on the significance of statistically insignificant results in consumer behavior experiments

Experimentation is the sine qua non of consumer behavior research, and much of what is thought to be known about the behavior of consumers is based on findings from experiments. however, many articles that report consumer behavior experiments contain one or more results that are significantly insignificant. that is, one or more experimental results are so unusually weak or minuscule that they are unlikely to have come about by chance. as such, significantly insignificant results can be due to the “failure” of the theory underlying an experiment and/or the flawed design or implementation of an experiment. consequently, significantly insignificant results have implications for the theories and methodologies employed in consumer behavior experiments, the quality of conclusions drawn from the experiments, and the credibility of the consumer behavior research discipline as a whole.