CABUGATAN, HANIF S. MENDEZ, MARIANO RYAN J. What is a Z TEST?
AZ-testis anystatistical testsfor which
thedistributionof thetest statisticunder thenull hypothesis can be approximated by anormal distribution. Because of thecentral limit theorem, many test statistics are approximately normally distributedforlargesamples. For each significance level, theZ-test has a single critical value (for example, 1.96 for 5% two tailed) which makes it more convenient than theStudent'st- testwhichhasseparatecriticalvaluesforeachsample size A z-test is a statistical test used to determine whether two population means are different when the variances are known and the sample size is large. The test statistic is assumed to have a normal distribution, and nuisance parameters such as standard deviation should be known for an accurate z-test to be performed. The Hypothesis Tests The z-test is a hypothesis test in which the z-statistic follows a normal distribution. The z-test is best used for greater than 30 samples because, under the central limit theorem, as the number of samples gets larger, the samples are considered to be approximately normally distributed. When conducting a z-test, the null and alternative hypotheses, alpha and z-score should be stated. Next, the test statistic should be calculated, and the results and conclusion stated. BUT BEFORE THAT, Here is how you run a one sample z- test State the Null Hypothesis. This is one of the common stumbling blocksin order to make sense of your sample and have the one sample z test give you the right information you must make sure youve written the null hypothesis and alternate hypothesis correctly. For example, you might be asked to test the hypothesis that the mean weight gain of pregnant women was more than 30 pounds. Your null hypothesis would be: H0: = 30 and your alternate hypothesis would be H , sub>1: > 30. All you do is put in the values you are given into the formula. Your question should give you the sample mean (xx ), the standard deviation (), and the number of items in the sample (n). Your hypothesized mean (in other words, the mean you are testing the hypothesis for, or your null hypothesis) is 0. PROBLEM Suppose it is up to you to determine if a certain state receives
significantly more public school funding (per student) than
the USA average. You know that the USA mean public school yearly funding is $6300 per student per year, with a standard deviation of $400. Next, suppose you collect a sample (n = 100) and determine that the sample mean for New Jersey (per student per year) is $8801. Use the z-test and the correct Ho and Ha to run a hypothesis test to determine if New Jersey receives significantly more funding for public school education (per student per year). HYPOTHESIS Problem: Do the new jersery receives significantly more funding for public school education (per student per year). Hypothesis: The mean per student per year funding in New Jersey is significantly greater than the average per student per year funding over the entire USA. Ho: mean per student per year funding for New Jersey = mean per student per year funding for the USA Ha: mean per student per year funding for New Jersey > mean per student per year funding for the USA NOTICE1: The Ha in this example is ONE-TAILED because we are interested in seeing if New Jersey is significantly GREATER than the population mean. In a two-tailed test, the Ha contains a NOT EQUAL and the test will see if there is a significant difference (greater or smaller). NOTICE2: The Ho is the null hypothesis and so always contains the equal sign as it is the case for which there is no significant difference between the two groups. Statistics Now, calculate the test statistic. In this example, we are using the z-test and are doing this by hand. However, there are many applications that run such tests. This Site has several examples under the Stats Apps link. z = (sample mean population mean) / [population standard deviation/sqrt(n)] z = (8801 6300) / [400/sqrt(100)] z = 2501 / [400/10] z = 2501 / [40] z = 62.5 So, the z-test result, also called the test statistic is 62.5. Formula to be used: The alpha value is the percentage chance that you will reject the null (choose to go with your Ha research hypothesis as you conclusion) when in fact the Ho really true (and your research Ha should not be selected).
This is also called a Type I error (choosing Ha when Ho
is actually correct). The smaller the alpha, the smaller the percentage of error, BUT the smaller the rejection regions and more difficult to reject Ho. Most research uses alpha at .05, which creates only a 5% chance of Type I error. However, in cases such as medical research, the alpha is set much smaller.
In our case, we will use alpha = .05
This is ONE-TAILED test, therefore the rejection region is any z-test value greater than the critical z value for a one-tailed test with alpha = .05 The critical value for one-tailed z-test at alpha = .05 is 1.645. Therefore, the rejection region is any value GREATER than 1.645. CONCLUSION
Our z-test result is 62.5.
This is very large! 62.5 is MUCH LARGER than 1.645 and so the result of the z test is INSIDE the rejection region. (The rejection region is all portions of the curve to the right of 1.645. Why to the right? Because this is a one-tailed test for which we are looking at whether the sample is GREATER than the population). Note: The p value in this case would be the probability of getting a result of 62.5 randomly given that the rejection region starts at 1.645.
The probability of getting 62.5 is nearly zero and so
the p value = 0. The actual calculation of the p value is mathematically complicated and is often done using a software like Excel, SPSS, SAS, R, etc. This z-test has a significant result. Conclusion: The funding for New Jersey public schools (per student per year) is significantly GREATER than the average funding per student per year for the USA. THANK YOU FOR NOT LISTENING