Experiment Design: Marie Kondo Effects

R, Statistics, mTurk, Choice-based Conjoint Design

April 20, 2019

In this project, I...

  • Design an experiment to determine the effect of Marie Kondo on minimalist attitudes.
  • Use a choice-based conjoint design to determine effect.
  • Recruit study participants using Mechanical Turk.
  • Analyze results using R

ABSTRACT: Marie Kondo is a self-improvement celebrity currently enjoying tremendous popularity. She promotes an aesthetic of minimalism focusing on retaining those things that “spark joy,” and discarding the rest. Our experiment aims to determine the extent--if any--that exposure to Kondo and her philosophy changes viewers’ attitudes towards the accumulation of things around them. Using a Choice-Based Conjoint analysis technique, we find that even a brief exposure to Marie Kondo’s technique has a small but significant effect, bringing attitudes closer to Kondo’s minimalism. This shift primarily affects attitudes towards books and clothing, and is stronger in women than in men.

This was a collaborative group project, and as such there was not a single area where I had full responsibility. We developed the survey questions as a team and tested them on a small pilot group before running the full experiment on 250 participants. In the final analysis, I focused primarily on the analysis by response group.

Read the full paper here.