAP Ads Algorithm Fails to Reach Spanish Speakers Effectively

Online SNAP Ads Algorithm Fails to Reach Spanish Speakers Effectively.A team of researchers led by Cornell University has uncovered a significant issue with the algorithm used by Google Ads in delivering online ads related to SNAP (Supplemental Nutrition Assistance Program), formerly known as food stamps.

Online SNAP Ads Algorithm Fails to Reach Spanish Speakers Effectively

This issue has led to increased advertising costs for reaching Spanish-speaking individuals interested in SNAP benefits. In conjunction with a survey showing broad support for a fairer approach to promoting SNAP, these findings prompted changes to target Spanish-speaking Californians more effectively, aiming to assist them in obtaining food assistance.

See also  New SNAP Payment System September 2023 Food Stamp Changes Guide

Challenges in SNAP Advertising Strategy

The algorithm powering Google Ads, when optimized for maximum SNAP enrollments per dollar spent, was found to disproportionately neglect Spanish-speaking audiences. The cost of delivering ads to Spanish speakers was significantly higher than that for English speakers, nearly four times as expensive. Although the precise reason behind this disparity remains elusive, it underscores the challenges posed by algorithms in the realm of online advertising.

Ethical Dilemma for GetCalFresh

This revelation presented GetCalFresh, the platform through which Californians apply for SNAP benefits, with a profound ethical dilemma. Should they prioritize cost-efficiency and reach as many applicants as possible, even at the expense of Spanish-speaking applicants?

See also  SNAP Benefits in Georgia: Payment Schedule and Eligible Purchases

Or should they allocate more resources to target Spanish speakers, albeit resulting in fewer total applicants? These trade-offs highlight the ethical quandaries inherent in algorithmic systems.

Balancing Efficiency and Equity

The study conducted by Allison Koenecke and her collaborators engaged the public’s opinion on how to strike a balance between efficiency and equity in advertising SNAP benefits. Remarkably, respondents from diverse backgrounds, including age, gender, race, welfare status, and political affiliation, expressed a preference for reducing total enrollments in favor of facilitating more Spanish-speaking enrollments. This preference reflects a growing recognition of the importance of equitable access.

The Call for Transparency and Community Involvement

Koenecke research underscores the need for greater transparency and community involvement in shaping algorithmic systems. Algorithms, even seemingly benign ones underpinning advertising platforms, can inadvertently exacerbate inequality or diverge from societal preferences.

See also  Do You Know How to Get a SNAP Monthly Income of $1,090 Through SNAP?

Consequently, she advocates for productive dialogues about the metrics used in these systems, with a focus on empowering the communities most affected by algorithmic decisions.

Impact and Future Actions

In response to these findings, Code for America adjusted its online advertising strategy to directly target Spanish-speaking prospective applicants.

This shift underscores the potential for positive change when addressing algorithmic biases. The research was made possible through funding from the National Science Foundation and Stanford University, highlighting the importance of continued research into the fairness and equity of algorithmic systems.

Collaborators and Acknowledgments

The research paper, titled “Popular Support for Balancing Equity and Efficiency in Resource Allocation: A Case Study in Online Advertising to Increase Welfare Program Awareness,” was authored by Allison Koenecke, Eric Giannella of Code for America, Robb Willer of Stanford University, and Sharad Goel of Harvard University’s Kennedy School. This research received partial funding from the National Science Foundation and Stanford University.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *