|Title||Puzzles and peculiarities: How scientists attend to and process anomalies during data analysis|
|Publication Type||Conference Paper|
|Year of Publication||2004|
|Authors||Trickett, SB, Schunn, CD, Trafton, JG|
We used an "in vivo " methodology to investigate how scientists working alone or with a single colleague in informal data analysis sessions deal with anomalous data. We focused on the extent to which several scientists paid attention to anomalies, both immediately and over time, and the processes by which they addressed them. We found that the scientists paid more immediate attention to anomalies than to expected phenomena and also that they continued to devote more attention to anomalies, in some cases over considerable periods of time. We further identified a pattern of response to anomalies that differed from the response to expected phenomena. This pattern included identifying specific features of the phenomenon, proposing a hypothesis to account for it, elaborating that hypothesis by reference to the visual display, and considering the anomaly in the context of the surrounding dataset. We consider how our results mesh with those of previous work and propose some possible explanations for the differences.
NRL Publication Release Number: