What does the indicator describe?
Unlike standard employment, atypical forms of employment comprise part-time employment incl. 20 or less working hours per week, marginal employment, fixed-term employment and temporary employment.
While standard employment is aimed at fully covering one’s own living expenses and probably those of relatives, atypical employment can usually not meet that requirement completely. However, atypical employment is not the same as precarious employment. The latter is characterised by an increased poverty risk of the persons employed, which additionally depends on the people’s working lives and their individual household context. The forms of atypical employment referred to here may indeed be based on deliberate decisions which are for instance aimed at better combining work and other personal interests.
How is the indicator calculated?
The number of people in atypical employment is calculated from the annual microcensus results.
The total of people in atypical jobs is determined based on the number of all employees to whom the following criteria apply:
- fixed term employment
- part-time employment including 20 or less hours
- temporary employment
- marginal employment
The results of atypical employment refer to persons aged 15 to 64 years who are not undergoing education or training. For the relevant calculations, the population is defined in the following way:
- persons aged 15 to 64 years at their main residence
- excluding pupils/students/apprentices
- excluding persons doing compulsory military or civilian service and persons doing voluntary military service or volunteer service
The members of this group (which is also referred to as core group of persons in employment) are in a phase of life where the focus is much more on employment than, for instance, during education and training or retirement. For this reason, the group is used as a reference in calculating rates, especially in the framework of reporting on atypical employment.
When is the indicator released?
Annual results of the microcensus are available roughly three months after the end of a survey year. Atypical employment results are published in the summer of the year following the reference year, after processing and the necessary analyses have been completed.
How accurate is the indicator?
Two types of errors occur in sample surveys like the microcensus: random errors and systematic errors. Random errors are deviations which are due to the fact that only a sample of persons is included in the survey; in the case of the microcensus the sample comprises one percent of the population. The so-called standard error serves as an estimate of the random sampling error; estimation is based on the individual data of the sample. Regarding grossed-up annual results for less than 5,000 persons, that is, for less than 50 cases in the sample, the simple relative standard error often exceeds 15%. As the informative value of such results is low, they should not be used for comparisons. Therefore, grossed-up cell frequencies of under 5,000 (annual basis) are not shown; they are rather indicated by a slash (“/”) in the microcensus publications.
Systematic errors are non-random errors; the respective deviations may result from errors at various levels of statistics production (relating e.g. to the design of data collection instruments, interviewer behaviour, incorrect data or errors in data capture).
The microcensus results concerning labour force participation partly differ from the results of other sources of labour market statistics. In the microcensus, the number of persons in employment in 2013 is 6.2% lower than in the employment accounts that are part of the national accounts. Apart from minor differences in definitions, the discrepancy is especially due to the methods and procedures applied. Detailed comparative analyses (e.g. using a follow-up survey) have shown that differences occur especially in smaller and marginal activities. Actually, recording these activities in household surveys like the microcensus involves problems. The differences between the microcensus and the employment accounts should be taken into consideration when interpreting the data on persons in employment. Regarding the number of people in atypical employment, the problems in recording smaller activities can be largely neglected as only people aged 15 to 64 are included who are not undergoing education or training.
When making comparisons with other data sources comprising information on atypical employment, it has to be taken into account that the microcensus focuses on employed people rather than jobs, that second jobs are not included and the group of persons covered is limited in terms of age and participation in education.