Using Analysis of Variance (ANOVA) on Excel Analysis of variance is one of the statistical methods used to determine if there is any

difference in means. Unlike t-test that we performed in Lab 1, Analysis of variance is used when determining the difference in means of more than two populations. However, this test can only tell you that if there is at least two population means are significantly different. Why not just use the t-test?1 The t-test tells us if the mean difference between two groups is "significant". Why not just do t-tests for all the pairs. Multiple t-tests are not the answer because as the number of means increases, the number of needed pair comparisons grows quickly. For 7 groups there are 21 pairs. In addition, the probability of falsely rejecting at least one of the null hypotheses increases as the number of t tests increases. Hypothesis Unlike t-test, the hypothesis for analysis of is always one-tailed test. Example: We want to know if there is any different in tensile strength of 4 films, LDPE, LLDPE, PP, and PET. Ho: LDPE = LDPE = LDPE = LDPE Ha: not all are equal Where = mean of tensile strength Procedure to use Excel for ANOVA 1. Check to see if you need to load the Data Analysis Add-in - Pull down the Tools menu to see if Data Analysis is there (at bottom) - If it is proceed to step # 2 - If not: pull down Tools menu and choose Add-Ins; put a check by the Analysis ToolPak; hit OK 2. To do a 2-sample t-Test: a. Enter Data into Excel datasheet For example, Manual 3.24 3.44 Auto Method I 3.53 3.74 Auto Method II 3.66 3.89

b. Pull down Tools to Data Analysis c. Select ANOVA: Single Factor

Modified from http://www.physics.csbsju.edu/stats/anova.html

d. In open window, look to see Variable 1 Range window; choose the little red arrow in right of window; after open window collapses, in the main data sheet, use the cursor to drag around all the data. This will enter the cell values into the window; click the red arrow in the small window to open again. For example, Manual Auto Method I Auto Method II 3.24 3.53 3.66 3.44 3.74 3.89 e. Back in the ANOVA: Single Factor to fill out the rest of the information o Grouped by: Columns (as your group is arranged by column) o Put check mark in front of labels in first row o Alpha: 0.05 o Output options: New Workssheet Ply

f. Click ok

You will have output looking like in Figure 1 or similar if you use equal variance. Excel will perform both one-tailed and two-tailed t-test Figure1. Result from unequal variance t-test performed by Excel

The first table shows the descriptive statistics of each group. The second table is ANOVA Table. This table contains information used to calculate Fvalue. The only information you will need from this table to test your hypothesis is Pvalue. In order to reject the null hypothesis at 95% confidence, you need P-value to be less than or equal to 0.05.

Suggested Reading: Ott, R. L.; Longnecker, M. An introduction to statistical methods and data analysis, 1st ed.; Duxbury: California, 2001,