README September 2021 A. GENERAL INFORMANTION TITLE: Delivery Gig Worker Interviews on Automation at Work AUTHOR INFORMATION: Diana Enriquez Department of Sociology, Princeton University de8@princeton.edu ORCID: https://orcid.org/0000-0002-6254-5503 Alternate contact: Professor Janet Vertesi Department of Sociology, Princeton University jvertesi@princeton.edu ORCID: https://orcid.org/0000-0003-4579-6252 SUMMARY These data include 39 structured interview transcripts. Each case is someone who worked at the time for Uber, UberEats, Lyft, and/or Amazon Flex (Amazon’s contractor delivery service). These data were collected between July and September 2019. All but one of the interviews occurred over the phone. My questions are focused on the structure of their gig work jobs and the technology they used at work or expected to use at work in the future. I included a description of the data, the recruitment methods, and the discussion guide in this ReadMe file.  B. SHARING AND INFORMATION ACCESS: a. LICENSE: Creative Commons with Attribution by 4.0: https://creativecommons.org/licenses/by/4.0/ b. RECOMMENDED CITATION FOR USE: Enriquez, Diana. "Delivery Gig Worker Interviews on Automation at Work." September 2019. https://doi.org/10.34770/4324-yn77. c. These data have been used to write the following two academic articles: 1. Vertesi, Janet A., Adam Goldstein, Diana Enriquez, Larry Liu, and Katherine T. Miller. “Pre-Automation: Insourcing and Automating the Gig Economy.” Sociologica, January 29, 2021, 167-193 Pages. https://doi.org/10.6092/ISSN.1971-8853/11657. 2. Enriquez, Diana, and Janet Vertesi. “Managing Algorithms: Partial Automation of Middle Management and Its Implications for Gig Worker.” Academy of Management Proceedings 2021, no. 1 (August 1, 2021): 16560. https://doi.org/10.5465/AMBPP.2021.16560abstract. C. ABOUT THE DATA AND SAMPLE These data include 20 interviews with Amazon Flex drivers, 3 from Lyft drivers and 16 interviews with Uber and/or UberEats drivers. Each file is labeled used the first letter of the Company name (A, L, U, or UE) and a number assigned to the specific file. The interviews are 20 minutes long and structured. There are a few interviews that went past the 20 minutes and/or include additional material sent to me via email after the interview. My co-authors and I selected this set of gig workers so we could study the technology these workers were already using and their perceptions of future technology, specifically self-driving cars. We selected these companies because Amazon and Uber have spoken about their desire to introduce autonomous vehicles into their delivery systems. This is an interesting moment to capture this kind of data from the workers because they are completing some of the key services sold to consumers by these companies but they are also completing the necessary data collection and testing for future technologies.  My recruitment took three forms: I posted a recruitment ad on Craigslist in New York City and New Jersey. I did this to mimic the recruiters who hire many gig workers to this platform. I was only able to complete one interview this way. Next, I ran an ad on a budgeting app for households who receive SNAP or WIC benefits. The ad led to a brief screener asking about the type of gig work this app user did and how long they had been doing this work. From there, I called, emailed, or texted (based on the preference they marked on the screener) the respondents who fit our criteria and scheduled interviews with 27 of our interview cases. Twelve of our interviews with Amazon interviews came through Facebook recruitment ads posted in Facebook groups built and managed by Amazon flex drivers. The subjects recruited through the budgeting app were all from low-income households and the Facebook groups provided a broader range of gig workers from different income brackets.  My sample includes cases from both urban and rural areas across the United States. The budgeting app has users in every state, which meant I was able to reach gig workers from every region of the US. The further recruitment effort on Facebook provided a sample of additional Amazon drivers from a similar coverage across the United States. Of these workers, 24 of them worked in urban areas and 15 of them work in rural areas.  My interview cases included 20 female and 19 male drivers. I did not ask specific questions about race and ethnicity but based on the profile images selected by drivers I believe this sample includes a mix of Black, White, Latinx, and Asian drivers.  **We dropped one interview from this final data set because there were inconsistencies in the interviewee’s responses that gave us pause. This file was A8. It is not missing from your downloads, it was omitted. WHAT WE REMOVED FROM THE DATA TO ANONYMIZE IT  On consent for the data to be shared publicly: I explicitly asked the interview subjects if they consent to being recorded for a transcript to be produced after the interview and I asked for permission to share their anonymized data once we completed our work.  In order to anonymize the data, we had several members of our team review and redact identifying information from the transcripts. Our team removed location and specific details about their location that could be used to infer their location. We also removed especially personal information that could be used to further identify them. We also tried our best to keep as much specific information in the interview data as possible, because we wanted to preserve the richness of the stories our interviewees shared. Our data initially meant to focus on the structure of these worker’s jobs and the technology they used. Our edits to maintain the individual’s privacy does not interfere with the important questions we included in our discussion guide.  D. ACKNOWLEDGEMENTS  This project was funded by the Sloan Foundation and the Princeton School of International Affairs. Thank you to Janet Vertesi for her support for the fieldwork and edits on the discussion guide. This project was made possible through the projects envisioned by Janet Vertesi. Thank you to Katie Miller and Joseph Isaac provided support during the early, in-person phase of this data collection in New York. Thank you also to Summer Crown, who helped anonymize our final interview data. E. DISCUSSION GUIDE  [Consent form read by the interviewer to the interviewee before the interview began. The interviewee provided verbal consent before the interview began. A written document with the consent form information and contact information was also provided to their email.] INTRO TO GIG WORK ROLE: • What is your job? • How did you find this job? • What other companies do you work for? • What do you do on those platforms? • How long have you worked for (this employer)?  • What kind of introduction to the job or training did you receive for this job? OTHER JOBS CONTEXT • How many jobs do you have?  • How do you find these other jobs? • How do you structure your week between different jobs?  • How many hours do you try to work per week? • How much do you get paid an hour? • Do you have a financial goal you try to hit for the day or week? STRUCTURE OF ROLE • Who is your manager? • What do they do as their job? • Did you have to sign a contract to work for [COMPANY]? • Do you remember anything that contract said? • What skills do you need or have you developed to do this job? DATA and TECHNOLOGY • Is there any data that they record or track while you work? • Do you check in with someone on completed tasks? • What devices or tools do you use to do this job? • Have you noticed any new features or software in the app this year? • Has anything changed since you first started working for them? • How does this compare to other work that you’ve done? WORK AND CAREER • Do you consider yourself an employee, self-employed, or running a small business? Why? • Do you expect to still be doing this job in a few years? • Are there other jobs you’d like to do in the future? FUTURE OF TECH • How do you feel about self-driving cars? • How would you feel if [COMPANY] was going to start using self-driving cars in their fleet? • Does it seem like that is something [COMPANY] might do? 1