An analysis of the effect of confirmation bias on industrial radiography

dc.creatorRomero, Henry Abraham
dc.date.accessioned2016-11-14T23:14:15Z
dc.date.available2011-02-18T19:48:47Z
dc.date.available2016-11-14T23:14:15Z
dc.date.issued1997-05
dc.degree.departmentIndustrial and Systems Engineeringen_US
dc.description.abstractThis experiment was undertaken to ascertain the current performance, the variables affecting performance, and performance improvements of a dynamic visual inspection process, real time radiography, used at a Department of Energy laboratory. The system was being used to determine contents of 5 5-gallon mixed waste drums prior to storage. This experiment started with a task analysis in which critical tasks necessary for successful completion of a barrel assessment were determined. One of the factors expected to shape the performance was the provision of a shipper's manifest. The reason is that the manifest induces a confirmation bias on the part of the operator concerning what to expect in the barrel. A literature review was performed concerning perception and industrial inspection processes. From this review, it was determined that this process had not been adequately studied in the literature. One of the main differences was that the confirmation bias was elicited in the literature by the use of tachistoscopic cues which were not feasible in this process. A repeated-measures design experiment was developed that utilized specifically created drums to test the confirmation bias as well as effects of experience and Mental Fatigue. The responses from this experiment were categorized into three binomial distributions: correct identifications, misclassifications. and detection failures. The results were that this experiment failed to note an effect due to the confirmation bias. In addition, this experiment failed to note impact from the Mental Fatigue or experience variables. In examining experience effects, it appears that between three and nine months time on task, the performance of the operators reached a level and did not significantly improxe. However, individual differences alone could have accounted for these results.
dc.format.mimetypeapplication/pdf
dc.identifier.urihttp://hdl.handle.net/2346/12036en_US
dc.language.isoeng
dc.publisherTexas Tech Universityen_US
dc.rights.availabilityUnrestricted.
dc.subjectImaging systemsen_US
dc.subjectQuality controlen_US
dc.subjectIndustrialen_US
dc.subjectRadiographyen_US
dc.subjectQuality control inspectorsen_US
dc.titleAn analysis of the effect of confirmation bias on industrial radiography
dc.typeDissertation

Files