Interorganizational dependence in marketing channels : a test of a model and methodology

A contrived laboratory situation called a parasimulation was used as the experimental setting chiefly because it offered a high level of control and allowed accessibility to data. Subjects were college students who were taking a marketing channels course. The experimental variable was the number of potential options available to the channel firms. It was posited that the greater the potential options, the less dependent; and the fewer the number of potential options, the more dependent. Variation in the number of options was provided by differing the number of firms at each level of the channel.

Dependence was measured by two methods. Actual costs and benefits of relationships were obtained from parasimulation firms. Dependence was then calculated utilizing the dependence model. Also, firm executives were asked their perceptions of their dependence on the other firms. Power was measured by asking the extent of control the other firms had over marketing decision variables as perceived by firm executives. Conflict was measured by the perception of executives with respect to certain conflictive behaviors exhibited by the other firms.

Hypotheses were then developed. Dependence hypotheses tested the validity of the dependence model by predicting levels of dependence. The relationship between the two methods of dependence measurement was also tested. Power hypotheses tested the relationship between power and dependence. Conflict hypotheses tested the relationship between conflict and dependence.

The hypothesized relationship between dependence and the number of options available was not supported by the test results. Since these hypotheses were used to develop the power and conflict hypotheses, their rejection indicated that the power and conflict hypotheses would not be supported. This contention was the case for the power and conflict hypotheses. The most promising cause for the lack of support concerned the executives' perception of the number of potential alternatives. Consumer behavior research has shown that consumers only perceive a subset (the evoked set) of all potential brands. If channel executives also employed this mechanism, the number of potential alternatives they perceived and the number of alternatives as determined by the parasimulation may not have been the same.

Firms were divided into large and small set groups based on the sizes of their evoked sets. . Firms with small set sizes should have been more dependent, while firms with large set sizes should have been less dependent. Hypotheses for dependence, power, and conflict were reformulated using the size of the evoked set as the experimental variable. Test results for the dependence and power hypotheses supported the operationalized dependence model and its relationship with power. These findings indicated that the problem with the original analysis was that of perception of the number of potential options.

The relationship between conflict and dependence was not supported. Failure to measure all the components specified by the overall channel relations model as producers of conflict was offered as the most probable cause.

There were several implications from the study. First, the further use of the parasimulation as a research tool was recommended. Methods for improving the measurement of dependence and conflict were also suggested.

Future research directions included suggestions for replication and effort in the study of goals, conflict causes, and the relationship between dependence and power bases.