VISUAL INSPECTION OF DATA REVISITED - DO THE EYES STILL HAVE IT

In behavior analysis, visual inspection of graphic information is the standard by which data are evaluated. Efforts to supplement visual ins pection using inferential statistical procedures to assess interventio n effects (e.g., analysis of variance or time-series analysis) have me t with opposition. However, when serial dependence is present in the d ata, the use of visual inspection by itself may prove to be problemati c. Previously published reports demonstrate that autocorrelated data i nfluence trained observers' ability to identify level treatment effect s and trends that occur in the intervention phase of experiments. In t his report, four recent studies are presented in which autoregressive equations were used to produce point-to-point functions to simulate ex perimental data. In each study, various parameters were manipulated to assess trained observers' responses to changes in point-to-point func tions from the baseline condition to intervention. Level shifts over b aseline behavior (treatment effect), as well as no change from baselin e (no treatment effect or trend), were most readily identified by obse rvers, but trends were rarely recognized. Furthermore, other factors p reviously thought to augment and improve observers' responses had no i mpact. Results are discussed in terms of the use of visual inspection and the training of behavior analysts.