A page from the OpenDepot.org service

Schuetze, Joerg (2009) Trust in Data - Trust to make an appointment Surveys - Weightings - Manipulations Telephone Survey to evaluate how satisfied customers were with their bank Graz 1998 [German]. In: Vertrauen in Daten, Vertrauen zum Termin. Lulu, Graz, pp. 1-216. ISBN 9781409259411

PDF - Published Version Restricted to Repository staff only until 09 May 2085. Download (3668Kb) | Request a copy

Abstract

A telephone survey was carried out in Graz between April and July 1998. The aim of this survey was to evaluate how satisfied customers were with their bank and to offer non-binding advice. This resulted in almost 2500 questionnaires which were used to analyse the possibilities and limits of new customer acquisition. Data collection must fulfil strict statistical requirements to be considered for the analysis used. One fundamental requirement for statistical inference is representativity (i.e. the non-biased random selection of a sample). This ideal situation is, however, in reality extremely rare. Here, comparisons with both internal and external data as well as weighting should be considered. These comparisons have been used insufficiently in the past. This thesis focuses on interviewer-specific biases. The sample size used is large enough to extend the conclusions to other surveys. The analysis concentrates on the stability of the sizes of certain variables. This could be regarded as a kind of sensibility analysis. Survey data ordered by interviewer ? are compared with external data using various weighting processes. By doing this, it is possible to achieve a certain trust in the data. This trust is the basis for a further analysis. The thesis is consistent in showing how interviewer bias and manipulations can be detected using different comparisons. When the results of all interviewers are considered together, these detected biases balance each other out to a large extent. This work delivers important guidelines for the analysis of data from small sample sizes. This is especially relevant for internet-based surveys and telephone surveys. The variables "investor type", "profession", "age" and "sex" were the variables which were most decisive in forming a customer's level of trust in his/her bank. People with a higher level of trust in their bank were more likely to agree to the telephone survey and make an appointment to see a consultant. A comparison of the appointments which lead to contracts (health insurance, extra services etc.) shows that the telephone could play an important role as a marketing instrument especially amongst the younger, perhaps less risk-averse people.

Actions (login required)