Download a PDF of this page here.
Essential Concepts
One-Sample Z-Test of Proportions
- Write out the null and alternative hypotheses.
- The null hypothesis, [latex]H_{0}[/latex], is what we assume to be true to begin with. It is often a statement of no change from the previous value or from what is expected (e.g., we expect a coin to be fair).
- The null hypothesis, [latex]H_{0}[/latex], is always given in the form: [latex]p = value[/latex] for population proportions.
- The alternative hypothesis, [latex]H_{A}[/latex], is what we consider to be plausible if the null hypothesis is false. Often, it is a change from the null hypothesis that we would like to test the accuracy of. With a null value of [latex]a[/latex], the alternative hypothesis, [latex]H_{A}[/latex], is written as an inequality:
- [latex]H_{A} : p>a[/latex],
- [latex]H_{A}: p \lt a[/latex], or
- [latex]H_{A}: p\neq a[/latex].
- The null hypothesis, [latex]H_{0}[/latex], is what we assume to be true to begin with. It is often a statement of no change from the previous value or from what is expected (e.g., we expect a coin to be fair).
- Check the conditions for the hypothesis test. For testing a one-sample z-test for proportions, we require:
- Large counts: Check that [latex]np\ge10[/latex] and [latex]n(1-p)\ge10[/latex].
- Random samples/assignment: Check that the sample is a random sample.
- 10% population size: Check that the sample size, [latex]n[/latex], is less than 10% of the population size, [latex]N[/latex]: [latex]n<0.10(N)[/latex]
- If not using technology, calculate a test statistic.
-
- A test statistic measures the distance between the sample statistic and the null hypothesis value in terms of the standard error of the null hypothesis value.
[latex]\text{test statistic }= \frac{\text{sample statistic }-\text{ null hypothesis value}}{\text{standard error of the null hypothesis value}}=\frac{\stackrel{ˆ}{p}-p}{\sqrt{\frac{p(1-p)}{n}}}[/latex]
where [latex]\stackrel{ˆ}{p}[/latex] is the sample statistics and [latex]p[/latex] is the null hypothesis value.
4. Use technology to calculate a P-value.
-
- We define the P-value as the probability of obtaining a test statistic at least as extreme (in the direction of the alternative hypothesis) as the one that is actually seen if the null hypothesis is true.
- The smaller the probability, the less likely it is that the sample occurred by chance alone, the more evidence we have against the null hypothesis.
- Compare the P-value to the significance level, [latex]\alpha[/latex], to make a decision.
- The significance level, [latex]\alpha[/latex], is the cut-off for P-values at which we have enough evidence to reject the null hypothesis. Typically, small significance levels such as 1%, 5%, or 10% are used in hypothesis testing.
Decision Conclusion If P-value [latex]\le\alpha[/latex], there is enough evidence to reject the null hypothesis. At the [latex]\alpha\times[/latex]100% significance level, the data provide convincing evidence in support of the alternative hypothesis. If P-value [latex]\gt\alpha[/latex], there is not enough evidence to reject the null hypothesis. At the [latex]\alpha\times[/latex]100% significance level, the data do not provide convincing evidence in support of the alternative hypothesis. - We never accept the null hypothesis.
- The significance level, [latex]\alpha[/latex], is the cut-off for P-values at which we have enough evidence to reject the null hypothesis. Typically, small significance levels such as 1%, 5%, or 10% are used in hypothesis testing.
5. Write a conclusion in context (e.g., we do/do not have convincing evidence…).
-
- If a hypothesis test results in rejecting the null hypothesis because the P-value is less than the significance level, we say we have statistical significance in favor of the alternative hypothesis.
- If the results are meaningful, we say that the results have practical significance. Having practical significance usually means the results show a significant improvement!
- Sometimes, due to chance, the result of the hypothesis test does not align with reality. If we reject a correct null hypothesis, we are committing a type I error. If we do not reject a null hypothesis that is actually incorrect, we are committing a type II error.
-
Reject the null hypothesis Fail to reject the null hypothesis Null hypothesis is correct Type I error No error Null hypothesis is incorrect No error Type II error
Two-Sample Z-Test of Proportions
- Write out the null and alternative hypotheses.
- Null hypothesis: [latex]H_0: p_1=p_2[/latex] or [latex]H_0: p_1-p_2=0[/latex]
- Alternative hypothesis:
- [latex]H_A: p_1\lt p_2[/latex] or [latex]H_A: p_1-p_2\lt 0[/latex]
- [latex]H_A: p_1>p_2[/latex] or [latex]H_A: p_1-p_2>0[/latex]
- [latex]H_A: p_1\ne p_2[/latex] or [latex]H_A: p_1-p_2\ne0[/latex]
- Check the conditions for the hypothesis test. For testing a one-sample z-test for proportions, we require:
- Large counts: For [latex]\hat{p}_c = \frac{x_1+x_2}{n_1+n_2}[/latex], check that: [latex]n_1\hat{p}_c \ge 10[/latex], [latex]n_2\hat{p}_c \ge 10[/latex], [latex]n_1(1-\hat{p}_c) \ge 10[/latex], and [latex]n_2(1-\hat{p}_c) \ge 10[/latex].
- Random samples/assignment: Check that the two samples are independent and random samples or that they come from randomly assigned groups in an experiment.
- 10%: Check that [latex]n_1<0.10(N_1)[/latex] and [latex]n_2<0.10(N_2)[/latex].
- Calculate a test statistic.
- Calculate a P-value.
- Compare the P-value to the significance level, [latex]\alpha[/latex], to make a decision.
- Write a conclusion in context (e.g., we do/do not have convincing evidence…).
- Typically, the conclusion drawn from a two-tailed confidence interval is usually the same as the conclusion drawn from a two-tailed hypothesis test. If a confidence interval contains the hypothesized parameter, a hypothesis test at the 0.05 level will almost always fail to reject the null hypothesis. If the 95% confidence interval does not contain the hypothesized parameter, a hypothesis test at the 0.05 level will almost always reject the null hypothesis. While this does not always hold for tests of proportions, a confidence interval typically provides more information about reasonable values of the parameter.
Key Equations
pooled sample proportion
[latex]\hat{p}_c = \frac{x_1+x_2}{n_1+n_2}[/latex]
test statistic
[latex]\text{test statistic } = z =\frac{\stackrel{ˆ}{p}-p}{\sqrt{\frac{p(1-p)}{n}}}[/latex]
Glossary
hypothesis testing
the process of forming hypotheses, collecting data, and using the data to draw a conclusion about the hypotheses
outcomes of hypothesis tests
reject the null hypothesis, fail to reject the null hypothesis
practical significance
the results of a hypothesis test are meaningful
P-value
the probability of obtaining a test statistic at least as extreme (in the direction of the alternative hypothesis) as the one that is actually seen if the null hypothesis is true
significance level
the cut-off for P-values at which we have enough evidence to reject the null hypothesis
statistical significance
enough evidence against the null hypothesis to convince us to reject the null hypothesis
type I error
rejecting a correct null hypothesis
type II error
not rejecting an incorrect null hypothesis