Official name of the survey/data source:
Czech Microcensus, 1992

Administrative Unit responsible for the survey:

The following people with specific expertise for various parts of the survey can be contacted at the above address:

Sampling and sample selection:

Data Collection Process/ Income Variables / Demographic Variables:

Editing/Survey Processing:

The survey was first conducted in 1956. and is conducted every 3-5 years. A new series was issued in 1958 in which the sample size was cut by one-fourth. Subsequent surveys occurred in 1960, 1966, 1970, 1973, 1976, 1980, 1985, 1989, 1992.

The survey was first made available as microdata in 1970 when it ha been released to the Institute of Sociology, Czechslovak Academy of Sciences.

The main focus of the survey is to measure household standards of living.

The main source of funding for the survey is the central budget of the Czech Republic.

The process of data collection began February 25, 1993 (note: the income for 1992 was declared) and was completed March, 18, 1993

The principle users of the data were the Czech Statistical Office which produced a series of tables which were publicly distributed. The Ministry of Labor and the Institute of Sociology have also been granted waves of the Czech Microcensus.

A book is released by the Central Statistical office entitled Microcensus1992.

Statistical surveys are not released to the public domain. Special contracts are necessary for release of the data to the Ministry of Labor and other institutions.

These microdata been made available to the Luxembourg Income Study Project under the condition that all names and addresses would be destroyed from records released to outside institutions.

B POPULATION AND SAMPLE SIZE, SAMPLING METHODS back

1. The Sample Design and Sampling Frames

The sampling frame was the Census of Population 1991. Only those living in the household were included. Excluded from the sample are all individuals which are absent from the household for greater than six months of the year. All institutions are not considered households. Those individuals in the military are excluded if they are not household members for more than six months of the year. Officers who live with their families are included in the survey.

The Microcensus is a stratified random sample. The two step stratification is done by size of the locality (urban if greater than 2300 apartments of greater than 29 census tracts) and rural otherwise, and then further stratified by the number of apartments within the census tract (three stage breakdown).

Weights are not yet produced by the Statistical Office.

The basis unit of observation is the flat. There are 3.8 million economic households. Unoccupied units were excluded.

C. MEASURES OF DATA QUALITY back

1. Response Rates

Total number of respondents actually surveyed :

Unweighted N=_15677__ (which represents .5% of the total population)

Weighted N=_3836040__

Total number of sample units in the sampling not interviewed:

Unweighted N=_21912___

Weighted N=__________

Information is available on the address, reasons for non-response, and the number of household members.

The sampling frame includes all geographic areas in the country.

Estimates of sampling variability, that is, standard errors have been computed for this survey.

They are not published in reports

Overall non-response rates to the survey : 15.7%

2. Reporting and Under-Reporting and Income Data Quality

Income data was aggregated and compared to national income accounts. Microcensus incomes were found to be much lower than should be.

The demographic characteristics have been compared with other sources.

Several checks were made. Individual reported income was compared to household income. Also the interviewer was independently requested to report the "trustworthiness" of the information. Cases were dropped if the interviewer viewed them as "untrustworthy".

When earnings were not reported interviewers imputed an amount according to the individuals job, age, and gender. Also for those families witch did not declare family benefits were received, they were imputed for the individuals.

Special techniques were used to reduce error, or correct for item non-response in order to reduce the underreporting of incomes. In some cases, values has been replaced by consulting alternative data sources (e.g., administrative records): family allowances.

When earnings were not reported interviewers imputed an amount according to the individuals job, age, and gender. Also for those families witch did not declare family benefits were received, they were imputed for the individuals.

Most important problems regarding the quality of the income data:

1. The instruction were not clearly given to those individuals who were entrepreneurs as to how to designate net income (after taxes and investments).

2. There is large incidence of underreported income.

3. Boarder workers have an inflated income due to exchanging foreign wages. Often these wages are not reported. Rents are also much higher along border areas.

D. DATA COLLECTION AND ACQUISITION back

Interviewers would make one personal visit to the household. The survey is organized in to sections: household characteristics, personal income, agriculture, housing.

The survey is currently voluntary, previously participation was mandatory. Anonymity is guaranteed by law.

The interviewer interviews only one member of the household. Others members are allowed, but not required, to make inputs.

Respondents were not required to consult or use pay records, tax returns, etc. to help provide the most accurate income information possible.

E. WEIGHTING PROCEDURES back

A survey weight has been assigned to each sample case to allow recalculation to represent the total population. Each case is weighted by about 200 according to registers. Institutional publications were considered in recalculation also.

Estimates derived from the sample are representative of the total population defined by the sampling frame?

F. DETERMINATION OF SURVEY UNIT MEMBERSHIP back

The survey unit is the economic household. People living in the apartment are asked to declare if they are members of a common household, thus the decision is based on individual judgment. There could exist more than one household at a given address. The tenant is the individual who holds the lease or deed to the dwelling. In the case of a cooperative apartment between a couple, the male is considered the tenant. Separate households are considered subtenants within the dwelling.

The basic unit is the flat. Each flat can be divided into economic households and further into individuals. The head is the male member of the couple which owns the flat. In subfamilies the male is always the head. There are individual records and details of the relationships to both the head of the flat household and the economic household. Relationships within the household include: Head, spouse, children, step children, grand parents, parents, step-parents, brothers and sisters and outsiders

G. CHILDREN AND SPOUSES back

Children are defined to be under age 18 or are 26 and pursuing education/training without a break. It is possible to distinguish between children of the unit head (or spouse) and children of some other household member or someone living outside the household by the relation to the head.

The spouse is the legal partner of the head. Note that marriage is not a necessity but individuals must be of the opposite sex.

Both siblings and parents are identified.

H. AVAILABILITY OF BASIC SOCIAL AND DEMOGRAPHIC INFORMATION back

I. AVAILABILITY OF LABOUR MARKET INFORMATION back

J. AVAILABILITY OF GEOGRAPHIC INFORMATION back

Geographic location can be broken by regions (10 divisions), districts (85 divisions), size of locality (9 categories), and number of apartments in the census tract (3 divisions).

K. SOURCES AND AMOUNTS OF CASH INCOME back

L. TAXES back

M. BIBLIOGRAPHY OF MAIN PUBLICATIONS BASED ON THE CSP PANEL back