Hypothesis Testing for CFA Level 1: The Steps, Type I and II Errors, and Which Test to Use
Skip the comparison - test yourself on 50 free questions
The first 50 questions of our 180-question CFA Level I paper, free: Ethics, Quantitative Methods and the start of Economics, in the exam's three-option format. The full paper covers all ten 2027 topics.
CFA-EXAM-FGE
Try 50 CFA Level I questions free
Already have an account? Log in to redeem
A hypothesis test asks whether sample evidence is strong enough to reject a null hypothesis. Compute a test statistic, compare it with a critical value or compare the p-value with the significance level α, and reject if it is more extreme. A Type I error rejects a true null (probability α); a Type II error fails to reject a false one. Quant rises to 11-14% of Level I in 2027.Checked against CFA Institute's published pages on 15 November 2026.
In this guide
The six steps
- State the hypotheses: H0 always contains the equals sign; Ha is what you want to show.
- Choose the test statistic and its distribution.
- Set the significance level α.
- State the decision rule: the critical values, or reject if p < α.
- Collect the data and compute the statistic.
- Decide, and say what it means in plain terms.
Type I, Type II and power
| H0 true | H0 false | |
|---|---|---|
| Reject H0 | Type I error (probability α) | Correct (probability = power) |
| Fail to reject H0 | Correct | Type II error (probability β) |
Raising α makes rejection easier: more Type I errors, fewer Type II errors, more power. A larger sample raises power without changing α.
Which test for which question
| Question | Test | Degrees of freedom |
|---|---|---|
| One mean, population variance unknown | t = (X̄ - μ0) / (s / √n) | n - 1 |
| One mean, large sample or known variance | z | - |
| Difference between two means (independent) | t | Depends on the variance assumption |
| Mean difference (paired samples) | t on the differences | n - 1 |
| One variance | Chi-square = (n - 1)s2 / σ02 | n - 1 |
| Two variances | F = s12 / s22 | n1 - 1, n2 - 1 |
| A correlation | t = r√(n - 2) / √(1 - r2) | n - 2 |
Your free CFA Level I sample code is ready
Sit the first 50 questions at the real 90-second pace, about 75 minutes, and find out where you stand before you spend on a full course. The full 180-question paper is $39 once, lifetime, updates included.
CFA-EXAM-FGE
Redeem the code
Already have an account? Log in to redeem
Critical values to know
| Significance | One-tailed z | Two-tailed z |
|---|---|---|
| 10% | 1.282 | 1.645 |
| 5% | 1.645 | 1.96 |
| 1% | 2.326 | 2.576 |
Worked example
Is a fund's mean monthly return different from zero?
36 months, mean 1.2%, standard deviation 4%. H0: μ = 0, Ha: μ ≠ 0, α = 5%.
Standard error = 4 / √36 = 0.667. t = 1.2 / 0.667 = 1.80. The two-tailed 5% critical t with 35 degrees of freedom is about 2.03.
1.80 is inside ±2.03, so we fail to reject H0: the evidence is not strong enough at 5%. "Fail to reject" is not "accept": the test cannot prove the mean is zero.
Five Level I questions
Five questions in the Level I format. Answer before you read each explanation.
Frequently asked questions
What is the difference between a Type I and Type II error?
Type I: rejecting a true null hypothesis, with probability equal to the significance level. Type II: failing to reject a false null hypothesis.
What is the power of a test?
The probability of correctly rejecting a false null hypothesis: 1 minus the probability of a Type II error.
What does a p-value mean?
The smallest significance level at which the null hypothesis can be rejected. If the p-value is below α, reject.
Which test do I use for a variance on CFA Level 1?
Chi-square for one variance against a hypothesized value; F for comparing two variances.
Is hypothesis testing still on CFA Level 1 in 2027?
Yes. It sits in Quantitative Methods, which rises to 11-14% of the exam in 2027.