Question 1

Your head of clinical practice has followed your advice and now wants you to measure effect sizes. You report a Pearson's r of 0.50 for the impact of increased consumption of oily fish supplements on the symptoms of dementia. Your head of clinical practice wants to know if this is bad, as she remembers that a p-value of 0.30 is not good. What do you tell her?

Accepted Answer

Effect size and p-values are not the same. A Pearson's r of 0.50 indicates a large effect size, regardless of the p-value. Therefore, the results suggest that increased consumption of oily fish supplements may have a positive impact on the symptoms of dementia, and your head of clinical practice should encourage their use in patients.

Question 2

In our previous example, the doctor had already calculated the probability of patient life expectancy based on national cancer data. In a Bayesian approach, what sort of probability is this?

Accepted Answer

The probability calculated based on national cancer data is a prior probability, which is an initial probability used as a starting point in Bayesian analysis.

Question 3

A researcher presented a recent study, which showed a statistical significance between increased consumption of sugar and increased tests scores in young children. How can she explain to the chief medical officer that advertising the importance of increasing sugar in children's diets should not be broadcast in the media?

Accepted Answer

A significant result only indicates that there is a relationship between two variables, in this case, sugar consumption and test scores. It does not imply that this relationship is important or causative. In fact, promoting increased sugar consumption in children's diets can have negative health consequences, and further research should be done before making any public health recommendations.

Question 4

You are the head of a large medical group. You have decided to adopt a Bayesian approach to data analysis and modelling. When you announce this new policy, your staff are unhappy and unconvinced, as they are used to a NHST approach. You stress that the Bayesian approach has several key advantages, including which of the following.

Accepted Answer

The Bayesian approach allows for evaluating the likelihood of the null hypothesis being true, which is not possible with NHST.

Question 5

You lead a product-testing unit for a large pharmaceutical company. Your team has conducted forty trials of a new antibiotic but you are not sure if the results are conclusive enough to urge the company to start producing the new drug. A new data analyst has joined your team suggesting that meta-analysis might be a good idea, do you agree?

Accepted Answer

Meta-analysis can be useful in this case because the forty trials were identical and tested the same research question, allowing for the calculation of an average effect size for the new drug. This can provide a more precise estimate of the drug's effectiveness and help determine whether it is worth producing.

Question 6

A midwife conducted a Bayesian analysis of stork numbers and declining birth rates. She calculated a Bayes factor of 1. Should she use stork numbers to predict birth rates?

Accepted Answer

A Bayes factor of 1 suggests that there is no evidence to support the hypothesis that stork numbers are a predictor of declining birth rates. Therefore, it is not advisable to use stork numbers to predict birth rates. Other factors may be at play, and more data is needed to make a stronger case for any correlation between stork numbers and birth rates.

Question 7

A researcher was assessing patient healing time when using a new form of surgical dressing called MediDressX. He had a sample size of 48 and a p-value of 0.20. Does the researcher recommend that hospitals stop using this product?

Accepted Answer

While a p-value of 0.20 may suggest that there is no significant difference in healing time between patients using MediDressX and those using other dressings, the small sample size of 48 means that the results are less reliable and subject to change with a larger sample size. Therefore, the researcher cannot recommend hospitals stop using MediDressX based on this study alone.

Question 8

A trainee data analyst for a large hospital, which has higher than average patient mortality rates, has just completed an examination of key factors that may be influencing these high mortality rates. However, he finds only one statistically significant factor, which he includes in his report but deliberately, omits the other six non-significant findings. What is the term for what the data analyst has done?

Accepted Answer

The data analyst has engaged in p-hacking, which involves selectively choosing which statistical tests to perform and which results to report to reach a desired conclusion. By only reporting the statistically significant finding and omitting the non-significant ones, the data analyst has distorted the results and potentially misled others. HARKing (Hypothesizing After the Results are Known) refers to the practice of formulating a hypothesis after analyzing data, which can also introduce bias. Meta-analysis is a statistical technique for combining data from multiple studies to arrive at a summary estimate. Bayesian analysis is a statistical approach that incorporates prior knowledge and assumptions to update probability estimates based on new data.

Question 9

A researcher was interested in assessing gender and response to pain, so conducted a t-test. The mean for males was 66.25 and the mean for females was 78.24, with both groups having a standard deviation of 7. What is the effect size using Cohen's d?

Accepted Answer

Cohen's d = (mean difference) / (pooled standard deviation)
mean difference = 78.24 - 66.25 = 11.99
pooled standard deviation = sqrt(((n1-1) * s1^2 + (n2-1) * s2^2) / (n1 + n2 - 2))
= sqrt(((49-1) * 7^2 + (49-1) * 7^2) / (49 + 49 - 2))
= sqrt(2*48*49/96) = 7.00
Cohen's d = 11.99 / 7 = 1.712. 
Therefore, the answer is D.

Question 10

You have joined the data modelling team for a health policy unit. Your boss has decided that from now on the team will adopt a Bayesian approach. However, not all staff understand what this is; your boss asks you to present a training session. How would you explain s Bayesian appaorch in your session introduction?

Accepted Answer

Bayesian approach is an approach that allows you to update the likelihood of your statistical model as more data is collected. It involves the use of probability theory to model uncertainty. Unlike classical statistics, Bayesian approach allows for the incorporation of prior knowledge and beliefs into the analysis, making it a flexible and powerful tool for decision making in various fields, including health policy.

Question 11

Which of the following is not a factor in researchers' over-use of p-values and NHST in health research?

Accepted Answer

Statistical software itself does not encourage the over-use of p-values, as it simply provides a tool for calculation. The other options - time constraints, pressure to find significant results, and incentivized career bonuses - are all factors that contribute to the over-reliance on p-values and NHST in health research.

Question 12

A doctor is concerned about the life expectancy rates of cancer patients in his clinic. There were forty patients in the clinic all of whom were in remission from cancer. Theoretically, all the patients had an equal probability of living for at least ten years post-treatment. However, the doctor had data that suggested that younger patients were more likely to live longer than older patients. This data was from national cancer data. However, the doctor had used the same and newly developed approach to treatment for all his patients and therefore wanted to determine the probability that all patients would live for at least ten years post-treatment, irrespective of their age. What formula could he use to determine this probability?

Accepted Answer

Bayes' theorem can be used to compute the probability of an event (in this case, living for at least ten years post-treatment) given prior knowledge of conditions related to the event (in this case, age). By using Bayes' theorem, the doctor can incorporate the national cancer data on age and life expectancy to determine the probability of all patients living for at least ten years post-treatment, while also considering the individual characteristics of each patient in the clinic.

Question 13

Your health research lecturer has devoted the past ten weeks to teaching you the Bayesian approach and is now asking that you offer a critique of it. What key criticism could you raise?

Accepted Answer

The key criticism of Bayesian approach is that it relies on prior probability, which can be overly subjective and therefore open to a researcher's degrees of freedom. This means that the results obtained from a Bayesian analysis can be influenced by the researcher's personal beliefs and opinions, leading to biased outcomes. In contrast, frequentist approach relies on a null hypothesis, which is more objective and less prone to researcher bias.

Question 14

Your head of clinical practice has just read a book on criticisms of the NHST and worries that all clinical data analysis is now flawed. How might you reassure her?

Accepted Answer

NHST does have its flaws, but incorporating an examination of effect sizes can help address these flaws and increase the reliability of our research outputs. This approach is recognized as important by leading statistical organizations such as the American Statistical Association. It is important to continue to critically evaluate different statistical approaches and choose the best one for each research question, but NHST can still be a useful tool when used appropriately.

Question 15

A person involved in health promotion wanted to examine the likelihood that health promotion messages on social media would be more effective than in print media. She conducted one study where the probability of making a Type I error was 0.05 and a Type II error was 0.2. Does her research have empirical probability?

Accepted Answer

Empirical probability refers to the probability that an event will occur based on the results of repeated trials of an experiment. In this case, the researcher conducted one study, so it does not have empirical probability. However, in order for an experiment to have empirical probability, it does not necessarily require repeated identical experiments with the same Type I and Type II error probabilities. The probability values of Type I and Type II errors are specified by the researcher, and the experiment can be repeated with different values for these probabilities.

Question 16

A researcher was interested in patient use of pop-up Flu vaccine clinics across thirty sites nationally. Different researchers collected and analysed data across each of the sites but the resultant thirty reports showed differing p-values, some sites found a statistical significance between opening hours of clinics and patient usage and others did not. Which of the following would is useful for her to review?

Accepted Answer

Confidence intervals provide information about the precision and reliability of the estimated effect size, which can help understand the variability in p-values across different sites.

Quiz 3: The Phoenix of Statistics