Dr Tom Cromarty
Interests: Paediatric Emergency Medicine, Medical Engagement and Leadership, Simulation, Quality Improvement, Research
Welsh Research and Education Network
Hot topics in research and medical education, in Wales and beyond
Dr Celyn Kenny
WREN Christmas Quiz
Davide Paccagnella, ST4
The days are shorter, the wards are busier, but on the bright side…Christmas is approaching!
Take a quick break from work with something light, like a statistics quiz! (there is also some trivia in there, don’t worry)
There is often more than one correct answer, as you will see. It all comes down to perspective.
1. How might confounding factors be avoided?
a. They just can’t be avoided
b. By matching individuals in the groups according to potential confounders
c. By mixing all the data together
d. By clarifying everything beforehand
2. A study where neither the patient nor the researcher knows which treatment the patient has been randomised to receive, is called a:
a. Good study
b. Single-blinded study
c. Double-blinded study
d. Not a clue study
3. What is the “number needed to treat”?
c. Depends on the circumstance
d. Depends on how much the researcher wants to get published
4. How can the confidence interval be calculated?
a. By using the standard error
b. By asking a statistician for advice
c. By subtracting your own confidence level from the level of confidence you have in the Health secretary
d. It does not matter, so it is not worth calculating
5. Likelihood ratios can be multiplied by pre-test odds to give post-test odds. True or false?
d. Likelihood ratios are an old-fashioned concept and, frankly, irrelevant.
a. measures the proportion of true positives that are correctly identified by the test
b. measures the proportion of false positives that are incorrectly identified by the test
c. Is often undervalued in Medicine
d. Is always a waste of time, according to some people
7. Type 1 error can be defined as...
a. The first error a researcher makes
b. The incumbent US president’s interpretation of the election result
c. The rejection of a true null hypothesis as the result of a test procedure
d. The content of this entire quiz
1 – b; 2 – c; 3 – c; 4 – a + b can both be true; d – may be true, if you are that kind of person; 5 – b; 6 – a, c, d may all be true depending on your views; 7 – you should know this, but I will respect your opinion.
Dr Annabel Greenwood