Chapter Listing | Return to 1997 Topical Index


EMPIRICAL ANALYSIS

Data used for the empirical analysis were obtained from a survey conducted among its members by the Forum of Private Business (FPB) in January/February 1996. This is a biennial survey which provides one of the most comprehensive sets of data relating to the bank-small business relationship in the UK1. Approximately 14,000 questionnaires were distributed in a single mailing to a random sample of the membership. This resulted in 3843 usable responses which represents a response rate of around 27%. Of these around 1000 were used for subsequent analysis because of the completeness of the data. A follow-up telephone survey of non-respondents was used to check for non-response bias. The sample itself was not truly representative of the national population of small businesses as respondents tended to be larger (whether measured by sales revenue or number of employees), more likely to be located in the south east and the sample had a significantly higher proportion of manufacturing firms (and a lower proportion of agricultural businesses). None of these problems was considered to be sufficiently substantial to threaten the generalizability of the subsequent findings.

In general the operationalisation of variables was straight forward and clearly apparent from the empirical analysis, though aspects of quality and the nature of participation in the banking relationship require some discussion.

The Quality of Service Provision

For the purposes of this analysis, quality of service provision was conceptualised in terms of both functional and technical quality (Gronroos, 1984) and a set of variables were used to cover both the core product itself and aspects of its provision. Respondents were required to provide (on a 1 to 5 scale) their assessment of the importance and quality associated with a range of aspects of bank services2. To identify underlying dimensions, customers importance ratings were factor analysed (principal components extraction and varimax rotation) to identify those aspects of quality which were viewed similarly by customers. The factor analysis identified five underlying dimensions based on the perceived price, the core product, knowledge and the advice offered, efficiency in operations and accessibility. For reasons of space, subsequent analyses rely on a simple aggregate across the five dimensions.

Participation in the Banking Relationship

Respondents were asked to rate 11 aspects of their own behaviour towards their bank (provision of information, welcoming visits, negotiating on charges etc) and 13 aspects of their perceptions of their bank managers behaviour towards them (threatening reduced overdraft, ready to help in a crisis etc). These variables were used to provide a preliminary indicator of participation with respondents perceptions serving as a proxy for service provider participation. Both sets of data were subject to an exploratory factor analysis (principal components extraction and varimax rotation) to establish whether there were any underlying dimensions which corresponded to the conceptual definitions presented earlier. Items which cross loaded were excluded from the analysis, resulting in the solutions presented in tables 1 and 2.

Three dimensions of firm behaviour were identified and these are to some degree consistent with the initial conceptualisation. The first dimension concerns information flows, although no distinction is made between necessary and desirable information. The second dimension deals with the absence of fear or threat within the relationship and provides a very crude indicator of the atmosphere of the relationship (personal interaction). The third dimension is akin to the idea of duty and responsibility and measures the extent to which the respondent monitors their relationship with their bank.

The interpretation of the dimensions of bank manager behaviour is rather more problematic. The factor analysis identifies two dimensions; one of which relates to the extent to which the bank manager appears to actively or positively help while the second is more concerned with the extent to which the bank manager avoids doing anything negative which might harm the relationship. This latter dimension has some similarities with the notion of responsible behaviour while the first dimension appears to include both information sharing aspects of the relationship (advice and understanding) and personal aspects (trust and confidence). This is a reflection of the limitations and constraints on the measurement instruments used; nevertheless as the dimensions identified do not appear to be too unrealistic, they can provide the basis for further exploratory analysis.

1Thus, the data were not dedicated to research questions specifically posed in this paper. While this does raise some difficulties with respect to measurement these were thought to be outweighted by the extent and quality of the dataset.
2The set of variables available and the format of the likert scale were pre-determined by earlier surveys and the need to retain consistency.

Top of page | Chapter Listing | Return to 1997 Topical Index


1997 Babson College All Rights Reserved
Last Updated 06/06/98