Informants in Organizational Marketing Research

Organizational research frequently involves seeking judgmental data from multiple informants within organizations. Researchers are often faced with determining how many informants to survey, who those informants should be and (if more than one) how best to aggregate responses when disagreement exists between those responses. Using both recall and forecasting data from a laboratory study involving the MARKSTRAT simulation, we show that when there are multiple respondents who disagree, responses aggregated using confidence-based or competence-based weights outperform those with data-based weights, which in turn provide significant gains in estimation accuracy over simply averaging respondent reports. We then illustrate how these results can be used to determine the best number of respondents for a market research task as well as to provide an effective screening mechanism when seeking a single, best informant.