Learning to experiment

Though I have been fascinated by the potential of controlled experiments since the first time I learnt about Milgram and Stanford prison experiments, it is only in last few days that I have begun to get initiaited into this methodology. Incidentally most of the research on power and status in the recent past have used lab experiments to establish causal effects. With a scope for greater creativity, experiments somehow seem more exciting than the typical field survey method. Of course a study that best combines experimental design with "reality check" (as in the Motherhood Penalty study) or a quasi-experimental design (as the famous blind auditions for symphony orchestra) seem to be even more interesting.

Anyways, as I begin to work with my research mentor on how one such experiment can be designed for studying the differential effects of power on various group outcomes, I seem to be stuck with one fundamental question. Consider an experimental design where one would like to study the effect of psychology of power on various negotiation outcomes. It appears from the literature that there are many ways to get the subject into condition of high power. But does when and how the subject is power-primed have differential impact.

For instance, consider a buyer-seller negotiation context. In one scenario, a subject playing the role of buyer can be primed with high power (e.g. through semantic prime or making him/her recollect a situation where he/she felt powerful, such as in interactions with subordinates or partner) before he/she gets into the negotiation context and takes the assigned role. Alternatively, the subject could be directly assigned to the negotiation context, with his/her buyer-role described as powerful (perhaps due to lesser dependency on seller or due to past successes in similar negotiations).

Some questions stemming from this -

• Are the two manipulations of power-priming different? If so, why?
• Are the activated self-concepts in subjects different in either cases?
• In the latter case, does the power manipulation continue to remain active even after negotiation role-play?
• What happens when a combination of both approaches is used? For instance, subject is semantically power-primed, but the buyer role he/she plays is described as low-power?

Perhaps, it is time to read more and brainstorm.