The role of inferences in consumer decision making : factors mediating process and outcome
Consumer research has frequently considered consumers' use of information during decision making. In instances where available information (both from memory and external sources) is inadequate for product choice, the consumer may make inferences to fill in informational gaps. Inference making, or consumers' generation and use of beliefs which go beyond available information, has received relatively little attention in the consumer information processing literature. Thus, the purpose of this dissertation is three-fold: 1) to examine the extent to which consumers make inferences during product evaluation and choice, 2) to better understand the nature of underlying inferential processes, and 3) to identify the individual and situational conditions under which inferences are likely to be made. A laboratory study was designed to examine naturally occurring inferences during product choice. The study used a 2X2X2 design, where 100 subjects 1) received either factual or evaluative product information, 2) chose the product for either a specific or a less specifically defined usage situation, and 3) were later classified into either high or low prior product knowledge groups. Subjects' verbal protocols were tape recorded while they chose between three hypothetical brands of 35mm cameras. They were at no time instructed to make inferences. A coding scheme was developed to measure subjects' inferential processes and outcomes. The study results indicate that, even when unprompted, subjects often made inferences during decision making. However, the extent and type of inference making varied across circumstances. First, high product knowledge consumers made significantly more inferences and used relatively different inferential processes than those with low product knowledge. In addition, type of information made little difference in low knowledge subjects' inferencing, but high knowledge subjects made significantly more inferences from factual than evaluative information. Finally, although not significantly different, high knowledge subjects made more inferences for specifically than less specifically defined usage situations, while there was little difference for the low knowledge subjects.