Study Designs
Describing Data
Screening Tests
Sampling and Measurement
Estimation
Randomised cluster design and
intention to treat analysis
intention to treat analysis
Introduction to Hypothesis
Testing
Testing
Hypothesis Testing II
100
This type of observational study investigates individuals who are observed only once and is used to describe the prevalence of exposures and outcomes.
What is a cross-sectional study?
100
This type of categorical variable can take values that can be logically ordered or ranked, such as "strongly agree, agree, disagree, strongly disagree" or "small, medium, large."
What is an ordinal variable?
100
Screening tests target this population and prioritize high sensitivity to avoid missing potential disease, while diagnostic tests target at-risk populations and prioritize higher specificity to prevent over-treatment.
What is the healthy population?
100
BONUS WISDOMS TRIVIA
This person has been in a hot air balloon
Who is Becky Breau?
100
This type of estimation provides one number which estimates the true value of the population parameter and is considered the 'best guess' of an unknown population parameter.
What is point estimation?
100
This is the 'social' equivalent of the placebo effect, occurring when subjects know they are being investigated or 'watched', causing them to feel special and potentially changing their behavior.
What is the Hawthorne effect?
100
BONUS WISDOMS TRIVIA
This person learned how to play drum kit when they were 12 years old.
Who is Adele Bibault.
100
This is the probability of detecting a true effect by correctly rejecting the null hypothesis when it is false, and it depends on sample size, effect size, and variability. Low values of this increase the risk of Type II errors
What is statistical power?
200
This type of bias occurs when subjects are interviewed and asked to recall past exposures, and cases may have thought more about their exposure than controls.
What is recall bias?
200
When a distribution has a longer right tail with a few extreme high values, it is described as having this type of skew, and examples include income or sperm concentration data.
What is positive skew (or right skew)?
200
BONUS WISDOMS TRIVIA
This person once had their ears pierced.
Who is Simon Boushel?
200
This characteristic of a measurement tool refers to consistency—meaning it gives the same results when repeated under the same conditions, as opposed to validity which measures what it claims to measure.
What is reliability?
200
RANDOM TRIVIA
The name of the University of Limerick's Mascot.
Who is Wolfie?
200
This method of analysis is the default in randomized controlled trials, following the principle "once randomized, always analysed" by analyzing patients in their originally assigned group regardless of which treatment they actually received.
What is intention to treat (analysis)?
200
This hypothesis is the one being tested and is usually the hypothesis of no difference, such as "there is no difference between the means for two groups" or "the risk of an event is the same for two groups."
What is the null hypothesis (or H₀)?
200
BONUS WISDOMS TRIVIA
This person's cousin's husband went to school with the actor who plays Neville Longbottom.
Who is Leanne Kong?
300
In this experimental study design, each participant acts as their own control and receives both treatments separated by a washout period, making it useful for stable chronic conditions.
What is a cross-over trial?
300
For positively skewed distributions, this measure of centrality is preferred over the mean because the mean can be misleadingly inflated by extreme values, and it represents the point below which 50% of data lies.
What is the median?
300
This outcome occurs when a screening test incorrectly classifies someone without the condition as high risk, and its rate equals 100 minus the specificity.
What is a false positive?
300
RANDOM TRIVIA
The length of The Living Bridge, in metres.
What is 350m?
300
The formula for an interval estimate is calculated as the point estimate plus or minus this quantity, which represents the range of uncertainty around the estimate.
What is the margin of error?
300
BONUS WISDOM'S TRIVIA
This person has a condition known as Exercise Urticaria.
Who is Morgan Wilson?
300
This statistical measure is a probability from 0 to 1 that determines the likelihood that an observed outcome is the result of chance, and the smaller it is, the more evidence there is for rejecting the null hypothesis.
What is a p-value?
300
When data within one or more groups are not normally distributed, you should use these types of hypothesis tests that compare medians rather than means across groups.
What are non-parametric tests (or non-parametric hypothesis tests)?
400
BONUS WISDOMS TRIVIA
This person has an IgE-mediated hypersensitivity to chinchilla allergens.
Who is Mattew Garland?
400
In the Axelsson et al. (2011) study comparing sperm concentrations between two cohorts of Swedish men (2000-2001 vs 2008-2010), researchers used this graphical summary to show that both groups had similar medians and variability, with positively skewed distributions.
What is a box plot (or side-by-side box plots)?
400
RANDOM TRIVIA
The number of litres of water that pass through the River Shannon everyday.
What is 18 billion litres.
400
This type of validity assesses how a questionnaire relates to other external criteria, such as using a well-being questionnaire to screen for depression.
What is criterion validity?
400
A wide confidence interval points to lack of this quality, whether the difference is statistically significant or not, and serves as a warning against over-interpreting results from small studies.
What is precision (or information)?
400
These trials test the effectiveness of an intervention in a real world setting with typical patients and clinicians, and often randomize at a group level such as hospitals or clinics, making them generalizable to routine practice.
What are pragmatic trials (or pragmatic clinical trials)?
400
This decision level, usually set at 0.05, is the probability below which you reject the null hypothesis, resulting in a statistically significant result.
What is the significance level?
400
This test is used to compare means of two dependent groups, such as when the same group is measured twice at baseline and follow-up, as opposed to an independent samples t-test.
What is a paired t-test?
500
A researcher wants to investigate whether a training programme for community mental health therapists reduces challenging behaviours in children with autism spectrum disorder. To prevent assessment bias, what blinding technique should be used so that those evaluating the children's behaviour don't know which therapists received the training?
What is double-blind (or blinding assessors to treatment allocation)?
500
BONUS WISDOMS TRIVIA
This person once came 4th place in a wakeboarding competition.
Who is Meaghan Walsh?
500
This measure, also called the detection rate or true positive rate, represents the percentage of those with the condition who are correctly identified as high risk by the screening test.
What is sensitivity?
500
This probability sampling method is mainly used to improve precision by dividing the population into groups based on known information like age or gender, with sample sizes proportional to population breakdown, and units sampled randomly within each group.
What is stratified random sampling?
500
In the resistance training study for neck/shoulder pain, the 2-minute daily training group had a mean change in pain intensity of -1.3 compared to 0.1 in the control group, with a 95% CI for the difference of [-2.0, -0.7]. Since this interval does not include this value, it provides evidence of a pain reduction.
What is zero?
500
In the Know Your Midwife trial, the caesarean rate was 16% in refusers, 7% in acceptors, and 7% in controls. The higher rate in refusers was most likely because women who refused were different from those who accepted, illustrating why this type of analysis should be avoided as it deviates from original randomization.
What is "as treated" analysis?
500
RANDOM TRIVIA
This is the most visited attraction in Ireland.
What is the Guinness Storehouse?
500
In Example 3, researchers investigated whether there is a relationship between having diabetes (Yes/No) and hypertension control (Yes/No) in 78 patients, with a p-value of 0.008 from the test. Since both variables are categorical and some cells had small frequencies, what statistical test was used?
What is Fisher's exact test?
600
In the famous Framingham Heart Study (1948), researchers identified a group of subjects and followed them over time to see if cardiovascular disease occurred, comparing subgroups with different levels of exposure to risk factors like smoking and diet. What type of observational study design is this?
What is a cohort study (or prospective cohort study)? 
600
RANDOM TRIVIA
The number of days in total will be be in medical school, not including weekends, holidays, and vacations.
What is approximately 340 days?
600
In the Neural Tube Defect (NTD) screening example with 46 true positives, 25 false positives, 15 false negatives, and 1208 true negatives, the positive predictive value was 65%. This means that for every 100 women screened as high risk, how many are expected to actually have a baby with NTD?
What is 65 (women)?
600
In the Australian Melanoma Family Study investigating genetic risk factors, researchers were concerned that case participation might depend on whether patients believed sun exposure caused their melanoma, while population-based controls might be more likely to participate if they had a family history. This is an example of what type of bias?
What is non-response bias?
600
A lipid-lowering medication study of 100 patients showed an average cholesterol reduction of 0.3 mmol/L with a 95% CI of [0.1824, 0.4176]. Given that a reduction of 0.5 mmol/L is considered clinically important, what can you conclude about this medication's clinical significance despite the interval showing statistical significance?
What is the medication may not be clinically important (or the confidence interval does not include 0.5 mmol/L, so we cannot be confident it achieves the minimum clinically important difference)?
600
RANDOM TRIVIA
What condition is commonly experienced by medical students?
What is Medical student's disease?
600
In the sleeping pill study with 50 participants, the sample mean sleep duration was 7.5 hours compared to the population mean of 7 hours, with a p-value of 0.0049. Since 0.0049 is less than 0.05, what decision should be made about the null hypothesis that the pill does not affect sleep duration?
What is reject the null hypothesis (or conclude the sleeping pill does affect sleep duration)?
600
RANDOM TRIVIA
This was a common nickname for Limerick in the 19th and early 20th century.
What is Pigtown?
700
RANDOM TRIVIA
This is the year Nancy Blake's bar was built.
When is 1956?
700
In a Tanzanian fertility study by Kasililika et al. (2021), researchers examined 96 women to explore the relationship between supplement use (Yes/No) and fertility status (Infertile/Fertile). Since both variables are categorical, what type of numerical summary table displays frequencies and percentages for each combination of these two variables?
What is a cross-tabulation (or contingency table or two-way table)?
700
In the PSA screening study of 100 men for prostate cancer, the test had a sensitivity of 90% but a specificity of only 31%, resulting in 55 false positives out of 73 positive tests. This demonstrates the clinical challenge of balancing what two competing priorities when selecting a screening test cutoff point?
What is sensitivity (detecting the disease) and specificity (avoiding false alarms or unnecessary anxiety/treatment)?
700
In the Australian Melanoma Family Study, researchers minimized measurement error when assessing sun exposure—which is complex and difficult to measure accurately—by using three strategies: collecting both self and parent reports of childhood sun exposure, collecting objective clinical measures of skin phenotype, and doing what with their questionnaire?
What is minimising measurement error by using a questionnaire that has been used elsewhere and has good reliability (or demonstrated good reliability for capturing total lifetime sun exposure)?
700
BONUS WISDOMS TRIVIA
This person was once in a movie with Sean William Scott (pictured below).
Who is Geordie Keane?
700
The AFFIRM trial involving over 400,000 pregnant women used a cluster randomized design with 33 maternity hospitals as clusters, testing an e-learning package for staff and leaflets for women to reduce stillbirth. Why was cluster randomization chosen instead of individual randomization—what type of intervention characteristic made this necessary?
What is the intervention was delivered to and affected groups (healthcare staff/hospitals) rather than individuals (or risk of contamination between individuals within the same hospital)?
700
In the resistance training study for neck/shoulder pain, researchers found a statistically significant difference in muscle strength with p = 0.008. However, if the null hypothesis were actually true in the population and this significant result occurred by chance alone, what type of error would the researchers have made?
What is a Type I error?
700
A doctor investigating HDL cholesterol levels across three age groups (20-29, 40-49, and 60-69) with 12 individuals in each group used one-way ANOVA. If this test shows a significant result, what important limitation exists—what does ANOVA NOT tell you about the differences between groups?
What is ANOVA does not indicate which specific means differ from one another (or how the means differ from each other)?
