Generations and Gender Survey Georgia Wave 1 & Wave 2

Method: Face-to-Face (personal interview) Technique: Paper and pencil (PAPI)

WAVE 1 ACTIONS

1. Dealing with nonresponse 1.1 Screening: Yes, as a mini-census of population in selected PSU has been conducted. 1.2 Refusal conversion: Majority of the interviewers were working in the regions where they permanently live and were able to establish good contacts with people due to the knowledge of local pecularities and attitudes between people. This knowledge made easier for them to convince potential respondents for interview. 1.3 Incentives: None.

2. Tracking of sampled units 2.1 Respondent contact information: Yes, contact details mainly fix and mobile telephone numbers of the respondent were collected for following contacts. 2.2 Other contact information: Yes, contact details of any person who we may apply later were collected. 2.3 Cards: Only a brochure with results of Wave 1 was sent. 2.4 Additional surveys: No. 2.5 Administrative records: No

WAVE 2 ACTIONS: metadata are forthcoming

WAVE 1 DATA COLLECTION

1. Interviewers 1.1 Total number of interviewers: 131. 1.2 Number of interviewers in the field: Each interviewer had a fixed number of interviews to carry out. Some of them could finish the work earlier than others. 1.3 Network organization: Field coordinators - 10 regional coordinators in all administrative regions of Georgia, aside of Tbilisi, coordination of fieldwork in Tbilisi, capital city - by project team. 1.4 Working arrangement of interviewers: Fully contracted for period of fieldwork people, mostly professional interviewers, some specially trained and fully employed. 1.5 Payment of interviewers: Per interview, some additional benefits for those working in montain areas and areas with poor transport communication.

2. Interviewer training: 2.1 General interviewing: No. 2.2 Survey specific: Interviewers in Tbilisi, capital city, were mostly the experienced interviewers, so one-week of training in GGS-specific issues has been conducted for them with home works during the training course and exams at the end. In regions, two stages-training has been implemented. At the first stage, interviewers were trained by regional supervisors, then the project team visited each region with purpose of organizing of exams for regional interviewers, and for those successfully passed this exam additional two-day training with home works and experimental interviewing. All regional supervisors were trained in advance in Tbilisi, provided with instruction of procedure of training at place. Groups of interviewers were designed with taking into account their experience. People having the experience of dealing with the economic surveys, as interviewers of household surveys of Georgian statistical office were trained by distinctive program than those worked before mainly on sociological surveys. 2.3 Length: Duration of each training session was one week, some interviewers who did not successfully pass exams were invited repeatedly to attend part of the training session on definite questionnaire chapters with the next group 2.4 Control of performance: Yes, 3 respondents in each PSU consisted 25 interviewed households have been visited for control. In case of problems found all addresses were visited for control. All filled questionnaires before the data-entry passed the procedure of logical control by the special group of trained staff. 2.5 Interviewer survey: Permanent discussions with interviewers and supervisors during fieldwork allowed us to adjust the operating procedures during the fieldwork.

3. Contact protocols 3.1 Advance letter: Household visit during micro-census of population in selected PSU several months before survey. Interviewers were required to mark any family members unable to be interviewed due to health conditions. 3.2 Cold contacts: During micro-census of population in selected PSU several months prior to survey using face-to-face contact with any family member where respondent had chance to be chosen for survey. 3.3 Scheduling / scattering: The scheduling procedure was up to interviewers and it was different in urban and rural areas and dependent on respondent's employment schedule and his/her personal preferences. Interviewers in big cities were advised not to call to potential respondents in advance, as pilot survey has shown that probability of refusal is higher in this case, while if potential respondents meet with interviewers they are more likely to do a favour and accept interviewing, or in case of time-limits for the respondent to make an appointment for another time. 3.4 Contact history: No, as interviewers were allowed to replace the potential respondent. However, regional supervisors were responsible for monitoring of all cases of replacement and proper following the procedure of replacement. 3.5 Min number of contacts: 3 visits. 2.6 Max number of contacts: NA.

4. Questionnaire localization 4.1 Validation: Re-translation was used. 4.2 Pre-test: A pre-test was carried out in September 2004 (150 respondents). The design effect characteristics and response rates found by the pilot study confirmed that proposed sampling design is sufficient to meet the GGP requirement. Furthermore, it was found that some questions in the Core Questionnaire needed to be modified and reformulated. 4.3 Length of interview: Average length of interview: 71.1 minutes.

WAVE 2 DATA COLLECTION

• Total number of interviewers: 165
• Network organization: 10 regional supervisors in all administrative regions of Georgia, aside of Tbilisi, coordination of fieldwork in Tbilisi, capital city - by project team.
• Control of performance: Three stages of control:

1. Controllers have visited 8 randomly chosen addresses in each of 400 PSU (Nmax=25) provided by GCPR staff – 3200 addresses in total.
2. Interviewers having at least one problem on at least one controlled address were additionally screened. In total the second stage of control have been implemented in 36 PSU.
3. In case of significant discrepancies in age of respondents between two waves GCPR staff was making comparison of personal information on respondent and his household composition by I and II wave.