Generations and Gender Survey Estonia Wave 1

Method: Face-to-Face (personal interview) was the main method. In addition, a drop-off-mail-back questionnaire was used to collect data mainly on attitudinal questions. Return rate of the drop-off questionnaire was 80.3% of the respondents. Technique: Paper and Pencil (PAPI).

1. Dealing with nonresponse 1.1 Screening: No. 1.2 Refusal conversion: Usual techniques of refusal conversion, i.e., insisting on the importance of the survey, its international dimension, addressing specific concerns of the respondent, etc. 1.3 Incentives: No.

2. Tracking of sampled units 2.1 Respondent contact information: The addresses of the respondents were recorded. 2.2 Other contact information: Since the funding was available for a single wave, other contact information was not recorded. Respondents can be followed up by means of the population register. 2.3 Cards: No. 2.4 Additional surveys: No. 2.5 Administrative records: No.

1. Interviewers 1.1 Total number of interviewers: 130. 1.2 Number of interviewers in the field: The number varies according to the stage of fieldwork. Some interviewers quitted at an early stage of the work. At the later stages of the fieldwork the number of active interviewers decreased as more and more interviewers completed their task. 1.3 Network organization: Centralized coordination. 1.4 Working arrangement of interviewers: The network was project-based and consisted of both full- and part-time interviewers. Most of the interviewers were working or had previously worked as interviewers for Statistical Office or polling firms. They were hired by the institute on fixed term basis. 1.5 Payment of interviewers: Per interview + travel expenses.

2. Interviewer training: 2.1 General interviewing: The majority of the interviewers (nearly 90%) had previous training in interviewing either in Statistical Office or polling firms (interviewing techniques, appointments, conversion of refusals, etc.). 2.2 Survey specific: Before the fieldwork started, a series of GGS-specific training seminars was organized in several locations in Estonia. The training was focused on the GGS survey instrument (questionnaire) and procedures of the fieldwork. 2.3 Length: General training varied depending on survey organization (typically 2-3 days). The GGS-specific training was 1.5 day. 2.4 Control of performance: The quality was monitored by checking the filled-in questionnaires. The interviewers received personal feedback from the coordinator. A letter describing typical problems that were detected was circulated to all interviewers at the early stage of the fieldwork. 2.5 Interviewer survey: Yes. A brief questionnaire was distributed among the interviewers once the fieldwork was completed. It collected background information about interviewer, the contacting procedures, and experiences with respondents, comments on questionnaire modules, etc. Altogether 106 interviewers responded.

3. Contact protocols 3.1 Advance letter: All respondents received an advance letter introducing the survey and informing about the coming of the interviewer. The letter explained the purpose, subject of the survey, the institutions involved. The letter requested respondent's cooperation. 3.2 Cold contacts: Mixed practice. Usually the interviewers tried to make a telephone contact and schedule an appointment for the face-to-face interview. But this was not always feasible and part of the interviewers preferred to start with face-to-face contact. 3.3 Scheduling / scattering: Yes. To get a higher response rate, contact attempts were scattered over different days of week and parts of the day. 3.4 Contact history: Yes. The interviewer documented attempts to contact the respondent on a special form. For each case of non-response, the form was checked by the coordinator. 3.5 Min number of contacts: Number was set for the attempts of face-to-face contacts (5 for respondents in urban areas, 3 in rural areas). 2.6 Max number of contacts: No.

4. Questionnaire localization 4.1 Validation: Retranslation of the questionnaires. 4.2 Pre-test: A pilot was carried out in June 2003 (120 respondents). The pilot study confirmed the possibility to increase the number of event histories and to also expand the target population to include the population groups of foreign origin using a Russian language questionnaire. 4.3 Length of interview: Respondents who agreed for the interview were generally cooperative. Average length of interview was 99 minutes but it varied according to the complexity of life history and household composition of the respondent. The presence of another person was reported at 24% of the interviews. Difficulties were reported at 4% of the interviews (in most cases recall difficulties).