[Paper Review] Trust in Data Science: Collaboration, Translation, and Accountability in Corporate Data Science Projects

Samir Passi, Steven J. Jackson|arXiv (Cornell University)|Feb 9, 2020

Research Data Management PracticesComputer Science96 references62 citations

TL;DR

The paper analyzes how trust is built and negotiated in real-world corporate data science work, emphasizing collaboration, translation, and accountability amid conflicting data and models.

ABSTRACT

The trustworthiness of data science systems in applied and real-world settings emerges from the resolution of specific tensions through situated, pragmatic, and ongoing forms of work. Drawing on research in CSCW, critical data studies, and history and sociology of science, and six months of immersive ethnographic fieldwork with a corporate data science team, we describe four common tensions in applied data science work: (un)equivocal numbers, (counter)intuitive knowledge, (in)credible data, and (in)scrutable models. We show how organizational actors establish and re-negotiate trust under messy and uncertain analytic conditions through practices of skepticism, assessment, and credibility. Highlighting the collaborative and heterogeneous nature of real-world data science, we show how the management of trust in applied corporate data science settings depends not only on pre-processing and quantification, but also on negotiation and translation. We conclude by discussing the implications of our findings for data science research and practice, both within and beyond CSCW.

Motivation & Objective

• Investigate how trustworthiness emerges in applied data science within corporate teams.
• Identify the tensions that challenge trust: equivocal numbers, counterintuitive knowledge, credible data, and inscrutable models.
• Examine the practices of skepticism, assessment, and credibility in negotiating trust.
• Highlight the role of collaboration and translation in managing trust in messy analytic contexts.

Proposed method

• Ethnographic fieldwork with a corporate data science team over six months.
• Theoretical framing from CSCW, critical data studies, and history/sociology of science.
• Qualitative analysis of distributed, practical work around data and models.
• Examination of everyday trust practices—skepticism, assessment, and credibility—within organizational settings.

Experimental results

Research questions

• RQ1 What tensions (numbers, knowledge, data credibility, model opacity) affect trust in corporate data science?
• RQ2 How do organizational actors establish and renegotiate trust under uncertain analytic conditions?
• RQ3 What practices (skepticism, assessment, credibility) support or hinder trust during data science work?
• RQ4 How do collaboration and translation influence trust management in corporate data science projects?

Key findings

• Trustworthiness in applied data science emerges from ongoing, situated work that resolves tensions through practical negotiation.
• Trust management depends on more than data preprocessing and quantification; it relies on collaborative processes and translation across actors.
• Skepticism, assessment, and credibility practices are central to re-negotiating trust under messy analytic conditions.
• Collaborative and heterogeneous nature of data science is a key resource for building trust in corporate settings.