Non-response in research on the destination of people who started up new businesses : correction procedure concerning failure to respond taking as an example formerly unemployed people who started up businesses with the aid of the bridging allowance

Abstract

"Five years after a business start-up, it becomes increasingly difficult to obtain information on the destination of people who set up businesses with the aid of the bridging allowance. At an earlier stage of the project - after an average of three years of self-employment - it was still possible to obtain a response rate of nearly 90 per cent of the initial sample when questioning the people starting up businesses. Now, however, at the end of the study, it is not only the 'natural decline' in sample units - for example due to addresses that can no longer be investigated - which poses methodological problems for the analysis. Systematic biases occurred also as a result of the survey organisation. For the final telephone interviews were only carried out with those business founders who had signalled in the penultimate wave of the survey that they would be willing to take part in the survey again. This results in a positive bias which requires appropriate corrections. With the aid of a worst-case scenario, a restricted extrapolation, data comparisons and a weighting procedure, a 'corridor' was defined in which the 'true' survival rate of the former bridging allowance recipients is most likely to range. According to that, at least one in two of the former unemployed people is still running his company a good five years after the start-up. All in all when comparing these findings with those of other surveys on start-ups, it can be stated that the outcome of the assisted start-ups is similar to that of start-ups initiated without the handicap of previous unemployment. Despite the remaining lack of clarity in the data stock and the growing difficulties in correcting these biases, the former bridging allowance recipients still show a respectable balance - at least at gross level. It is not possible to make a statement regarding the net effects of the support programme as no appropriate control group is available. Also, when looking at the factors for success, little change can be detected over time. Just as two years before, the critical factors continue to be in the area of human capital and start-up capital. Whether further support programmes might provide sensible additional measures must be assessed by subsequent research. The various initiatives and offers that have been newly created over the past years can help to tap further potential for start-ups." (Author's abstract, IAB-Doku) ((en))

Cite article

Wießner, F. (2003): Nonresponse bei Verbleibsuntersuchungen. Korrekturverfahren zu Antwortausfällen am Beispiel ehemals arbeitsloser Existenzgründer, die mit dem Überbrückungsgeld (§ 57 SGB III) gefördert wurden. In: Mitteilungen aus der Arbeitsmarkt- und Berufsforschung, Vol. 36, No. 1, p. 77-96.