European Commission

SILC_ESQRS_A_CY_2011_0000 - Version 1

National Reference Metadata in ESS Standard for Quality Reports Structure (ESQRS)

Compiling agency: Statistical Service of Cyprus (CYSTAT)

Time Dimension: 2011-A0

Data Provider: CY1

Data Flow: SILC_ESQRS_A

For any question on data and metadata, please contact: EUROPEAN STATISTICAL DATA SUPPORT

1. Contact

Top

1.1. Contact organisation

Statistical Service of Cyprus (CYSTAT)

1.2. Contact organisation unit

Demography, Social statistics and Tourism division

1.5. Contact mail address

2. Introduction

Top

The production of quality reports is part of the implementation of the EU-SILC instrument. In order to assess the quality of data at national level and to make a comparison among countries, the National Statistics Institutes give detailed information mainly on: the entire statistical process, sampling and non-sampling errors, and potential deviations from standard definition and concepts.

This document follows the ESS standard for quality reports structure (ESQRS), which is the main report structure for reference metadata related to data quality in the European Statistical System. It is a metadata template, based on 13 main concepts, which can be used across several statistical domains with the purpose of a better harmonisation of the quality reporting requirements in the ESS.

For that reason the template of this document differs from that one stated in the Commission Reg. 28/2004.

CYSTAT completed the sections of ESQRS that were also covered by the Commission Reg. 28/2004. Therefore sections such as 3, 4, 6 and 7 remained empty.

3. Quality management - assessment

Top

Not requested by Regulation 28/2004

4. Relevance

Top

4.1. Relevance - User Needs

Not requested by Reg. 28/2004

4.2. Relevance - User Satisfaction

Not requested by Reg.28/2004

4.3. Completeness

Not requested by Reg. 28/2004

4.3.1. Data completeness - rate

Not requested by Reg. 28/2004

5. Accuracy and reliability

Top

5.1. Accuracy - overall

In terms of precision requirements, the EU-SILC framework regulation as well the Commission Regulation on sampling and tracing rules refers respectively, to the effective sample size to be achieved and to representativeness of the sample. The effective sample size combines sample size and sampling design effect which depends on sampling design, population structure and non-response rate.

5.2. Sampling error

EU-SILC is a complex survey involving different sampling design in different countries. In order to harmonize and make sampling errors comparable among countries, Eurostat (with the substantial methodological support of Net-SILC2) has chosen to apply the "linearization" technique coupled with the “ultimate cluster” approach for variance estimation. Linearization is a technique based on the use of linear approximation to reduce non-linear statistics to a linear form, justified by asymptotic properties of the estimator. This technique can encompass a wide variety of indicators, including EU-SILC indicators. The "ultimate cluster" approach is a simplification consisting in calculating the variance taking into account only variation among Primary Sampling Unit (PSU) totals. This method requires first stage sampling fractions to be small which is nearly always the case. This method allows a great flexibility and simplifies the calculations of variances. It can also be generalized to calculate variance of the differences of one year to another .

The main hypothesis on which the calculations are based is that the "at risk of poverty" threshold is fixed. According to the characteristics and availability of data for different countries we have used different variables to specify strata and cluster information. In particular, countries have been split into four groups:

1)BE, BG, CZ, IE, EL, ES, FR, IT, LV, HU, NL, PL, PT, RO, SI, UK and HR whose sampling design could be assimilated to a two stage stratified type we used DB050 (primary strata) for strata specification and DB060 (Primary Sampling Unit) for cluster specification;

2) DE, EE, CY, LT, LU, AT, SK, FI, CH whose sampling design could be assimilated to a one stage stratified type we used DB050 for strata specification and DB030 (household ID) for cluster specification;

3) DK, MT, SE, IS, NO, whose sampling design could be assimilated to a simple random sampling, we used DB030 for cluster specification and no strata;

5.2.1. Sampling error - indicators

	AROPE			At risk of poverty (60%)			Severe Material Deprivation			Very low work intensity
	Ind. value	Stand. errors	Half CI (95%)	Ind. value	Stand. errors	Half CI (95%)	Ind. value	Stand. errors	Half CI (95%)	Ind. value	Stand. errors	Half CI (95%)
Total	23,5	0,85	1,66	14,5	0,65	1,28	10,7	0,70	1,36	4,5	0,36	0,71
Male	21,5	0,92	1,80	12,6	0,71	1,39	10,6	0,77	1,51	4,0	0,43	0,84
Female	25,4	0,92	1,80	16,3	0,72	1,41	10,7	0,73	1,44	5,0	0,40	0,79
Age0-17	21,8	1,54	3,03	12,0	1,22	2,40	13,5	1,34	2,64	2,8	0,53	1,03
Age18-64	20,8	0,87	1,70	11,0	0,62	1,22	10,6	0,72	1,42	5,1	0,38	0,74
Age 65+	40,4	1,45	2,84	36,9	1,42	2,79	6,0	0,67	1,30	na	na	na

5.3. Non-sampling error

Non-sampling errors are basically of 4 types:

Coverage errors: errors due to divergences existing between the target population and the sampling frame.
Measurement errors: errors that occur at the time of data collection. There are a number of sources for these errors such as the survey instrument, the information system, the interviewer and the mode of collection
Processing errors: errors in post-data-collection processes such as data entry, keying, editing and weighting
Non-response errors: errors due to an unsuccessful attempt to obtain the desired information from an eligible unit. Two main types of non-response errors are considered:

– Unit non-response: refers to absence of information of the whole units (households and/or persons) selected into the sample

– Item non-response: refers to the situation where a sample unit has been successfully enumerated, but not all required information has been obtained

5.3.1. Coverage error

Coverage errors include over-coverage, under-coverage and misclassification:

Over-coverage: relates either to wrongly classified units that are in fact out of scope, or to units that do not exist in practice
Under-coverage: refers to units not included in the sampling frame
Misclassification: refers to incorrect classification of units that belong to the target population

Sampling frame and coverage errors

The list of households from the 2001 Census of Population was used as sampling frame with a supplementary list of newly constructed houses (built after 2001 up to 2010). The Statistical Service of Cyprus was provided by the Electricity Authority of Cyprus (E.A.C.) with a list of domestic electricity consumers, which contained all the new connections of electricity between 2002 and 2010 (last update September of 2010). The E.A.C. distinguishes domestic consumers from other consumers (e.g. industrial etc). It has been established that each domestic electricity consumer registered by the E.A.C. corresponds to the statistical definition of a housing unit. Each of these new electricity meter connections represented one new household.

Coverage problems encountered were:

The frame of the 2001 Census of Population was somehow outdated and as a result some housing units were found to be empty or to be used for other purposes other than housing.
Some houses included in the E.A.C. list were used as secondary residence, so they were out of scope of the survey.
Some houses listed by the E.A.C. were impossible to be located due to incomplete information regarding their addresses.
Housing units built after September 2010, were not included in our sampling frame.

5.3.1.1. Over-coverage - rate

Main problems

Size of error

Cross sectional

data

·Over-coverage

·Under-coverage

·Misclassification

5.3.2. Measurement error

Cross sectional data

Source of measurement errors

Building process of questionnaire

Interview training

Quality control

Possible sources of measurement errors are the questionnaire (design, content and wording), the method of data collection, the interviewers and the respondents. As the 2011 EU-SILC round was the 7^th in the series, quality has considerably improved due to interviewers’ feedback, continuous data analysis and research.

The questionnaire for EU-SILC was developed on the basis of the EU-SILC Doc. 065 and Doc. 055. Even though, the questionnaire was well tested and despite the fact that this was the 7^th wave of the survey, some questions were still difficult to be answered with precision. Difficulties due to memory lapses were encountered in questions regarding income, housing cost, main activity each month as well as for the age at first job especially with older persons. In an effort to minimise these problems respondents were requested to prepare pay slips and utility bills when the interviewer was making an appointment. In the case that the respondents could have the pay slips at a later date then they could send them by fax at the central offices. Difficulties were also encountered in distinguishing the various benefits and pensions. In order to overcome these difficulties a part of the training of the interviewers was focused specifically on social benefits and pensions.

As the method of data collection was Computer Assisted Personal Interviewing (CAPI) many validation and consistency checks were implemented during the interview. This had a positive impact on the quality of the data collected. Additionally, problems usually accounted to the routing of the questionnaire were fully avoided because of CAPI.

In order to reduce interviewer effects a two week training session for all the interviewers and an extra week training for newly recruited interviewers (i.e. those working for the first time in EU-SILC), was organised at the head offices of the Statistical Service. The training was conducted by permanent staff, Statistics Officers responsible for the EU-SILC survey. The aim of the training was to ensure that all interviewers were uniformly trained both in regard to the content of the questionnaire, as well as their behaviour during the interview. The extra week training for the newcomers focused mainly on the terminology of the survey giving also general information on the previous rounds of the survey. In this way the newcomers were able to follow the other interviewers who worked the year before in the survey. In the second week where all interviewers were together, the training mainly focused on refreshing the terminology used in the questionnaire and on the understanding of new terminology used for the first time in the questionnaire (e.g.Intergenerational trasmission of disadvantages). Main emphasis was given on difficult definitions and on explaining the various public benefits as well as the importance of the accuracy of the information collected. On the third week the interviewers had intensive sessions on working with their laptops and the electronic questionnaires in the environment of BLAISE. An interviewer manual was prepared explaining each and every single question of the questionnaire as well as their respective possible answers.

Apart from the 23 interviewers the training sessions were also attended by 6 supervisors. Each one of them was responsible for a group of 3 or 4 interviewers. During the fieldwork period the supervisor had meetings with each one of the interviewers in his/her group at least once a week. During these meetings, apart from discussing problems or questions raised during the week, the supervisors also collected (from the interviewers´ laptops) all completed questionnaires. Their main duty during the data collection period was to examine the interviewers’ work and refer back to them for inconsistencies or for problems identified in connection with terminology. Furthermore the supervisors had to double check some of the answers with respondents either by telephone or by personally visiting the household in question, especially in the case of unusual answers or missing data. Additionally from 2^nd wave onwards, data for households in the survey for 2 years or more were further checked based on the data from previous years.

5.3.3. Non response error

Non-response errors are errors due to an unsuccessful attempt to obtain the desired information from an eligible unit. Two main types of non-response errors are considered:

1) Unit non-response which refers to the absence of information of the whole units (households and/or persons) selected into the sample. According the Commission Regulation 28/2004:

Household non-response rates (NRh) is computed as follows:

NRh=(1-(Ra * Rh)) * 100

Where Ra is the address contact rate defined as:

Ra= Number of address successfully contacted/Number of valid addresses selected

and Rh is the proportion of complete household interviews accepted for the database

Rh=Number of household interviews completed and accepted for database/Number of eligible households at contacted addresses

Individual non-response rates (NRp) will be computed as follows:

NRp=(1-(Rp)) * 100

Where Rp is the proportion of complete personal interviews within the households accepted for the database

Rp= Number of personal interview completed/Number of eligible individuals in the households whose interviews were completed and accepted for the database

Overall individual non-response rates (*NRp) will be computed as follows:

*NRp=(1-(Ra * Rh * Rp)) * 100

For those Members States where a sample of persons rather than a sample of households (addresses) was selected, the individual non-response rates will be calculated for ‘the selected respondent’, for all individuals aged 16 years or older and for the non-selected respondent.

2) Item non-response which refers to the situation where a sample unit has been successfully enumerated, but not all the required information has been obtained.

5.3.3.1. Unit non-response - rate

Cross sectional data
Address contact rate (Ra)*		Complete household interviews (Rh)*		Complete personal interviews (Rp)*		Household Non-response rate (NRh)*		Individual non-response rate (NRp)*		Overall individual non-response rate (NRp)*
A*	B*	A*	B*	A*	B*	A*	B*	A*	B*	A*	B*
0,9970	0,9901	0,9013	0,8304	1,00	1,00	0,1015	0,1779	0,0	0,0	0,1015	0,1779

* All the formulas are defined in the Commission Regulation 28/2004, Annex II

A* = Total sample; B = * New sub-sample

5.3.3.2. Item non-response - rate

The computation of item non-response is essential to fulfil the precision requirements concerning publication as stated in the Commission Regulation No 1982/2003. Item non-response rate is provided for the main income variables both at household and personal level.

5.3.3.2.1. Item non-response rate by indicator

	Total hh gross income (HY010)	Total disposable hh income (HY020)	Total disposable hh income before social transfers other than old-age and survivors benefits (HY022)	Total disposable hh income before all social transfers (HY023)
% of household having received an amount	100,0	100,0	99,3	91,6
% of household with missing values (before imputation)	0,0	0,0	0,0	0,0
% of household with partial information (before imputation)	2,2	0,2	0,2	0,2

	Imputed rent (HY030)	Income from rental of property or land (HY040)	Family/ Children related allowances (HY050)	Social exclusion payments not elsewhere classified (HY060)	Housing allowances (HY070)	Regular inter-hh cash transfers received (HY080)	Interest, dividends, profit from capital investments in incorporated businesses (HY090)
% of household having received an amount	90,6	8,7	52,8	0,5	3,5	9,1	13,7
% of household with missing values (before imputation)	na	0,0	0,0	0,0	0,0	0,0	0,0
% of household with partial information (before imputation)	na	0,0	0,0	0,0	0,0	0,0	0,0

	Cash or near-cash employee income (PY010)	Other non-cash employee income (PY020)	Income from private use of company car (PY021)	Employers social insurance contributions (PY030)	Cash profits or losses from self-employment (PY050)	Value of goods produced for own consumption (PY070)	Unemployment benefits (PY090)	Old-age benefits (PY100)	Survivors benefits (PY110)	Sickness benefits (PY120)	Disability benefits (PY130)	Education-related allowances (PY140
% of household having received an amount	49,7	6,4	1,0	45,9	11,4	0,8	3,8	20,3	4,6	1,1	2,7	6,1
% of household with missing values (before imputation)	0,1	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0
% of household with partial information (before imputation)	0,8	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0	0,0

5.3.4. Processing error

Data entry and coding

Editing controls

Processing errors were reduced because of CAPI and the implementation of validation and consistency checks during the data collection phase (BLAISE software). The processing errors were further reduced as the questionnaires were edited and coded by the supervisors prior to finalising the data files for processing. For the households which were in the survey for at least 2 years an additional tool during editing was the preloading of certain variables from the previous survey. Inconsistencies were further examined with interviewers and in many cases with the households directly. The coding requested was minimal, i.e. occupation (2 digits ISCO), economic activity (2 digits NACE rev. 2) and country of birth; and was carried out using drop down lists.

The finalised data files prepared by supervisors were then processed using SAS programs with various other logical and consistency checks. The main errors found were connected to self-employment income and the recording of the various benefits and pensions under the correct income variable according to EU-SILC Doc. 065.

Before sending the final D-, R-, H- and P- files, data files were further checked using EUROSTAT’s SAS programs.

5.3.4.1. Imputation - rate

Not requested by Reg. 28/2004

5.3.4.2. Common units - proportion

Not requested by Reg. 28/2004

5.3.5. Model assumption error

Not requested by Reg. 28/2004

5.3.6. Data revision

Not requested by Reg. 28/2004

5.3.6.1. Data revision - policy

Not requested by Reg. 28/2004

5.3.6.2. Data revision - practice

Not requested by Reg. 28/2004

5.3.6.3. Data revision - average size

Not requested by Reg. 28/2004

5.3.7. Seasonal adjustment

Not requested by Reg. 28/2004

6. Timeliness and punctuality

Top

6.1. Timeliness

Not requested by Reg. 28/2004

6.1.1. Time lag - first result

Not requested by Reg. 28/2004

6.1.2. Time lag - final result

Not requested by Reg. 28/2004

6.2. Punctuality

Not requested by Reg. 28/2004

6.2.1. Punctuality - delivery and publication

Not requested by Reg. 28/2004

7. Accessibility and clarity

Top

7.1. Dissemination format - News release

Not requested by Reg. 28/2004

7.2. Dissemination format - Publications

Not requested by Reg. 28/2004

7.3. Dissemination format - online database

Not requested by Reg. 28/2004

7.3.1. Data tables - consultations

Not requested by Reg. 28/2004

7.4. Dissemination format - microdata access

Not requested by Reg. 28/2004

7.5. Documentation on methodology

Not requested by Reg. 28/2004

7.5.1. Metadata completeness - rate

Not requested by Reg. 28/2004

7.5.2. Metadata - consultations

Not requested by Reg. 28/2004

7.6. Quality management - documentation

Not requested by Reg. 28/2004

7.7. Dissemination format - other

Not requested by Reg. 28/2004

8. Comparability

Top

8.1. Comparability - geographical

Not requested by Reg. 28/2004

8.1.1. Asymmetry for mirror flow statistics - coefficient

Not requested by Reg. 28/2004

8.1.2. Reference population

Reference population	Private household definition	Household membership
There is no difference to the standard EU-SILC definition, hence the reference population is defined as all the households and their members living in the areas under the effective control of the Government of the Republic of Cyprus. Population in collective households and institutions is excluded.	No deviation from the standard EU-SILC definition. A private household is a person living alone or a group of persons living together in the same dwelling sharing expenses, including the joint provision of the essentials of living.	The definition of household membership is the one recommended by EUROSTAT. Students (either in Cyprus or abroad) are considered to be members of their parents´ household given they are fully financially supported by them.

8.1.3. Reference Period

Period for taxes on income and social insurance contributions

Income reference periods used

Reference period for taxes on wealth

Lag between the income ref period and current variables

The period for taxes payments/refunds and social insurance contributions was 2010. Tax refunds received during 2010 referred to income received in previous years.

For EU-SILC 2011 the income reference period was 2010.

The reference period for taxes on wealth was 2010.

Since EU-SILC 2011 was carried out during the beginning of March and the end of July 2011, the time lag between the income reference period and current variables varied between 3 to 7 months.

8.1.4. Statistical concepts and definitions

Total hh gross income (HY010)	Total disposable hh income (HY020)	Total disposable hh income before social transfers other than old-age and survivors' benefits (HY022)	Total disposable hh income before all social transfers (HY023)
F	F	F	F

Imputed rent
(HY030)

Income from rental of property or land
(HY040)

Family/ Children related allowances
(HY050)

Social exclusion payments not elsewhere classified
(HY060)

Housing allowances
(HY070)

Regular inter-hh cash transfers received
(HY080)

Interest, dividends, profit from capital investments in incorporated businesses
(HY090)

Interest paid on mortgage
(HY100)

Income received by people aged under 16
(HY110)

Regular taxes on wealth (HY120)

Regular inter-hh transfers paid
(HY130)

Interest paid on mortgages is collected asking directly the amount. Over and above, a double check is carried out with an estimation of the amount, which is calculated on the basis of the following questions: year the housing loan was taken, the initial amount borrowed, years of repayment of the initial loan, the monthly payment, the outstanding amount at the end of the previous year, the actual total amount paid on the previous year and the interest rate applied for the loan.

Cash or near-cash employee income
(PY010)

Other non-cash employee income
(PY020)

Income from private use of company car
(PY021)

Employers social insurance contributions
(PY030)

Cash profits or losses from self-employment
(PY050)

Value of goods produced for own consumption
(PY070)

Unemployment benefits
(PY090)

Old-age benefits
(PY100)

Survivors benefits
(PY110)

Sickness benefits
(PY120)

Disability benefits
(PY130)

Education-related allowances
(PY140)

Gross monthly earnings for employees
(PY200)

Gross monthly earnings for employees were not collected as the gender pay gap is calculated from other sources than EU-SILC.

The source or procedure used for the collection of income variables	The form in which income variables at component level have been obtained	The method used for obtaining target variables in the required form
Data on income variables were collected by Computer Assisted Personal Interviewing. Each and every income component was separately collected.	The instructions to the interviewers were to collect each income component as gross and to record separately taxes on income at source and social insurance contributions. In the very few cases where gross income was impossible to collect, net income was recorded.	In the cases where gross income or taxes on income at source or social insurance contributions were impossible to collect, at least net value was collected for the specific income component. It was then converted to gross by applying the existing tax system and social insurance contributions rules.

8.2. Comparability - over time

Not requested by Reg. 28/2004

8.2.1. Length of comparable time series

Not requested by Reg. 28/2004

8.3. Comparability - domain

Not requested by Reg. 28/2004

9. Coherence

Top

9.1. Coherence - cross domain

The income results of EU-SILC 2010 were compared with the income results of the 2009 Household Budget Survey. For both surveys the income reference period was 2009. When comparing the two survey results it is essential to keep in mind the differences between the concepts and methodologies. Discrepancies may further arise by the fact that they serve different purposes; HBS targets household expenditure whereas EU-SILC targets household income.

In the two tables that follow, income results from both surveys are shown. They present the percentage of households and persons having received an amount on a specific income target variable as well as its mean value per household. It should be stated that income questions in HBS were answered by persons aged 15 and over whereas in EU-SILC by those 16 and over. Some income variables were grouped so that comparison could be more relevant. The results of the two surveys are favourably compared.

9.1.1. Coherence - sub annual and annual statistics

Table : Comparison between Household Budget Survey 2009 and EU-SILC 2010 for income variables at household level

Income target variable	EU-SILC 2010		HOUSEHOLD BUDGET SURVEY 2009
Income target variable	% of households having received an amount	Mean income per household (EURO)	% of households having received an amount	Mean income per household (EURO)
Total household gross income HY010	99,9	40.308	100,0	38.358
Total disposable household income HY020	99,9	35.682	100,0	34.564
Income from rental of a property or land HY040G	8,2	726	6,8	685
Family/children related allowances HY050G / Social exclusion not elsewhere classified HY060G	52,5	975	54,4	856
Housing allowances HY070G	2,5	200	1,1	135
Regular inter-household cash transfer received HY080G	8,4	450	9,1	382
Interest, dividends, profit from capital investment in unincorporated business HY090G	12,5	502	12,7	426
Regular taxes on wealth HY120G	58,7	48	61,5	64
Regular inter household cash transfer paid HY130G	14,9	581	12,7	482
Tax on income and social contributions HY140G	74,8	3.997	72,5	3.248

Table : Comparison between Household Budget Survey 2009 and EU-SILC 2010 for income variables at individual level

Income target variable	EU-SILC 2010		HOUSEHOLD BUDGET SURVEY 2009
Income target variable	% of persons 16+ having received an amount	Mean income per household (EURO)	% of persons 15+ having received an amount	Mean income per household (EURO)
Employee cash or near cash income PY010G	49,2	26.612	54,4	26.147
Non-cash employee income PY020G	6,0	191	n.a	193
Cash benefits or losses from self-employment PY050G	11,6	4.298	n.a.	2.995
Unemployment benefits PY090G	3,5	383	2,9	313
Old-age benefits (PY100G)/ Survivor benefits (PY110G)/ Sickness benefits (PY120G)/ Disability benefits (PY130G)	26,4	6.207	22,4	5.748
Education-related allowances PY140G	5,8	337	6,6	449

The next table presents the labour force participation rates as they were recorded by Labour Force Survey 2011 and EU-SILC 2011. There is one main methodological difference between the two surveys, for LFS students studying abroad or national guards (compulsory army service) are not considered to be part of the population, whereas they are part of the EU-SILC population. Thus, the totals as well as the rates of the ages 16-24 are not comparable. The rest of the results up to the age of 59 fit very well. EU-SILC seems to underestimate the rates for persons aged 60 years and over, but this is understandable since LFS is the core survey with main objective to collect information on employment.

Table : Comparison between Labour Force Survey 2011 and EU-SILC 2011 for the labour force participation rates
Age Groups	Total		Males		Females
Age Groups	LFS	EU-SILC	LFS	EU-SILC	LFS	EU-SILC
16 - 19	10,1	10,4	11,2	9,4	9,2	11,4
20 - 24	65,8	55,3	68,3	56,0	63,5	54,6
25 - 29	88,2	88,5	89,4	89,2	87,0	87,8
30 - 34	89,8	91,7	95,1	98,2	85,0	86,0
35 - 39	90,0	93,4	95,7	98,7	85,3	88,9
40 - 44	88,1	90,0	94,6	95,2	82,5	85,5
45 - 49	86,4	86,5	95,2	96,9	78,3	76,8
50 - 54	80,1	80,3	89,3	91,7	71,2	69,3
55 - 59	69,8	71,0	85,6	86,6	54,3	55,6
60 - 64	44,8	39,5	59,5	51,5	30,6	27,9
65+	11,2	5,2	17,5	8,7	5,8	2,2
Total	63,7	61,8	70,7	67,6	57,5	56,6

9.1.2. Coherence - National Accounts

Not available

9.2. Coherence - internal

In the tables that follow, we compare the results on income components between EU-SILC 2008, EU-SILC 2009, EU-SILC 2010 and EU-SILC 2011 at both household and personal level. More specifically in the two tables that follow the percentage of households and persons having received an amount on specific income target variables, as well as their mean value per household are presented.

The results show that the percentage of either households or persons receiving an amount between the four surveys are very close and hence consistent.

Table : Comparison between EU-SILC 2008, 2009, 2010 and 2011 for all income target variables at household level
Income target variable	EU-SILC
	2008		2009		2010		2011
	% of households having received an amount	Mean (weighted) income per household (EURO)	% of households having received an amount	% of households having received an amount	Mean (weighted) income per household (EURO)	Mean (weighted) income per household (EURO)	Mean (weighted) income per household (EURO)	Mean (weighted) income per household (EURO)
Total household gross income HY010	100,0	38.652	100,0	39.677	99,9	40.308	100,0	41.094
Total disposable household income HY020	100,0	34.625	100,0	35.496	99,9	35.682	100,0	36.142
Total disposable household income before social transfers other than old-age and survivor's benefits HY022	99,5	32.475	99,5	33.113	99,4	33.245	99,3	33.388
Total disposable household income before social transfers including old-age and survivor's benefits HY023	90,0	27.838	89,1	27.939	90,5	27.532	91,6	27.506
Imputed rent HY030G	91,8	5.994	92,7	7.055	90,7	6.851	90,6	5.929
Income from rental of a property or land HY040G	8,9	804	8,5	740	8.2	726	8,7	696
Family/children related allowances HY050G	50,1	733	51,3	843	51,9	936	52,8	1.022
Social exclusion not elsewhere classified HY060G	0,7	40	0,6	42	0,6	39	0,5	46
Housing allowances HY070G	1,9	127	2,0	138	2,5	200	3,5	307
Regular inter-household cash transfer received HY080G	8,3	365	7,6	338	8,4	450	9,1	519
Interest, dividends, profit from capital investment in unincorporated business HY090G	11,1	572	11,4	504	12,5	502	13,7	629
Interest repayments on mortgage HY100G	13,6	525	11,9	571	10,3	540	10,8	589
Regular taxes on wealth HY120G	61,2	54	60,0	49	58,7	48	61,5	50
Regular inter household cash transfer paid HY130G	11,5	467	12,2	461	14,9	581	17,1	732
Tax on income and social contributions HY140G	75,1	3.505	73,6	3.670	74,8	3.997	75,1	4.170
Value of goods produced for own consumption HY170G	N.A.	N.A.	N.A.	N.A.	5,7	15	7,3	18

Table : Comparison between EU-SILC 2008, 2009, 2010 and 2011 for all income target variables at individual level

Income target variable	EU-SILC
	2008		2009		2010		2011
	% of persons 16+ having received an amount	Mean (weighted) income per household (EURO)	% of persons 16+ having received an amount	Mean (weighted) income per household (EURO)	% of persons 16+ having received an amount	Mean (weighted) income per household (EURO)	% of persons 16+ having received an amount	Mean (weighted) income per household (EURO)
Employee cash or near cash income PY010G	50,3	24.870	48,7	25.550	49,2	26.112	49,7	26.498
Non-cash employee income PY020G	7,3	230	6,2	196	6,0	191	6,4	216
Company car PY021G	1,4	83	1,1	73	0,9	68	1,0	75
Employer´s social insurance contribution PY030G	45,9	3.179	44,7	3.200	55,0	3.417	45,9	3.538
Cash benefits or losses from self-employment PY050G	12,2	4.947	11,9	4.608	11,6	4.298	11,4	3.863
Unemployment benefits PY090G	3,6	434	2,7	516	3,5	383	3,8	429
Old-age benefits PY100G	21,2	4.682	22,5	5.277	22,0	5.550	20,3	4.992
Survivor benefits PY110G	1,0	177	0,8	204	0,7	163	4,6	890
Sickness benefits PY120G	0,9	50	1,0	55	1,1	64	1,1	43
Disability benefits PY130G	2,5	420	2,5	441	2,6	480	2,7	518
Education-related allowances PY140G	6,4	344	6,3	347	5,8	337	6,1	388

10. Cost and Burden

Top

Not requested by Reg.28/2004

11. Confidentiality

Top

11.1. Confidentiality - policy

Not requested by Reg.28/2004

11.2. Confidentiality - data treatment

Not requested by Reg.28/2004

12. Statistical processing

Top

12.1. Source data

Sampling frame and coverage errors

Coverage problems encountered were:

The frame of the 2001 Census of Population was somehow outdated and as a result some housing units were found to be empty or to be used for other purposes other than housing.
Some houses included in the E.A.C. list were used as secondary residence, so they were out of scope of the survey.
Some houses listed by the E.A.C. were impossible to be located due to incomplete information regarding their addresses.
Housing units built after September 2010, were not included in our sampling frame.

12.1.1. Sampling design and procedure

Type of sampling design

The sample design was one-stage stratification.

Stratification and sub stratification criteria

Geographical stratification criteria were used for the sample selection. The households were stratified in 9 strata based on District (Urban / Rural), i.e. 1) Lefkosia Urban, 2) Lefkosia Rural, 3) Ammochostos Rural⁽¹⁾, 4) Larnaka Urban, 5) Larnaka Rural, 6) Lemesos Urban, 7) Lemesos Rural, 8) Pafos Urban, 9) Pafos Rural.

⁽¹⁾Ammochostos Urban is an area not under the effective control of the Government of the Republic of Cyprus.

Sample selection schemes

The sample was selected from each stratum with simple random sampling.

Sample distribution over time

Sample distribution over time
Period	Addresses in initial sample	Addresses out of scope	Addresses used	Addresses not successfully contacted	Non-response	Household Questionnaire Completed
01/03 – 15/03	236	5	231	0	19	212
01/03 – 31/03	781	39	742	0	43	699
01/03 –15/04	1.332	75	1.257	4	75	1.178
01/03 – 30/04	1.670	103	1.567	4	99	1.464
01/03 – 15/05	2.358	158	2.200	4	142	2.054
01/03 – 31/05	3.058	202	2.856	7	202	2.647
01/03 – 15/06	3.737	249	3.488	9	287	3.192
01/03 – 30/06	4.317	281	4.036	9	355	3.672
01/03 – 15/07	4.537	288	4.249	11	399	3.839
01/03 – 31/07	4.645	290	4.359	13	429	3.917

12.1.2. Sampling unit

The sampling units are private households, which were selected with simple random sampling within each stratum.

12.1.3. Sampling rate and sampling size

Concerning the SILC instrument, three different sample size definitions can be applied:

- the actual sample size which is the number of sampling units selected in the sample

- the achieved sample size which is the number of observed sampling units (household or individual) with an accepted interview

- the effective sample size which is defined as the achieved sample size divided by the design effect with regards to the at-risk-of poverty rate indicator

Given that the effective sample size has been already treated in the section dealing with sampling errors, in this section the attention focuses mainly on the achieved sample size.

Sample size and allocation criteria

According to the Regulation (EC) No 1177/2003 Article 9, the minimum effective sample size for Cyprus is 3.250 households and 7.500 persons aged 16 or over. As the sample is based on a rotational design of 4 replications with a rotation of one replication per year, the selection of one new sub-sample was required. More specifically, for 2011 one sub-sample of 2010 survey was dropped (R3), and a new sub-sample (R2) was separately selected in the same manner as in 2005, so as to represent the whole population. Due to the non-response of 2010 survey and the number of non existent or not successfully contacted addresses, the initial sample of 2011 survey was 4.645 households. The status of our sample for the 2011 round in each rotational group is as follows:

	Total	R1	R2	R3	R4
Status of sample	4.645	1.560	1.600	726	763

The allocation of the sample in the 9 strata is shown in the table below:

Population and sample distribution

DISTRICT	N			n
	NUMBER OF HOUSEHOLDS 2011			DISTRIBUTION OF THE SAMPLE
	TOTAL	URBAN	RURAL	TOTAL	URBAN	RURAL
TOTAL	300.100	206.500	93.600	4.649	3.174	1.475
LEFKOSIA	118.400	89.700	28.700	1.755	1.319	436
AMMOCHOSTOS	15.600	0	15.600	264	0	264
LARNAKA	49.200	29.600	19.600	789	473	316
LEMESOS	84.500	65.000	19.500	1.364	1.071	293
PAFOS	32.400	22.200	10.200	477	311	166

For the data collection 23 interviewers were appointed, 8 in Lefkosia district, 5 in Larnaka/ Ammochostos, 7 in Lemesos and 3 in Pafos. The sampled households were grouped as much as possible in small areas so as to minimise travelling expenses. Each interviewer had to visit on average 15 households per week.

The 2011 sample results are shown in the table below:

Table 2.1.4.2 : Sample size
Addresses in initial sample	4.649
Addresses used for the survey	4.359
Addresses out of scope	290

Addresses used	4.359
Addresses successfully contacted	4.346
Addresses not successfully contacted	13

Addresses successfully contacted	4.346
Household questionnaire completed	3.917
Refusal to cooperate	353
Entire household away for the duration of fieldwork	23
Household unable to respond	41
Other reasons for not completing the Household questionnaire	12

Household questionnaire completed	3.917
Interviews accepted for database	3.917
Interviews rejected for database	0

The 290 addresses that were out of scope of the survey correspond to vacant accommodation, or buildings used as secondary residences or for business purposes, or demolished housing units. Furthermore, 13 addresses were not successfully contacted. Out of the 4.346 addresses successfully contacted, 3.917 households completed the Household questionnaire and were all accepted for the database. This was above the minimum effective sample size (3.250 households) requested by the Regulation (EC) No 1177/2003 Article 9. Thus, the achieved sample size was 3.917 households, 11.443 persons in total and 9.500 persons aged 16 or over. In order to achieve this, the number of households of the new sub-sample selected was 1.600.

Achieved sample size

The table below presents the achieved samples of persons aged 16 years and over, as well as of households, within each rotational group.

Sample Size and Accepted Interviews
	Total	R1	R2	R3	R4
Persons 16 years and over	9.500	3.493	2.603	1.681	1.723
Number of accepted personal questionnaires	9.500	3.493	2.603	1.681	1.723
Accepted household interviews	3.917	1.438	1.077	691	711

Substitutions

No substitution procedures were applied.

Method of selection of substitutes

Not applicable.

Renewal of sample: rotational groups

The sample in the first round was divided in 4 sub-samples as it was based on a rotational design of 4 replications with a rotation of one replication per year. Each sub-sample was separately selected so as to represent the whole population. Every year one sub-sample is going to be dropped and substituted by a new one. Thus for 2011 one specific sub-sample, pre-selected from 2007 (R3), was dropped and substituted by a new one (R2). The new sub-sample was also separately selected, so as to represent the whole population.

The size of each Rotational Group for the 2011 survey is shown in Table below:

Size of the Rotational Groups

	Total	R1	R2	R3	R4
Addresses in initial sample	4.649	1.560	1.600	726	763
Household Questionnaire completed	3.917	1.438	1.077	691	711
Interviews Accepted for database	3.917	1.438	1.077	691	711

12.2. Frequency of data collection

CYSTAT collects EU-SILC data annually.

12.3. Data collection

Mode of data collection

The mode of data collection for EU-SILC survey was CAPI. Paper Assisted Personal Interviewing (PAPI) was only used in the extreme case of a technical problem with the interviewer’s laptop (for 2011 only once). Of all completed personal questionnaires 18,9% were filled with proxy interviews; 49% of them corresponded to persons who were temporarily absent mainly national guards and students who were supported by their parents. For these cases we preferred to have a personal questionnaire filled with a proxy interview rather than a refusal. Also in many cases where a person was not temporarily absent and a proxy interview existed, the interviewer would communicate with the interviewee by telephone and some personal questions would be answered directly by the interviewee.

The following tables present the distribution of individuals aged 16 or over by data status and type of interview.

RB250 Data status	Total	R1	R2	R3	R4
RB250 Data status	Count %	Count %	Count %	Count %	Count %
Total	9.500 100	3.493 100	2.603 100	1.681 100	1.723 100
information completed only from interview (11)	9.491 100	3.491 99,9	2.603 100	1.680 99,9	1.717 99,7
information completed from full record imputation (14)	9 0,0	2 0,1	0 0,0	1 0,1	6 0,3
individual unable to respond and no proxy possible (21)	0 0,0	0 0,0	0 0,0	0 0,0	0 0,0
refusal to co-operate (23)	0 0,0	0 0,0	0 0,0	0 0,0	0 0,0
person temporarily away and no proxy possible (31)	0 0,0	0 0,0	0 0,0	0 0,0	0 0,0
no contact for other reasons (32)	0 0,0	0 0,0	0 0,0	0 0,0	0 0,0
information not completed: reason unknown (33)	0 0,0	0 0,0	0 0,0	0 0,0	0 0,0

Distribution of individuals aged 16 or over by type of interview and rotational group

RB260 Type of interview	Total	R1	R2	R3	R4
RB260 Type of interview	Count %	Count %	Count %	Count %	Count %
Total	9.491⁽¹⁾ 100	3.491 100	2.603 100	1.680 100	1.717 100
face to face interview-PAPI (1)	1 0,0	0 0,0	1 0,0	0 0,0	0 0,0
face to face interview-CAPI (2)	7.697 81,1	2.825 80,9	2.130 81,8	1.359 80,9	1.383 80,5
proxy interview (5)	1.793 18,9	666 19,1	472 18,2	321 19,1	334 19,5

(1) The total number of individuals aged 16 and over is 9.500. The information for 9 of these individuals was completed from full record imputation.

1-PAPI (% of total)	2-CAPI (% of total)	3-CATI (% of total)	4-Self administrated (% of total)
0,0	100,0	0,0	0,0

The mean interview duration

The mean interview duration per household is calculated as the sum of the duration of all household interviews plus the sum of the duration of all personal interviews, divided by the number of household questionnaires completed. Only households accepted for the database have to be considered.

Average interview duration =52 minutes

12.4. Data validation

Not requested by Reg. 28/2004

12.5. Data compilation

Please find below a description of the weighting and imputation procedures .

12.5.1. Weighting procedure

Design factor

Non-response adjustments

Adjustment to external data

Final cross sectional weights

The methodology that was used for the computation of the weights of the survey is the one proposed in Doc. EU-SILC 065/09. For a household in the new panel 2 (R2), the design weight is the inverse of its inclusion probability that is the probability belonging to the selected sample of households:

DB080_i = 1/π_i= 1 / (n_i / N_i) = N_i/ n_i , i=1,…,9

π= the probability of a household to be selected from stratum i

n= the sample size of stratum i

N= the total number of households in the sampling frame of stratum i

For households in the older panels, the household design weights were calculated by following the methodology proposed by Eurostat in Doc. 065/09. The general steps followed were:

Computation of panel person base weights
Correction for non response due to attrition
Computation of base weights for persons entering panel households for the first time, i.e. newborns of sample women or persons moving into sample households from abroad
Non-panel persons (co-residents) have a basic panel weight equal to zero
Computation of household weights by averaging within household over all household members

For new panel:

The aim of non-response adjustments is to reduce the bias due to non-response, i.e. household was contacted (DB120=11) but household questionnaire was not completed (DB130≠11). The empirical response rate within each stratum provides an estimate of the response probability for all the households of the stratum. The weight of a household after correction for the non-response at the household level is:

DB080_i*1/^p_i

DB080_i= the design weight of a household in stratum i before non-response adjustment

^p_i= the estimated response probability of the household in stratum i

The next step is to combine the entire sample (panels 1 – 4) and apply the calibration procedure. The target of the calibration procedure is to improve the accuracy of the estimated household and personal weights by using external known information. Eurostat recommends an “integrative” calibration. The idea is to use calibration variables defined at both household and individual level. The individual variables are aggregated at the household level by calculating household totals such as the number of male/female in the household, the number of persons aged 16 and over etc. After that, calibration is done at the household level using the household variables and the individual variables in their aggregate form.

The calibration variable used at household level was the household type:

One adult no dependent children.
At least two adults no dependent children.
One adult with at least one dependent child.
Two adults with one dependent child.
Two adults with two dependent children.
Two adults with at least three dependent children.
At least two adults and at least one dependent child.

At personal level the calibration variables used were the distribution of population by age (age≤15, 16≤age≤19, 20≤age≤24,…, 70≤age≤74, age≥75) and gender.

Based on this calibration procedure and using the weight after non-response adjustment as the initial weight, the household (DB090) and the personal (RB050) cross-sectional weights were calculated.

Calibration procedures were further used for the calculation of cross-sectional weights for household members aged 16 and over (PB040) and for the children aged 0 to 12 years (inclusive) (RL070). For both PB040 and RL070 the personal cross-sectional weight RB050 was used as the initial weight. The calibration variables used for the cross-sectional weight of household members aged 16 and over were the distribution of population aged 16 and over by age (five years age groups) and gender. The respective calibration variable for the children cross-sectional weight for childcare (RL070) was the distribution of population aged 0 to 12 by single years of age.

The final cross-sectional weights were calculated as described above, i.e. using DB080 after non-response adjustment as the initial weight for new panel and base weights adjusted for non-response due to attrition for older panels. The calibration methods were then applied on the total sample.

12.5.2. Estimation and imputation

Imputation procedure used

Imputed rent

Company car

In the very few cases where imputation required, the method used was deductive imputation. Imputation was necessary in the cases where only net income was collected and in the cases of personal refusals. Net income was converted to gross by applying the existing tax system and social insurance contributions rules. Personal refusals were imputed using existing data from previous waves as the starting point.

Imputed rent was calculated using Heckman Method as it was one of the methods proposed by Eurostat. The following variables were taken into account for the calculation: type of dwelling, number of rooms, area in square meters, year of construction, heating, air-conditioning and income brackets. Despite the fact that efforts were made to make correct estimates using the Heckman method, however we still have our reservations as regards to the accuracy of these estimates, due to the fact that the rental market in Cyprus is considered quite small.

To valuate the benefit of private use of company car the approach of ‘Valuation on the basis of accrued saving’ according to Doc. EU-SILC 065 was followed. In order to valuate the amount the recipient would have to pay over the reference period to enjoy the same benefit from the use of own vehicle the sum of (i) & (ii) below were computed:

(i) Depreciation over the reference period in the capital value of the car,

(ii) Coverage by the employer of other costs, which would normally fall on the user of his/her own car. The latter may cover car insurance and possibly maintenance and major repair costs, but would normally exclude fuel and other running costs.

External sources had to be used to construct suitable average schedules for (i) and (ii), rather than to collect (i) and (ii) from individual respondents.

The main requirement was to construct a ‘depreciation model’:

Depreciation = ( Purchase prices - Selling prices at X) / X ,

where X = ‘the average age of a company car’

To calculate the ‘Purchase price’ and the ‘Selling price’, the make, the model, the registration year and other characteristics of the car were used. A list of prices and manufacturer’s recommended retail prices (RRP) were also used for a wide range of new cars. If the RRP was not available, then it was estimated based on the price of a similar car or the price relative to other cars with a similar pricing structure. The list price included VAT and vehicle registration tax. For calculating ‘the average age of a company car’, an average of 5 was considered.

12.6. Adjustment

Not requested by Reg.28/2004

13. Comment

Top

National questionnaire is available in Circa BC at: https://circabc.europa.eu/ . Please select EU SILC section and then select the folder '06 National Questionnaire' in the library list. Additionally under the folder '02 Guidelines' and then under the folder '2.4 2011 Operation Guidelines' you can find information of the 2011 Ad-hoc Module variables.

Annexes

Top