2017 AAPM Annual Meeting
Back to session list

Session Title: Statistical Failings that Keep Us in the Dark & Practical Statistics
Question 1: Various publications have found a higher than expected incidence of p-values immediately below p=0.05 as compared to p-values immediately above p=0.05. Some possible reasons for this over-representation could include:
Reference:B. Ginsel, et al., The distribution of probability values in medical abstracts: an observational study., BMC Res Notes, Nov 26, 2015.
Choice A:Publication bias.
Choice B:Statistical fraud.
Choice C:Methodological errors (selective reporting, selective analyses, underpowered analysis).
Choice D:A and C.
Choice E:All of the above.
Question 2: A p-value can best be described as the probability, under a specified statistical model, that:
Reference:R.L. Wasserstein and N.A. Lazar, The ASA’s statement on p-values, context, process, and purpose, The American Statistician, accepted version published online 3/7/2016.
Choice A:The null hypothesis is true.
Choice B:A statistical summary of the data would be equal to or more extreme than its observed value.
Choice C:The statistical summary of the data was a random coincidence unique to the given data under analysis.
Choice D:The results would not hold up if the experiment is repeated.
Choice E:The data has sufficient statistical power to detect a given effect.
Question 3: Effect sizes are more important than p-values because:
Reference: H. Motulsky, Intuitive Biostatistics: A nonmathematical guide to statistical thinking, Oxford Univ Press, (2014), pp. 133-134.
Choice A:They are unitless measures.
Choice B:They are sample independent.
Choice C:They tell you how practically meaningful your results are.
Choice D:They are unrelated to the p-value.
Question 4: True or False: The ½ standard deviation method is an accepted criterion for determining clinical significance in quality of life (QOL) studies?
Reference:J. Sloan, et al., “Clinical significance of patient-reported questionnaire data: another step towards consensus”, J. Clinical Epidemiology 58, 2005.
Choice A:True.
Choice B:False.
Question 5: What does a high R2 value mean for a simple linear regression model?
Reference:An introduction to categorical data analysis 2nd edition. Alan Agresti, John Wiley and Sons, Inc. Hoboken, NJ 2007; p. 144.
Choice A:The model accounts for a lot of the data variability .
Choice B:The null hypothesis should be rejected.
Choice C:The data is linear with a positive slope.
Choice D:There are no outliers in the data.
Question 6: A researcher collects a database of T2W MR cases demonstrating glioblastoma. He wants to compare tumor area between manual outlines and a new computer algorithm. The most appropriate test would be:
Reference:Bland and Altman, “Statistical methods for assessing agreement between two methods of clinical measurement,” Lancet 327, 307 (1986).
Choice A:Pearson Correlation.
Choice B:Spearman Correlation.
Choice C:Fleiss’ Kappa.
Choice D:Bland-Altman Analysis.
Back to session list