Annex: Village research methodology
Our goal in these detailed case studies was to estimate sources and types of cash income in one relatively prosperous and in one relatively poor rural area. Another was to understand the kind of jobs that have emerged in post-war Kosovo, the prospects for job creation for the rural population, and the size, nature and impact of migration. In the course of the research, changing household structures came ever more sharply into focus as central to understanding the plight of the countryside. To obtain hard data, a house-by-house survey was undertaken in two villages, in close cooperation with members of the respective village councils. The survey was launched in spring 2004, and analysed and completed in 2005.
The ESI questionnaire had 44 questions (see below). Both villages were visited several times by teams of ESI researchers. In Cerrce and Lubishte, detailed maps of the village were drawn, and data was collected house-by-house.
In early 2005, the responses were analysed for consistency, and used to identify additional research questions. ESI researchers then visited many individual families and the village leaders, making corrections to the survey data and collecting additional qualitative information. The data was presented to local leaders to check for obvious mistakes and to obtain feedback. Co-operation from the villagers was high. In the case of Lubishte, the leader of the village council and an activist visited each family to support them in completing the forms. In the case of Cerrce, the village leader called a session of the village council and distributed the forms to the representative of each neighbourhood (mahalla). Nobody was paid for participating in the research, and there were no additional motivating mechanisms beyond the moral influence of village council members.
Obtaining answers to 44 questions for over 4,000 individuals (1,980 in Cerrce and 2,134 in Lubishte) was obviously a challenge. Where a complete household was absent, neighbours and members of the village council would supply basic information (number of household members, names, place of residence).
There are gaps in data where entire families were abroad. In Cerrce, we did not obtain the gender of 27 people, the first name of 109, the birthplace of 142 or the education level of 534. In Lubishte, on the other hand, much less information was missing: the age was missing in only 3 cases, while 202 chose not to answer the question on their level of education. Contradictions were checked and questions asked again. One reason for missing data was the absence of whole families, for which people did not know all answers. In most cases, the gaps in the data are not statistically significant. The only area where full information was not forthcoming was on the level of regular remittances, which some households were reluctant to disclose.
The questionnaire had 44 questions:
Questions related to the family relations, names and surnames, birth place, age and place of living:
Questions related to education, employment and local incomes:
Questions related to migration:
Questions related to transfers and remittances from diaspora (and IC):
Questions related to property, ownership, infrastructure: