The LIVES project logical framework requires that we conduct rigorous impact evaluation at the end of the project life. To this end, in the past few months, many of the LIVES project team members were out in the field to collect household level baseline survey data. The surveys made use of Computer Assisted Personal Interviewer (CAPI) technology to administer the questionnaires. A total of 80 people were involved in conducting the surveys that included well-trained enumerators who had good computer skills and knowledge of local cultures and languages. The actual field work took about two months. The data collected will also be used to conduct quantitative diagnostic analysis during the lifetime of the project.
The survey work was led by Dr. Berhanu Gebremedhin, the LIVES project Research Coordinator and senior scientist-agricultural economist. Professional inputs on research design, and instrument development were solicited from research staff in the project and partners from ILRI, IWMI and ICARDA. Planning and preparation for this extensive survey took more than four months.
The survey was conducted in all 31 project districts in the four regions of Ethiopia (Tigray, Amhara, Oromia and SNNP). Data were collected from 5,000 households, selected through stratified random sampling, and 500 community level focus group discussions to supplement the household interviews. The sites represented various agro-ecological zones and commodity potentials.
Imported specifically for the baseline survey, the 70 CAPIs facilitated data collection and analysis by reducing time spent on interviews, considerably minimizing human error and inconsistencies, and eliminating paper weight. Use of CAPIs was quite new to the project, although in the past there have been some experiences on the use of Open Data Kit (ODK) in ILRI for simpler data sets.
Once the survey instrument was developed, pre-testing in three of the LIVES zones (Sidama, East Shoa and West Shoa) was done. Necessary adaptations were included. A six-day training and review session was organized for 70 enumerators and supervisors as well as concerned LIVES staff located at regions and headquarters. The group reviewed the survey instrument, and also commented on the required adaptations and additions to the questionnaire. Also, enumerators and supervisors practiced data entry on the CAPIs. After the training, the baseline teams were dispatched to the sample peasant associations in the 10 zones.
After two months in the field, the team came back with tons of experience, challenges, great stories and lessons in addition to the baseline data. The team pointed out that for a successful survey of such magnitude, enumerators had to have data entry skills as they need to focus more on the interviewee and asking the questions, and less on figuring out how to enter data on the CAPI. This created smooth communications between the two. The team also shared their experience with gender issues where in the north of the country they came across a male household head who threatened to discontinue the interview when asked how much of a certain crop was consumed by that household. This individual was offended by this question and only agreed to take a wild guess after he was convinced of why he was asked that question. The need to go for a gender inclusive approach for quality data even if it requires additional resources is apparent.
Contributed by LIVES baseline survey team