Zur Schätzung des Inkonsistenz-Index als Maßzahl zur Fehlermessung bei klassierten Erhebungsergebnissen / On the Estimation of the Index of Inconsistency as Special Statistic to Measure Errors in Classified Survey Results

Strecker Heinrich
Additional contact information
Strecker Heinrich: emer. o., Universität Tübingen, und Honorarprofessor der Universität München, Rosenstr. 11, D-82319 Starnberg bei München

Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik), 2000, vol. 220, issue 6, 777-792

Abstract: The index of inconsistency is outlined as a special statistic for judging the reliability of survey data classified by size classes. The index estimates take on, as a rule, values between 0 and 1 according to a previously definded restriction for dichotomized proportions in the estimation formula (estimates should be ≤ 1). If the results of a main survey and a repetition show no variability of the individual responses regarding the assignment of units i to specified size classes the index value is 0, in case of total unreliability (random differences) it is, as a rule, 1. The extreme values 0 and 1 are demonstrated by means of simulated examples and their data show that the restriction required is fulfilled. Yet, there are exceptions. Another instructive example proves that in the case of total unreliability of responses the estimated index values may be larger than 1, here the restriction has not been fulfilled.Estimating the index values ist done on the basis of data of two surveys by means of fourfold tables. These can be set up without any effort by crossing out rows and columns from tables with classified results of two surveys. An application of the model to official statistics in Belgium rounds off the discourse.

Keywords: Index of inconsistency; response variability; fourfold tables; estimation with application in Belgium; Inkonsistenz-Index; Antwortvariabilität; Vier-Felder-Schemata; Schätzung mit Anwendung in Belgien; Index of inconsistency; response variability; fourfold tables; estimation with application in Belgium (search for similar items in EconPapers)
Date: 2000
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1515/jbnst-2000-0611 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:jns:jbstat:v:220:y:2000:i:6:p:777-792

DOI: 10.1515/jbnst-2000-0611

Access Statistics for this article

Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik) is currently edited by Peter Winker

More articles in Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik) from De Gruyter
Bibliographic data for series maintained by Peter Golla ().