The hypothesis holds that more options can backfire, cutting the motivation to choose at all or the satisfaction with whatever gets chosen. Strong instances turned up in the laboratory and in the field, and the idea became common knowledge. Other studies found nothing, or the opposite. Scheibehenne, Greifeneder and Todd pooled the lot. What is the average?
Virtually zero. Across sixty-three conditions from fifty published and experiments, with a combined five thousand and thirty-six participants, the mean came out essentially nil, with considerable variance between studies. Further analysis turned up several potentially important for choice overload, but no sufficient conditions: nothing that reliably produces it.
The choice overload hypothesis is specific about the harm it predicts. An increase in the number of options to choose from may lead to adverse consequences, among them a decrease in the motivation to choose, or in satisfaction with the option finally chosen. The idea moved quickly into general circulation, helped by memorable demonstrations.
The literature behind it was never that tidy. Many studies found strong instances of choice overload in the laboratory and in the field. Others found no such effects at all, and some found the reverse: that a wider range helps people choose and leaves them more satisfied.
Scheibehenne, Greifeneder and Todd pooled sixty-three conditions from fifty published and unpublished experiments, five thousand and thirty-six participants in all. The mean effect size came out virtually zero, with considerable variance between studies. That combination is the interesting part: a literature that disagrees with itself as much as this one averages to nothing.
They then looked for the conditions under which the effect does appear. Several potentially important preconditions showed up, but no sufficient conditions could be identified, which is to say nothing that makes overload happen dependably. The authors leave the door open. Idiosyncratic moderators proposed in single studies may still explain when and why choice overload reliably occurs, and they review those studies and point to where the work should go next.
wat. Too much choice sometimes paralyzes people. On average, across fifty experiments, it does nothing at all.
wat wat
A finding can be real in particular setups and still average to nothing. That gap is where a lot of popular psychology lives.
How do we know?
A meta-analysis in the Journal of Consumer Research covering sixty-three conditions from fifty experiments, published and unpublished, testing for overall effect size, for variance between studies, and for conditions under which the effect appears.
What is still unknown
A mean near zero with considerable variance is not proof the effect is absent. The authors allow that idiosyncratic moderators proposed in single studies may yet explain when and why it occurs reliably.
Learn something?
Where to next
Why did the people who paid full price show up to more plays?decision making
What happened when the Dunning-Kruger effect was run on random data?replication
What was left of the marshmallow test when they ran it on ten times as many children?replication