What is internal validity?
How confidently a relationship can be established within a study (did X cause Y?).
What is external validity?
The ability to generalise findings beyond the study to a wider population.
How can a representative sample be achieved?
By taking a random sample where every individual has an equal chance of being chosen.
Why is random sampling important?
It avoids bias and makes results more generalisable to the population.
What is the difference between unblinded and double-blind trials?
Unblinded: patients/researchers know who gets placebo; double-blind: neither know until the end.
Why does increasing sample size improve reliability?
It reduces random ‘noise,’ improves accuracy of estimates, and makes effects easier to detect.
Why is n = 1 meaningless?
A single observation cannot represent population variation or allow statistical inference.
What is pseudoreplication?
Lack of statistical independence — multiple measures from the same unit counted as separate samples.
Give an example of pseudoreplication.
Measuring 11 mouse brains 3 times each → n = 11 (not 33).
What is the sampling distribution?
The distribution of sample means obtained from repeated random samples of the same population.
What does the central limit theorem state?
The sampling distribution becomes approximately normal if sample size is large (n > 50), even if raw data aren’t.
What is the standard error of the mean (SEM)?
The standard deviation of the sampling distribution; measures precision of the sample mean estimate.
How do you calculate SEM?
SEM = s / √n (where s = sample SD, n = sample size).
How does sample size affect SEM?
As n increases, SEM decreases (estimates become more precise).
What does a 95% confidence interval (CI) mean?
It defines a range in which the true population mean likely lies; 95% of samples will include the true mean.
How do you calculate 95% CI for small samples?
95% CI = mean ± (t × SEM) using the t-distribution.
How do you calculate 95% CI for large samples?
95% CI = mean ± (2 × SEM) using the normal approximation.
Example: Given mean = 27.01, SD = 5.13, n = 12, what is SEM and 95% CI?
SEM = 5.13 / √12 = 1.48; 95% CI = 27.01 ± (2.201 × 1.48) = 23.75 – 30.27.
