: / Analytics / Local Governance Assessment in Kyrgyzstan (2005)

Kyrgyzstan Review, 10 years ago

2.3 Field Survey Design

The field survey was commenced in two stages:
The first stage was led by the consultant who tested the performance indicators and the survey tools. When the field tests were completed, the indicators and tools were refined.
The second stage was included into the actual field survey from selected sites using data collection and analyses. It was of a wider scale and had mobilized a large amount of human resources. The data was collected using the various survey tools: in-depth interviews, FGDs and report cards. All survey tools where developed from the general comprehensive questionnaire, which assesses the performance indicators set.
The questionnaire aside, the actual survey was oriented to gather information about the work of the 36 AOs on local budgetary processes and learn different views about their effectiveness. The primary sampling was based on the National administrative division of Kyrgyzstan and data was gathered from the AOs of two provinces in the country: Chui (North) and Jalalabad (South). The survey was undertaken in those AOs which were covered by the project activities (pilot AOs) and those which were not (Non-Pilot AOs).The following sampling method was used for selection of the survey sites:
Sampling unit
Actual # (Names)
2(Chui, Jalalabad)
2 pilot (IssykAta, Suzak)
3 non-pilot (Alamedin, Sokuluk, Bazarkurgon)
34 (see attachment 4 for the list of AOs)
810 (AO leader, Staff, AK members, Citizens)
The given sampling volume of 405 respondents in each province ensures that percentage of the error would not be exceeding 5% of the responses and could be extrapolated to the general quota of the targeted population of the research - approximately 100000 in two geographical zones.
Before the field survey had actually started in selected AOs, time was spent with AO leaders to explain the objectives and arrange all logistical matters for the planned interviews and FGDs with respondents. A minimum of 9 participants for each FGD was required in order to adjust the representation scale. However in the several survey sites this minimum representation, due to the inactiveness of the AO leadership and citizens, was not obeyed. Such issues were faced in the Bazarkurgon district of Jalalabad province and the Alamedin district of Chui province (non pilot areas).
Another limitation that has influenced the organizational part of the survey was the increased security level in the region, impacting the quality of several FGDs and consequent reporting. Due to the insecure situation in neighboring Uzbekistan, the facilitators reports from the Jalalabad province where submitted a week late.
Each visited site was surveyed using the different survey tools: in-depth interviews with AO leaders, FGDs and Report cards. The focus groups consisted of citizen representatives (in pilot sites project beneficiaries), AO staff, and AK deputies. After the completion of the each FGD the respondents were given report cards. The facilitators were to provide all necessary guidance and instructions to the respondents on how to fill out the cards. Additionally the local facilitators, which were implementing a survey of the site, also completed these cards to record their external view about the capacity level of the AO to do effective budgeting.
Bellow is a summary of survey tools and categories of the respondents that participated in the assessment:
Duration (hrs)
In-depth Interview
Citizens report card
Focus Group Discussion
AO leader
Director of AO
Chief accountant, cashier, financial controller, etc. Members of Ayil Kenesh
Active citizens
The data collected from the field survey was analyzed in the following ways:
v Factual information on the number of interviews, FGDs conducted and respondents covered;
v Amount of qualitative data that matches the specific performance indicators;
v The number of completed report cards submitted;
v The number of respondents who submitted uncompleted report cards.
v Questions which repeatedly remained unanswered;
v Questions which were not well understood by the respondent;
v The excerpts which illustrate the peoples views and perceptions