Successful application of statistics to obtain valid conclusions to research questions has many challenges. One of the challenges is to identify the right statistical techniques and to apply them correctly. Another common challenge is insufficient statistical training and/or lack of appreciation for appropriate statistical methods of the subject matter researchers. However, insufficient training and lack of appreciation of the statistician for the subject matter research can also be a problem. Approaches to overcome these challenges are often not obvious and maybe can not even be taught in graduate school. To overcome them, it is important that the researcher and statistician start interacting at the time of planning an experiment and not after all the data have been collected. Overcoming these challenges is part of the fun of working as an applied statistician.
    In this talk I will discuss some statistical and some non-statistical challenges of working as an applied statistician. The statistical challenge I will discuss is one aspect of survival analysis that is relatively straight forward from a theoretical perspective, but can be quite complex from an applied perspective: time-varying covariates. I will provide a brief introduction to survival analysis and discuss the conceptual issues and issues relating to analysis and interpretation of effects of time-varying covariates. In addition, I will provide examples and discuss some of the challenges of the collaboration of statisticians and investigators in answering research questions. Three take home lessons, are to (1) consult your friendly applied statistician at the design stage of experiments and not after all the results have been collected and to make sure that (2) the statistician understands the basic concepts of the applied research question and (3) the subject matter researcher understands the basic concepts of the statistics being used.

Faculty Coordinators

John E. Boyer, Jr.

Phone:  (785) 532-0518

E-mail: 

Staff

Student

Coordinators