Results: Validity of Virtual Tests

General Mills conducted one of the most comprehensive studies assessing traditional vs. emerging methodologies. When the results of over 80 case studies were compared, their researchers found an almost perfect correlation between live surveys and Internet based approaches. These findings were replicated in a recent Stanford University study which found “striking similarity of distributions across nearly all questions” and “remarkable comparability of results”.

In some cases, it can be argued that the virtual results are even more accurate predictors of future behavior. Virtual simulations let researchers manipulate and control almost every aspect of the environment and the respondent’s interaction with it. Variables that can adversely affect a real world study like bad weather, product out of stocks and competitive activity are no longer factors – unless they are designed to be part of the study.

A number of companies on both the client and supplier side have validated virtual results against market share data, most commonly dollar and unit sales, and have found the correlations to be excellent.