Question 1: What percentage of published research findings is false?
|
Reference: | J.P.A. Ioannidis. “Why most published research findings are false.” PLOS Medicine 2, no.8 (2005). |
Choice A: | <5% |
Choice B: | >5% but <25% |
Choice C: | >35% but < 50% |
Choice D: | >50% |
Question 2: Breast cancer incidence in 2012 compared to 1972 was:
|
Reference: | Cancer Statistics, 2016. CA Cancer J Clin 2016;66:7-30. |
Choice A: | Higher. |
Choice B: | About the same. |
Choice C: | Lower. |
Question 3: Mortality (per capita) from all cancers for males in 2012 compared to 1930 was:
|
Reference: | Cancer Statistics, 2016. CA Cancer J Clin 2016;66:7-30. |
Choice A: | ~50% lower |
Choice B: | ~10% lower |
Choice C: | About the same. |
Choice D: | ~10% higher. |
Question 4: High impact journals are more likely to publish results that later turn out to be false.
|
Reference: | B.Brembs, K.S. Button, M.R. Munafo. “Deep impact: unintended consequences of journal rank.” Frontiers of human neuroscience 7 (2013): 291.
|
Choice A: | False. |
Choice B: | True. |
Question 5: Assuming that 0.1% of women at the time of mammography exam have breast cancer, and that sensitivity and specificity of mammography is 95%, what’s the probability that a woman with a positive test has breast cancer?
|
Reference: | https://en.wikipedia.org/wiki/Bayes%27_theorem |
Choice A: | ~0.19% |
Choice B: | ~1.9% |
Choice C: | ~19% |
Choice D: | cannot be estimated. |
Question 6: Donald E. Herbert, Ph.D.....
|
Reference: | AAPM Report #43 |
Choice A: | Co-invented the Linear-Quadratic (LQ) model and demonstrated, using statistically adequate methods, that it is consistent with available radiobiological and clinical data. |
Choice B: | Demonstrated, using statistically adequate methods, that the LQ model “doesn’t fit” the relevant data. |