Chapter 7 Discussion of Results, Conclusions and Suggestions for
7.4 Possible Directions for Future Research
The thesis prompts the need for future research along three distinct strands.
Firstly, the Rebuy Case itself leaves several unaddressed aspects of enquiry for which potential actions to obtain answers are suggested. While the developed research method successfully addresses the research questions within the confines of the Rebuy Case, the second suggested strand of future research focusses on the expansion of the Rebuy Case to include much larger populations of buyers. The final research strand suggested addresses potential issues of replication.
7.4.1 Intra Case Refinement
Analysis of the results obtained through the Rebuy Case have identified relationship management behaviours and value perceptions that represent a divergence from those that would have been predicted by certain strands of extant thinking. The research design utilised in the study is not, however, intended to establish the underlying causes of these gaps43. A clear potential direction for future research effort would be the establishment of causal effects.
The first potential area for intra case refinement arises from the view expressed by Wilson (1995:344) and Cousins and Spekman (2003:26) who suggest, as noted previously, that the general transition from adversarial buying to a more co-operative model is occurring slowly but perceptibly.
Accepting this premise gives rise to an expectation that the related gap will continue to narrow. Replicating, as far as practicable, the investigation at some future time would therefore be expected to represent a vehicle whereby any reduction in the identified gap could be qualitatively examined.
The identification of possible casual variables represents a further area for intra case refinement of the research method. For example Anderson, Thomson and Wynstra (2000:325) and Kowalkowski (2011:278) suggest that the effect of strong budgetary constraints or the deployment of specific reward and measurement systems can influence buyer behaviours. Similarly, Hansen (2009:234) suggests that time constraints acting on the buyer will significantly modify their behaviour, while Cousins (2005:403) and Svahn and Westerlund (2009:174) observe that the firm’s choice of strategic direction will dictate
43 Potential causal effects are discussed in Section 7.2
195 the role that played by the purchasing function. Through the addition of diary and diary-interview prompts these potential variables could, with relative ease, be investigated utilising a broadly similar research design.
7.4.2 Extra Case Expansion
In order to improve the validity of the findings from the Rebuy Case it would be desirable to enlarge the scale of the study because, as Firestone (1993:17) and Yin (2003:19) note, a single study generally provides only ‘base level’
support for any resultant findings. One possible alternative would be to develop a survey in order to generate data across a wider population.
However, survey approaches, while ideally suited to exploring large populations, are extremely limited in their ability to explore ‘context’
because the survey designer constantly struggles to limit the number of variables under consideration (Yin, 2003:13) and are more appropriately used appropriate where the objective is to ‘generalise’, rather than to ‘develop’
the findings (Jankowicz, 1995:183). Onwuegbuzie and Leech (2006:245) advocate the use of confirmatory thematic analyses in which replication of qualitative studies are conducted to assess the replicability of previous emergent themes, considering that such confirmatory techniques help to legitimise previous qualitative findings, interpretations, and conclusions.
Exploring qualitative context should, therefore, remain a key focus of any further research in order to develop the potential theoretical implications of either the role of buyer value perception or the incidence of buyer relationship management behaviours.
Yin (2003:19) further cautions against failing to recognise that the objective of subsequent cases should focus on replication and not a misplaced application of sampling logic that assumes by simply increasing the number of cases examined either the validity or the reliability of the study will increase (Yin, 2003:47). Against this background any further Cases identified should clearly be selected with care.
A further possibility to expand the scale of the investigation would be to access what Lusch (2011:17) recognises to be an almost unlimited data source:
‘the most common data are words, as in the billions of conversations and communication that characterize Web 2.0. Thus tools are being developed to automatically analyse large textual databases’
196 Lusch, Liu and Chen (2010:72) concur not only as to the suitability of the characterisation of trading markets as conversations but also share the belief that computational linguistics, sentiment analysis, and network analysis will become increasingly important vehicles by which these conversations can be better understood and analysed (Lusch, Liu and Chen, 2010:75). An immediately attractive research proposition would be the expansion of the research method developed within this thesis to utilise the web based tools, as recognised by Lusch, Liu and Chen, to analyse the data contained within transactional ‘electronic conversations’. It is recognised that gaining access to web based sources of data may well be problematic, but the incentive to negotiate access would be the prospect of considerably expanding the number and nature of the transactions that are able to be researched.
An alternative means of extra case expansion is suggested by the observation of Moriarty and Bateson (1982:190), that:
‘Contacting potential decision participants is much more difficult than gaining their cooperation’
From this perspective conducting further research utilising respondents with membership of ongoing, large scale, related research programmes has obvious potential benefits. By way of illustrative examples, the Contract Management Benchmarking44 research project benefits from the input from over ninety organisations across a wide range of sectors, while the Cambridge Service Alliance45, whose research objectives include enhancing the value perceptions for service users, also enjoy the support of a wide range of businesses. The prospect of cross programme research collaboration is therefore a potential avenue worthy of exploration.
7.4.3 Replication
Bryman and Bell (2003:33) note that research protocols must be replicable by someone else, but acknowledge that it is difficult to ensure in business research that the conditions in a replication are precisely the same as in the original study (Bryman and Bell, 2003:85). Yin (2003:33) notes that the logic of replication (within case studies) should aim towards analytic generalisation and not necessarily aspire to reproduce the precise results of the initial study
44 http://www.linkedin.com/groups/Contract-Management-Benchmarking-wwwcmb2013-com- 4381858
45 http://www.cambridgeservicealliance.org/
197 but rather to support an emerging theory or initiate the generation of a rival theory. While Firestone (1993:22) notes that qualitative methods should not be avoided for fear that replication may appear problematic.
The task of replication is facilitated through the use of content analysis in which coding schemes can be developed and made available (Bryman and Bell, 2003:206). The further development of standardised coding schemes applicable to the understanding of buyer behaviour and buyer value perceptions represents a potentially fruitful direction for further research especially if these coding schemes can be coordinated with the expected developments in computational linguistics, sentiment analysis, and network analysis Lusch, Liu and Chen (2010:72).