8.4 Hypothesis Test for One Proportion

When you read a question, it is essential that you correctly identify the parameter of interest. The parameter determines which model to use. Make sure that you can recognize and distinguish between a question regarding a population mean and a question regarding a population proportion.

The z-test is a statistical test for a population proportion. It can be used when np ≥ 10 and nq ≥ 10.

Definition

z-Test

The formula for the test statistic is:

$Z = \frac{\hat{p} - p_{0}}{\sqrt{(\frac{p_{0} q_{0}}{n})}} .$

where

$n$ is the sample size

$\hat{p} = \frac{x}{n}$ is the sample proportion (sometimes already given as a %) and

$p sub 0$ is the hypothesized population proportion,

$q sub 0 equals 1 – p sub 0.$

Use the phrases in Figure 8-24 to help with setting up the hypotheses.

Hypothesis testing table for proportions with three columns: two-tailed test (H0: p = p0, H1: p ≠ p0, both tails shaded), right-tailed test (H1: p > p0, right tail shaded), and left-tailed test (H1: p < p0, left tail shaded), each with a normal curve. Below are common phrases for claims in the null hypothesis under =, ≤, and ≥ (is equal to, is at most, is at least) and in the alternative hypothesis under ≠, >, and < (is different from, more than, less than, decreased).

Figure 8-24

Note we will not be using the t-distribution with proportions. We will use a standard normal z distribution for testing a proportion since this test uses the normal approximation to the binomial distribution (never use the t-distribution).

If you are doing a left-tailed z-test the critical value will be negative. If you are performing a right-tailed z-test the critical value will be positive. If you were performing a two-tailed z-test then your critical values would be ±critical value. The p-value will always be a positive number between 0 and 1. The most important step in any method you use is setting up your null and alternative hypotheses. The critical values and p-value can be found using a standard normal distribution the same way that we did for the one sample z-test.

Example 1

It has been found that 85.6% of all enrolled college and university students in the United States are undergraduates. A random sample of 500 enrolled college students in a particular state revealed that 420 of them were undergraduates. Is there sufficient evidence to conclude that the proportion differs from the national percentage? Use $α$ = 0.05. Show that all three methods of hypothesis testing yield the same results.

Show solution

At this point you should be more comfortable with the steps of a hypothesis test and not have to number each step, but know what each step means.

Critical Value Method

Step 1: State the hypotheses: The key words in this example, “proportion” and “differs,” give the hypotheses:

H₀: p = 0.856

H₁: p ≠ 0.856 (claim)

Step 2: Compute the test statistic. Before finding the test statistic, find the sample proportion $\hat{p} = \frac{420}{500} = 0.84$ and q₀ = 1 – 0.856 = 0.144.

Next, compute the test statistic:

$z = \frac{\hat{p} - p_{0}}{\sqrt{(\frac{p_{0} q_{0}}{n})}} = \frac{0.84 - 0.856}{\sqrt{(\frac{0.856 \cdot 0.144}{500})}} = - 1.019 .$

Step 3: Draw and label the curve with the critical values. See Figure 8-25.

Use $α$ = 0.05 and technology to compute the critical values $z sub α / 2$ and $z sub 1 minus α / 2$ .

Excel: $z sub α / 2$ =NORM.S.INV(0.025) = –1.96 and $z sub 1 minus α / 2$ =NORM.S.INV(0.975) = 1.96.

TI-Calculator: $z sub α / 2$ = invNorm(0.025,0,1) = –1.96 and $z sub 1 minus α / 2$ = invNorm(0.975,0,1) = 1.96.

Figure 8-25

Step 4: State the decision. Since the test statistic is not in the shaded rejection area, do not reject H₀.

Step 5: State the summary. At the 5% level of significance, there is not enough evidence to conclude that the proportion of undergraduates in college for this state differs from the national average of 85.6%.

P-value Method

The hypotheses and test statistic stay the same.

H₀: p = 0.856

H₁: p ≠ 0.856 (claim)

$Z = \frac{\hat{p} - p_{0}}{\sqrt{(\frac{p_{0} q_{0}}{n})}} = \frac{0.84 - 0.856}{\sqrt{(\frac{0.856 \cdot 0.144}{500})}} = - 1.019$

To find the p-value we need to find the P(Z > |1.019|) the area to the left of z = –1.019 and to the right of z = 1.019. First, find the area below (since the test statistic is negative) z = –1.019 using the normalcdf we get 0.1541. Then, double this area to get the p-value = 0.3082.

Since the p-value > $α$ the decision is to not reject H₀.

Summary: There is not enough evidence to conclude that the proportion of undergraduates in college for this state differs from the national average of 85.6%.

There is a shortcut for this test on the TI Calculators, which will quickly find the test statistic and p-value.

TI-84: Press the [STAT] key, arrow over to the [TESTS] menu, arrow down to the option [5:1-PropZTest] and press the [ENTER] key. Type in the hypothesized proportion (p₀), x, sample size, arrow over to the $\neq$ , <, > sign that is the same in the problem’s alternative hypothesis statement then press the [ENTER] key, arrow down to [Calculate] and press the [ENTER] key.

Three TI-84 screens for a one-proportion z-test: the TESTS menu with 5:1-PropZTest highlighted, the input screen with p0 = .856, x = 420, n = 500 and prop ≠ p0 highlighted, and the results screen showing prop ≠ .856, z = −1.019029734, p = .3081888764, p̂ = .84, n = 500.

The calculator returns the z-test statistic and the p-value. Note: sometimes you are not given the x value but a percentage instead. To find the x to use in the calculator, multiply $\hat{p}$ by the sample size and round off to the nearest integer. The calculator will give you an error message if you put in a decimal for x or n. For example, if $\hat{p}$ = 0.22 and n = 124 then 0.22*124 = 27.28, so use x = 27.

TI-89: Go to the [Apps] Stat/List Editor, then press [2nd] then F6 [Tests], then select 5: 1-PropZ-Test. Type in the hypothesized proportion (p0), x, sample size, arrow over to the $\neq$ , <, > sign that is the same in the problem’s alternative hypothesis statement then press the [ENTER] key to calculate. The calculator returns the z-test statistic and the p-value. Note: sometimes you are not given the x value but a percentage instead. To find the x value to use in the calculator, multiply $\hat{p}$ by the sample size and round off to the nearest integer. The calculator will give you an error message if you put in a decimal for x or n. For example, if $\hat{p}$ = 0.22 and n = 124 then 0.22*124 = 27.28, so use x = 27.

Adapted from Mostly Harmless Statistics by Rachel Webb (Portland State University), hosted on LibreTexts (stats.libretexts.org) and licensed under CC BY-SA 4.0. Changes were made. License: CC-BY-SA-4.0.