Monadic – not as simple as it might seem

It would seem that there is no simpler method than a monadic test if you intend to choose the optimal product formula, or a new packaging design. All you need is 2-3 cells with at least 100 respondents in each, and you can go ahead. Sure thing, but…. Monadic tests are characterised by very restrictive assumptions, which affects the sample selection process. Each person within a given leg should have an appropriate demographic and usage  equivalent in another leg.

We have to remember that users of a given brand have entirely different preferences than people reaching for competitive products. Heavy users of a given category shall evaluate the product formula far more positively than its light users or non-users will. Hence, if a given research agency does not ensure proper sample selection in order to include in each cell ‘the same type of respondents’ as regards their demographic and usage variables, it shall not be possible to compare the results between cells.

I have several times had an opportunity to view monadic test results we have been provided with by our Clients. Unfortunately, it was impossible to correctly interpret them, because the sample selection procedure was incorrect. I would not mention this problem if it were not relatively common. I remember a very nice speech given at the Research Congress presenting the results of a very interesting experimental test carried out for one of the Agency’s Clients.

I was really impressed with the idea of methodology used, scope of analysis …but when I asked how the respondents were selected for particular monadic cells, I was informed that the respondents had not been asked screening questions about the use of brands within the category. This is why researchers did not know who they actually surveyed, i.e. whether, for instance, in one cell there were no A brand users, while in another there were only brand B users. The study aimed at testing reactions to branded concepts of brand A. Hence, the methodology as such was fine, the approach to the research problem was creative…. but the results were totally useless.