Hypothesis testing is the formal logic behind every 'is this difference real?' question in biology. This set trains you to frame null and alternative hypotheses, interpret p-values honestly, recognize Type I and Type II errors, and match common tests (chi-square, t-test) to the data type. Examiners reward precise statistical language — 'fail to reject' rather than 'prove' — and this set drills exactly that.
Q1.
A researcher wants to test whether a new drug lowers blood pressure. Which is the correct null hypothesis (H0)?
Multiple choice · Foundations
Q2.
In a clinical trial, p = 0.08 for the difference between treatment and control groups. Using the conventional significance level of 0.05, what should the researchers conclude?
Multiple choice · Intermediate
Q3.
A Type I error in hypothesis testing is best described as:
Multiple choice · Intermediate
Q4.
A Type II error in hypothesis testing is best described as:
Multiple choice · Intermediate
Q5.
A geneticist counts kernel colors in corn (purple vs yellow) and wants to test whether they fit a 3:1 Mendelian ratio. Which test is appropriate?
Multiple choice · Foundations
Q6.
For a chi-square test with 4 phenotypic categories, what are the degrees of freedom, and why does the value matter?
Multiple choice · Advanced
Q7.
Why must the null hypothesis and significance level be stated BEFORE data are collected?
Multiple choice · Intermediate
Q8.
A study with 50,000 participants finds that people who drink Brand X tea have a 0.2% lower risk of headaches, p < 0.001. Which statement is most scientifically sound?
Multiple choice · Advanced
Q9.
A researcher measures enzyme activity (a continuous variable) in 12 treated and 12 untreated plants and wants to know if the means differ. Which test fits?
Multiple choice · Intermediate
Q10.
After a well-powered experiment, researchers reject the null hypothesis at p = 0.01. Which conclusion is valid?