Statistics are a central tool of researchers. Fortunately, there are many computer statistical pro- grams available designed to help us "crunch the numbers." One widely available program is Microsoft's Excel. Excel is a spreadsheet program that can be used by researchers to manage and analyze data sets. The following exercise is designed to introduce you to some of the basic descriptive and inferential statistical resources within Excel that may be useful in research methods courses. Suppose that we conducted an experiment comparing a treatment group with a control group. We could enter the data into Excel as an array of numbers. A B C 1 Control Treatment 2 45 64 3 65 76 4 75 83 5 56 69 6 46 66 7 8 9 10 Now that the data are entered, you can perform a number of operations. Suppose that you wanted to compute the mean of the control group. You can do this in several different ways. One way would be to enter the equation. For example, in the A7 cell, you can enter = (A2 + A3 + A4 + A5 + A6)/5 This will return the mean of these five numbers. You can do most of your statistical analyses of your data by entering the appropriate equa- the tions (sometimes using multiple steps-e.g., compute the mean first, then use it to compute standard deviation, then use that to compute your standard error). However, Excel also has several statistical functions built into it. Some of the most common ones include the following (and can be found under the insert menu: Function): Descriptive statistics: = sum (array) = average (array) Mean = median (array) Median = StDevA (array) StandardDeviation (sample) = StDevP (array) StandardDeviation (population) (Notice: Array in this case would be something like A2:A6 to mean A2, A3, A4, A5, and A6.) Part I. Research Methods Exercises 63 Excel also includes some inferential statistical procedures. t test: = ttest(array1,array2,tails, type) For the first equation, you must arrange the data into two arrays for this analysis. (tails = 1 or 2 for one-tailed or two-tailed) (type = 1, 2, or 3 for paired test, independent samples with equal variances, independent samples with unequal variance) (Notice: The value that is returned is the p value, not the computed t score) correlation: = correl(array1,array2) You must arrange the data into two arrays of the same size for this analysis. Exercise Using Excel Enter the data set from the above example into an Excel spreadsheet. 1. Using Excel, compute the mean and sample standard deviation for each group. 2. Using Excel, compute the correlation between two groups. 3. Using Excel, compute a paired samples t test. Based on your results, using an alpha level of 0.05 and a two-tailed test, should you reject the null hypothesis?