Skip to main content
Thirty-Third Annual Meeting of the Society for Research in Psychopathology

Full Program »

152. Toward dimensional conceptualizations of psychopathology: The role of data visualization

Each year, 2 million undergraduates in the US take an introductory or abnormal psychology course. To what degree do the graphs used in common Introductory and Abnormal Psychology textbooks reflect two hallmarks - distributions and statistical uncertainty - of the less categorical, more dimensional, conceptualization of psychopathology that has gained traction in recent years (e.g. RDoC)? In a survey of current texts, we found that these hallmarks were present in only 13 of 466 graphs. In a series of online studies (total N = 116), we assessed the potential impact of these omissions on readers’ conception of differences. We compared the recently invented sinaplot – which incorporates both of the abovementioned hallmarks via clever arrangement of individual data points – to the common bar graph, which, in the textbook context, usually lacks both hallmarks. We found that estimations of difference from bar graphs, far more than sinaplots, were exquisitely sensitive to data-irrelevant features such choice of y-axis range. Such features produced both inflation and compression of perceived difference, relative to ground-truth, depending on the context. We suggest that greater use of modern data visualization techniques in psychology textbooks may foster a more dimensional view of psychopathology that reflects progress in the field.

Heesu Kim
McLean Hospital

Jeremy Wilmer
Wellesley College

Elizabeth Grinspoon
McLean Hospital

Laura Germine
McLean Hospital and Harvard Medical School

 


Powered by OpenConf®
Copyright ©2002-2018 Zakon Group LLC