“People-Centred Analyses” Report, Issue Nov. 2008

Terms of Reference - Polling Agency

Context:

The project “Strengthening evidence-based policy processes” in the former Yugoslav Republic of Macedonia aims to contribute to more effective policy-making in the country through the preparation processes of the “People-centred Analyses” report series.
The “People-centred Analyses” report series will be built around a set of recurrent topics of “social cohesion”, “quality of governance” and “ethnic cohesion” with a specific theme will be selected for each publication. 
The People-centred Analyses issue Nov. 2008 will continue to examine the trends and social inclusion aspects as initiated from its first issue of March. 2008. This issue, however, will examine on territorial disparities and the concept of regional development in the Macedonian context. The Report will be built on the results of an extended survey, an adapted set of questionnaires based on the European Quality of Life Survey (EQLS), and selected trend questions from the previous People-centred Analyses.
The envisaged survey (planned for July 2008) is a representative, questionnaire-based household survey covering a wide range of topics which will allow measurement of different dimensions of quality of life, both objective and subjective elements, and enable an accurate picture of the social situation in the country. While the standard EQLS aiming at EU level analysis and too small to allow for detailed analysis of sub-groups or below country level, the Report’s survey will extend the sample (EQLS sample of 1000 households) to about 3000 households in FYR Macedonia and include boosters of ethnic minority groups. The data obtained will be used extensively for in-depth analysis of regions and sub-groups (such as different age groups, family typologies) in FYR Macedonia.

The current project is implemented with the South East European University of Tetovo (SEE University) being the Responsible Party, who will coordinate the production of the Report, provide research support and promote independent research and collaboration with other research and think-tank institutions in promoting the concept of “People-centred Analyses”. 

Information on the previous People-centred Analyses report is available at http://www.undp.org.mk/Default.asp?where=news&id=403&start=1

Objective:
The polling agency will conduct a representative survey (and selected boosters) with the provided questionnaires and in line with the Project’s guidelines on sampling design and data entry/processing.

Duties and responsibilities:

Under the guidance of the South East European University (SEEU) and direct supervision of the Project Coordinator, the polling agency shall be responsible for the following:

•   Take part in the methodology development meeting and discussions for finalization and adaptation of the questionnaires, sampling designs and data processing. Provide inputs to the methods of in-depth data processing and indicator computation.

•    Design the survey and sampling methods, which should be approved by the SEEU and the expert team. The sampling should be based on a national and regional representative samples, with the definition of “region” in line with the official statistical regions (refered by EU’s NUTS3 region and definition of regions by the State Statistical Office) which will allow for analysis and comparison across regions.

•   Translate the final versions of questionnaire into the local languages which should be approved by the SEEU. The survey should be piloted and report to SEEU before the full scale of the survey.

•    Conduct the representative survey (and booster surveys) with the provided questionnaires and the approved sampling design. The surveys will be conducted on the basis of face-to-face interviews. The polling agency will use its own network and means - transportation and logistics – to complete the survey.

•   Construct survey database and undertake data entry of the survey results into the statistical software (SPSS).

The data entry should satisfy the following:

• Database must comprise two files—“households” and “individuals”, corresponding to parts of questionnaire. “Households” database should include all information related to households and information on the main respondents (who answer perception questions). “Individuals” database should include all and only information related to individuals.
• A unique identifier must be assigned to each household (household ID) and recorded both in “Households” and “Individuals” databases to allow linking them. A unique identifier (i.e. individual number (ID)) must be assigned for each individual within each observed household and recorded in the “Individuals” database.
• Each record in “Households” database must correspond to only one household observed. Each record in “Individuals” database must correspond to only one individual observed.
• Missed observations must be appropriately coded, using different codes for missing observation, “don’t know” and “refuse to answer” replies.
• Databases must be accompanied with codebook, showing variable titles and possible values. In addition, script of variables and value labels assignment must be provided to ease translation of database in Macedonian/English.

•    Provide the basic interpretation of results from the surveys in consultation with SEEU. Use in-house facilities to process the data and submit all analyzed results to researchers and the Project Coordinator.

•    Compute the list of basic variables provided in Annex A. All basic indicators must be generated using scripts/syntax based on guidelines from SEEU and not manual computations. During the calculation of the basic indicators the collected variables must not be modified and cases must not be deleted. Contrary, new variables must be used and cases must be filtered. The scripts used must be provided to Project Coordinator and SEEU for following comparability checks.

•    Assist the retrieval and presentation of the trend data (from EWR and People-centred Analyses surveys).

• Provide additional clarification on the survey and data upon request from SEEU.

• Provide SEEU with copies of database and all reports in electronic format (7 copies in 7 CDs).

• Participate in expert team meetings in the process to discuss the data issues and the interpretation of the data.

• Participate in the launch event of the report to provide needed clarification on the survey design and data collection.

Technical specification:

The survey should be designed to allow disaggregation of data according to geographical division, NUTS 3 level. The sample of the representative survey is estimated to be around 3000 respondents.

Boosters should be designed to allow disaggregation of data according to ethnic division, at least to the following ethnic groups: ethnic Macedonians, ethnic Albanians, ethnic Turks, ethnic Serbs, and ethnic Roma.

Submitted data and results should be presented in both SPSS and Word/Excel format.

Deliverables and timeline:

The Polling Agency shall provide SEEU with the following deliverables in accordance with the below timeline (within June to beginning of August 2008):

• Agreed and approved questionnaire in local languages and English—by 18 June
• Database of survey with codebook (in electronic copy on CD/DVD)— by 4 August
• Reports on basic indicators with scripts used for their generation— by 4 August
• Final reports, including all generated graphs, tables, maps, and other illustrations as needed— by 4 August.

Application procedures:

Applications should be sent directly to the Procurement Office, to Ms. Nermine Reshidi at SEEU, Ilindenska bb. 1200 Tetovo at 301.03 office (contact telephone 044 356 032). The package of application should include an competitive offer (total cost including VAT in MKD, separate cost of the main representative survey and the cost for boosters), a summary of experience in undertaking surveys and data processing (with details information on projects, partners, products), CV of the team leader and at least reference letters from the past partners in similar projects (preferable).  The offer should be enveloped, sealed and specified projects name (PCA project).

Payment will be made upon the satisfactory delivery of products.

Deadline for the expression of interest is 15th June 2008.

Annex A:

List of basic variables to be computed and processing steps to be undertaken by the polling agency. Additional of 4-5 variables (of similar level of complexity) to be computed and possible specification adjustment shall be specified at a later stage.

In the “households” database:  

• Number of household’s members (household size); Number of children (=< 14 years old); number of working age members (>=15); number of members participating in the labour force (according to the selected answers (given options) on employment status in the questionnaire); number of inactive members (in working age but not look for job – link to question xx); number of unemployed members (in the category of labour force); number of employed members (in the labour force); number of members with income;
• Create a new variable which specifies the highest educational attainment of the households (ie. educational attainment of the most educated members who is no longer in school);
• Income: total (monthy) income of the household (including all sources of income); and equivalized income. Equivalized income is derived by dividing the total household disposable income by the household size, calculated according to the modified OECD scale: weight 1 for the first household member, 0.5 for the second member, and 0.3 for any additional member.
• Expenditure: total (monthly) expenditure of the household (including education); and equivalized expenditure. Equivalized expenditure for each household member is derived by dividing the total household disposable expenditure by the household size, calculated according to the modified OECD scale: weight 1 for the first household member, 0.5 for the second member, and 0.3 for any additional member.
• Create another variable which specifies the level of vulnerability of each household according to income (see Methods note below);
• Create another variable which specifies the level of vulnerability of each household according to expenditure (see Methods note below)

In the “individuals” database: 

• Generate new variables which specify the age groups that the individual belongs to. The age groups are: 15-19, 20-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 60-64, “65 and more”.

• Individuals: if invididuals no longer in the school/education (question xx), then calculate the total years in school (from the answer of educational attainment, e.g.  primary education is equivalent of 8 years, secondary education: 8+4 years, high-school: 8+4+2; university: 8+4+4;
• Create another variable which specifies the level of vulnerability of each individual (see Methods note below)

Methods note: Criteria used for poverty and vulnerability assessment

The below criteria is suggested based on the internationally accepted methodologies and methodology for calculation of EU-SILC indicators.

Assessment on the level of vulnerability of each household or individual uses the following method:

• high level of vulnerability (meeting all criteria),
• medium level of vulnerability (meeting 1-2 criteria)
• and not vulnerable (the rest).