Human Perception of Outliers in Correlated Scatterplots Public
Downloadable Contentopen in viewer
Despite ample research analyzing how people recognize differences in data, one aspect that has largely gone unmeasured is how outliers affects these comparisons. This paper aims to provide a better understanding how people recognize differences in data by having participants decide which correlation is stronger (forced choice) when comparing scatterplots at different correlations with outliers. With 67 participants, we calculated a just noticeable difference (JND) at different correlation values. The results indicate that at all levels of correlation (e.g., .4 or .8) tested, people were less able to detect the stronger correlation for the scatterplots with outliers compared to scatterplots without outliers.
- This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.
- Date created
- Resource type
- Rights statement
Permanent link to this page: https://digital.wpi.edu/show/b2773x343