Effect of perceived risk and context on consumer inferencing about missing information

Abstract

Consumers adopt different strategies for evaluation of a brand when information regarding an attribute is not available. One of those is differencing strategy in which consumers infer or ‘construct’ values for the missing attributes and use the same for evaluation of the brand. Past researchers have shown that inferencing has significant effect on likelihood of purchase of a brand. However, determination of the proportion of consumes who adopt inferencing strategy has received less attention. Simmons and Lynch (1991) attempted this and found that inferencing occurred only in 13 to 19 percent of the cases. They measured incidence of inferencing by asking the respondents to describe their thought process during evaluation and identifying those who had mentioned the world infer in their description of the process. Since inferencing is a cognitive process, such elicitation of the process is likely to result in under reporting. There has been no other study to corroborate the above findings. The present study attempted to measure inferencing using indirect methods. The study also examined influence of consumer’s perception of risk associated with a choice and the composition of the set of brands available to the consumer at the time of evaluation, on inferencing. Respondents were classified into those who inferred and those who did not, by comparing their overall evaluation of brands with overall evaluations of the same brands by respondents in control conditions. The composition of the set of brands presented to the respondents was manipulated by varying the number of brands with complete and incomplete information. Different purposes of buying were used as proxy for different levels of perceived risk. Data from 292 first year postgraduate management students of three management institutes in Ahmedabad were analyzed using ANOVA.’ The major findings of this study are; • Inferencing occurred in about one third of the cases, far more than previously reported. • It was found that respondents were not risk averse while making an inference. This is in contrast to the normal tendency of people to be risk averse. • The composition of the set of brands available to the respondents at the time of evaluation had influence on inferred value and probability of making an inference As inferencing is a spontaneous cognitive process and influence consumer’s evaluation of brands, it is absolutely vital for marketers to understand the phenomena. The findings of the study would provide marketers a better understanding about inferencing. It has the potential to help them formulating their communication strategy and in designing advertisements in a comparative format. The findings regarding higher incidence of inferencing provide lead to further research under the Information Integration Framework. The study broadens the scope of inference making phenomena and opens up possibility of further research linking inferencing to choice of store, selection of criteria for evaluation etc.