Generations and Gender Survey 2020 Uruguay Wave 1

Mixed mode ● Face-to-Face (personal interview).
● Self-administered - CAWI (10% of the initial sample was allocated to surveys using the CAWI format)

Tablets

1. Dealing with non-response: Send letters to high income areas of Montevideo, supervisors visits to houses that rejected to respond, fieldwork trips in groups of about 8 interviewers and one supervisor, phone calls to ask for permission to enter buildings with lots of security. 1.1. Screening: No 1.2. Refusal conversion: Yes, we have a specific letter to give to those respondents that reject to answer, we suggest them to call to the office so they can talk with a supervisor and we suggest to check interviewers identities in our webpage, we offer them different times and days, if they were selected to answer face to face sometimes we offer to answer via web mode. 1.3. Incentives: Prize drawing

2. Tracking of sampled units 2.1. Respondent contact information: Yes, we collect contact information of every completed interview. 2.2. Other contact information: No 2.3. Cards: No. 2.4. Additional surveys: No 2.5. Administrative records: No

1. Questionnaire localization 1.1. Validation : The translation was based on an initial translation to Spanish made by a GGS team in Buenos Aires, Argentina. In Uruguay, we adapted some words from Argentinean Spanish to Uruguayan Spanish and tested the comprehensibility of the formulation of the questions in a pilot questionnaire that we applied before starting the fieldwork. 1.2. Pre-test : Yes pre-test with normal survey rules (pilot). 1.3. Pilot : September 2021, 100 cases in capital 1.4. Deviations from the baseline questionnaire : The GGP baseline questionnaire was been modified, added questionnaire modules including country-specific questions 1.5. Interview lenght : face to face: 1:39; web: 1:46 1.6. Breaks-off : 428 interrupted interviews: 175 face to face, 253 web

2. Interviewers 2.1. Total interviewers: 222 interviewers 2.2. Total interviewers in the field: Max number of interviewers: 90; Min number of interviewers: 50 2.3. Network Organization: Interviewers network was centralized in Montevideo, with 4 supervisors. In the last 2 months we worked with a regional supervisor in the north coast 2.4. Working arrangement of Interviewers: Contracted - other 2.5. Payment of interviewers: Per interview

3. Interviewer training 3.1. General Interviewer training: Interviewers participate in a training of 6 hours divided in two parts: one about GGS, the questionnaire and fieldwork methodology, the other about how device works and to make a trial of interview 3.2. Survey Specific training: Yes, they were trained about GGS specific questions (that differ from typical household interviews), and about GGP 3.3. Length of training: 3 hours each 3.4. Control of Performance (i.e. testing): We called 10% of completed interviews (balanced among interviewers) and check consistency of filled out questionnaires 3.5. Interviewer Survey: No

4. Contact protocols 4.1 Advance Letter Information: We sent 1000 letters at the begining of the fieldwork only to buildings in Montevideo. The letter has a brief introduction about the survey, and contact information 4.2 Cold Contacts: Face to face, we have no phones at all, only addresses 4.3 Scheduling/Scatter: We define a three visit protocol, interviewers have to visit each address three times before giving up on it. The three visits have to be on different days (weekdays and weekend) and at different times. 4.4 Contact History: Yes, it was recorded in individual sheets (one per interviewer). We attache it, it can have missings because it depends entirely on interviewers. 4.5 Min number of contacts: There were a minimum of visits, it was three. Each visit doesn't represent a contact since we consider a visit is done even when there's nobody in a house. 4.6 Max number of contacts: The maximum number of visits were three also.

5. Noteworthy characteristics of the data collection situation: It was hard to get the interview in the first visit, that slow down fieldwork because interviewers have to visit each house more than once. There were some specific regions more difficult to interview. In high income areas of Montevideo it was difficult to convince people to get interviewed, in specific localities of less than 20,000 inhabitants (Sarandi Grande and Sarandi del Yi) it was more difficult to hire people that live there, in Melo and Maldonado (cities of more than 20,000 inhabitants) we also have difficulties to hire people. Because of the sample size we have to use the last census that was done in 2011, so some addresses to contact people were old and make it harder to find those respondents.

6. Context: The Covid-19 pandemic The health emergency was declared on March 13, 2020. During the whole of 2020 there were few infections. The first wave started at the end of November and peaked during March and April 2021, when about 5,000 deaths were recorded. There were no specific measures, except for the recommendation to use facemasks in enclosed spaces with many people and to keep a physical distance as far as possible. We believe that COVID-19 negatively impacted the response rate in the face-to-face surveys. We asked the questions recommended by GGP