"Everything should be made as simple as possible, but not simpler." - Albert Einstein

Seminar Announcement

Developing a practical predictive model of commercial performance with digital video recorder (DVR) data.

Leigh Engelhardt, M.S. Candidate, Department of Statistics, Colorado State University.

Monday, April 6, 2009

11:00 am, Weber 223

ABSTRACT

The introduction of the Digital Video Recorder (DVR) in 1999 allowed viewers to fast forward and skip through television commercials more easily.  While commercial avoidance was not a new behavior, DVRs made this behavior far more measurable.  A benefit to the advertising industry amidst the gloom of commercial avoidance is that the DVR adds the ability to monitor real audience behavior.  As the usage of DVRs increased, manufacturers like TiVo made viewing behavior far more measurable by logging user behaviors and storing the data over time.  The availability of this new data begs us to ask, can we better understand consumer skipping behaviors as well as quantify a commercial’s success?  Using TiVo Stop||Watch ratings data, we explore DVR viewing data to better understand commercial performance and consumer skipping behavior.  The data are analyzed using linear models that investigate the relationship between commercial performance and individual predictors.  General linear models investigate the relationship of commercial performance with multiple predictors and interactions.

Advisery Committee:
Advisor: Phillip Chapman
Committee: Chihoon Lee
Outside: Kathleen Kelly, Marketing