Research Module

Question 1

What is the definition of epidemiology?

Answer 1

The quantitative study of the distribution, determinants, and control of health problems in human populations.

Question 2

Who is known as the “father of epidemiology” and why?

Answer 2

John Snow—he linked cholera outbreaks to a contaminated water pump on Broad Street, ending the epidemic.

Question 3

What are the three main aims of epidemiology?

Answer 3

(1) Describe disease distribution, (2) Identify etiological factors, (3) Provide data for planning and evaluating services.

Question 4

What is a sample in epidemiology?

Answer 4

A selected group meant to represent the larger population.

Question 5

Define simple random sampling.

Answer 5

Every individual in the population has an equal chance of being selected.

Question 6

What is stratified random sampling?

Answer 6

The population is divided into subgroups (e.g., age/gender), and random samples are drawn from each.

Question 7

What is cluster sampling?

Answer 7

Groups (not individuals) are randomly selected, e.g., tribes or clinics.

Question 8

What is convenience sampling?

Answer 8

Selecting participants who are easy to access, without ensuring representativeness.

Question 9

What is snowball sampling?

Answer 9

Participants recruit others from their network—used for hard-to-reach populations like drug users.

Question 10

Define selection bias.

Answer 10

Bias from differences in how participants are selected into the study.

Question 11

Define information bias.

Answer 11

Systematic errors in measurement or data collection (e.g., recall or interviewer bias).

Question 12

What is confounding bias?

Answer 12

A third variable distorts the true relationship between exposure and outcome.

Question 13

What is meant by “central tendency”?

Answer 13

A measure of the center of a distribution—mean, median, or mode.

Question 14

When is the median preferred over the mean?

Answer 14

When data are skewed or have outliers.

Question 15

What is variance?

Answer 15

The average squared deviation from the mean—indicates data spread.

Question 16

What is standard deviation?

Answer 16

The square root of variance—measures spread of individual data points around the mean.

Question 17

What is the standard error of the mean?

Answer 17

An estimate of how much the sample mean is likely to differ from the population mean.

Question 18

What is kurtosis?

Answer 18

A measure of how “fat” or “heavy” the tails of a distribution are.

Question 19

What is skewness?

Answer 19

A measure of asymmetry in the distribution.

Question 20

What is a null hypothesis (H₀)?

Answer 20

A statement that there is no difference or association—assumed true until disproven.

Question 21

What is an alternative hypothesis (H₁)?

Answer 21

The hypothesis that there is a real difference or association.

Question 22

What is a p-value?

Answer 22

The probability of obtaining the observed result if the null hypothesis is true.

Question 23

What does it mean if p < 0.05?

Answer 23

The result is statistically significant; we reject the null hypothesis.

Question 24

What is a Type I error (α)?

Answer 24

False positive—rejecting the null when it is actually true.

Question 25

What is a Type II error (β)?

Answer 25

False negative—not detecting a true difference.

Question 26

What is statistical power?

Answer 26

The probability of correctly rejecting a false null hypothesis (Power = 1 – β).

Question 27

What determines statistical power?

Answer 27

Sample size, effect size, and significance level (α).

Question 28

What is a confidence interval (CI)?

Answer 28

A range within which the true population value is expected to lie, with a given level of confidence (usually 95%).

Question 29

What is validity in a screening tool?

Answer 29

The extent to which the tool measures what it’s intended to.

Question 30

Define sensitivity and specificity.

Answer 30

Sensitivity = true positives / (true positives + false negatives); Specificity = true negatives / (true negatives + false positives).

Question 31

Define health informatics.

Answer 31

The discipline that stores, retrieves, shares and uses healthcare information, data and knowledge for communication and decision-making.

Question 32

What does the abbreviation PAS stand for in Maltese hospital systems?

Answer 32

Patient Administration System – the core electronic registry of patient episodes.

Question 33

HL7 messages are primarily used for what?

Answer 33

Standards-based electronic exchange of clinical and administrative data between health-information systems.

Question 34

Which 19th-century physician is hailed as a founder of modern epidemiology for tracing a cholera outbreak to a London pump?

Answer 34

John Snow.

Question 35

Give the textbook definition of epidemiology.

Answer 35

The study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to the control of health problems.

Question 36

Differentiate qualitative vs. quantitative research in one phrase each.

Answer 36

Qualitative = explores why & how via non-numerical data; Quantitative = measures how much/many using numerical data and statistics.

Question 37

What is lead-time bias in screening?

Answer 37

Apparent survival prolongation caused by earlier detection rather than true delay in death.

Question 38

Rank these designs from highest to lowest evidence for causality: cross-sectional, cohort, RCT.

Answer 38

RCT > Cohort > Cross-sectional.

Question 39

State one key attribute that makes an RCT the ‘gold standard’.

Answer 39

Random allocation balances known and unknown confounders between groups.

Question 40

When do you choose a paired t-test instead of an unpaired t-test?

Answer 40

When the same subjects provide two related measurements (e.g., before-and-after).

Question 41

Name two situations that call for a non-parametric test.

Answer 41

(1) Data are ordinal or categorical. (2) Continuous data are skewed and cannot be normalised.

Question 42

Which non-parametric test replaces an unpaired t-test?

Answer 42

The Mann–Whitney U (Wilcoxon rank-sum) test.

Question 43

The χ² (chi-square) test assesses the association between which type of variables?

Answer 43

Two categorical variables.

Question 44

Define confidence interval (CI) in one sentence.

Answer 44

A range of values calculated from the sample that is expected, with a given probability (e.g., 95 %), to contain the true population parameter.

Question 45

Formula for a 95 % CI of the mean (large n, known σ).

Answer 45

Mean ± 1.96 × (SE) where SE = σ/√n.

Question 46

Define standard deviation (SD) in one phrase.

Answer 46

The average distance of individual observations from the sample mean.

Question 47

Simple random sampling guarantees what statistical property?

Answer 47

Every individual in the population has an equal probability of selection, minimising selection bias.

Question 48

Instrument mis-calibration produces which class of bias?

Answer 48

Measurement (information) bias.

Question 49

Define the three measures of central tendency.

Answer 49

Mean = arithmetic average; Median = middle value when ordered; Mode = most frequent value.

Question 50

Which designs are prospective by nature?

Answer 50

Randomised controlled trials and most cohort studies.

Question 51

Give two examples of categorical data and two of continuous data.

Answer 51

Categorical: blood group, gender. Continuous: serum glucose (mmol/L), height (cm).

Question 52

What p-value threshold is conventionally considered statistically significant?

Answer 52

p < 0.05.

Question 53

Which graph best displays the distribution of a single continuous variable?

Answer 53

Histogram.

Question 54

Define positive predictive value (PPV).

Answer 54

Probability that a person with a positive test truly has the disease.

Question 55

In an ROC curve, what does the area under the curve (AUC) represent?

Answer 55

Overall test accuracy; the probability a randomly chosen case ranks higher than a control.

Question 56

What is the primary purpose of blinding in an RCT?

Answer 56

To prevent performance and ascertainment bias by keeping group allocation concealed.

Question 57

Name two advantages of electronic health records over paper charts.

Answer 57

Real-time data access & easier longitudinal tracking of outcomes.

Question 58

Categorise BMI ≥ 30 kg/m² (obesity) – is it nominal, ordinal, or interval?

Answer 58

Ordinal categorical (ordered classes).

Question 59

Explain the term intention-to-treat analysis.

Answer 59

Outcomes are analysed according to original group assignment regardless of protocol adherence, preserving randomisation benefits.

Question 60

Which statistical test compares survival curves between two groups?

Answer 60

The log-rank (Mantel–Cox) test.