Accuracy of results
The Structural Survey is conducted as a sample survey. Data are extrapolated to obtain statistical results for the whole population. These are estimates subject to a certain degree of uncertainty
The Structural Survey is conducted as a sample survey. Data are extrapolated to obtain statistical results for the whole population. These are estimates subject to a certain degree of uncertainty
Other facts that can affect the accuracy of these estimates are: the consistency of the variables recorded, the size and structure of the sample as well as the willingness of the persons interviewed to answer the questions (response rate). This means that there are limits to evaluating the sample and that results can only be interpreted with certain limitations.
By merging or cumulating samples (pooling) it is possible to analyse data together, gathered on several separate occasions for the same survey. Merging gives more accurate results than those obtained from one single sample. The results from the cumulated data, however, no longer refer to a specific reference date but are mean values for the whole period under observation and for an average population during this period.
The Structural Survey samples are merged for a period covering three to five years. After three years a pooled sample of 600'000 persons is obtained and after five years one consisting of a million people. The pooled estimates are updated every year so that "moving averages" are available.
Sample size of ... people |
Depth of focus (units of persons) |
Resolution | |
---|---|---|---|
1 year | 200'000 | approx. 15'000 | 140 |
3 years |
600'000 | approx. 5000 | 47 |
5 years |
1'000'000 | approx. 3000 | 28 |
Further information on this subject can be found in the document "Strukturerhebung: Aufstockungsmodelle für die Kantone" (in German and French) at point 3.
With the standard programme, the Confederation provides the cantons with an extensive range of information free of charge. Should the data provided prove to be insufficient for certain regional policy or planning purposes, the cantons can enlarge the sample. But the possibilities for supplementing the sample are subject to some organisational and methodological restrictions:
An enlargement of the sample size can be considered if a canton's needs are not met by the annual sample of 3.5% of the population or by the three or five-year pooled samples:
According to Article 21 of the Ordinance on the Federal Population Census, the cantons can request an enlargement of the sample size no greater than double the population living in their territory. In 2010 they were exceptionally allowed to enlarge their sample no more than fourfold, provided that they made no enlargement in the years 2011 and 2012 (Art. 30, para 1). The cantons must order the enlargement no later than 31 December of the previous year (enlargement equal to a maximum of double the population).
According to Article 14 of the Federal Act on the Federal Population Census, the costs for the enlargement of the sample size are to be borne in their entirety by the canton that requests them. The cantons assume the actual costs incurred by the enlargement of the sample in their territory. The modalities (including the payment modalities) are regulated in an individual contract with each canton requesting a sample enlargement.
Depending on the degree of enlargement and the size of the sample after enlargement, various prices are applicable. The price depends on the infrastructure necessary, which varies according to the size of the sample. Detailed information on this subject can be found in the document "Strukturerhebung: Aufstockungsmodelle für die Kantone" (in German and French) at point 4.
Monday to Friday
09.00–11.30 and 14.00–16.00