{"id":468,"date":"2015-07-16T08:06:54","date_gmt":"2015-07-16T12:06:54","guid":{"rendered":"http:\/\/sites.berry.edu\/vbissonnette\/?page_id=468"},"modified":"2016-06-22T11:40:24","modified_gmt":"2016-06-22T15:40:24","slug":"t-test-for-independent-groups","status":"publish","type":"page","link":"https:\/\/sites.berry.edu\/vbissonnette\/index\/stats-homework\/documentation\/t-test-for-independent-groups\/","title":{"rendered":"T Test for Independent Samples"},"content":{"rendered":"

Example homework problem:<\/h3>\n

Twenty participants were given a list of 20 words to process. The 20 participants were randomly assigned to one of two treatment conditions. Half were instructed to count the number of vowels in each word (shallow processing). Half were instructed to judge whether the object\u00a0 described by each word would be useful if one were stranded on a desert island (deep processing). After a brief distractor task, all subjects were given a surprise free recall task. The number of words correctly recalled was recorded for each subject. Here are the data:<\/p>\n\n\n\n\n
Shallow Processing:<\/td>\n13<\/td>\n12<\/td>\n11<\/td>\n9<\/td>\n11<\/td>\n13<\/td>\n14<\/td>\n14<\/td>\n14<\/td>\n15<\/td>\n<\/tr>\n
Deep Processing:<\/td>\n12<\/td>\n15<\/td>\n14<\/td>\n14<\/td>\n13<\/td>\n12<\/td>\n15<\/td>\n14<\/td>\n16<\/td>\n17<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Did the instructions given to the participants significantly affect their level of recall (alpha = .05)?<\/p>\n

If you would like help with the hand-written solution for this problem, then click here<\/a>.<\/p>\n

\n

Separate Variables Approach<\/h4>\n

Enter these data as two variables\u00a0in Stats Homework’s <\/i>data manager and rename the variables. Your screen should look like this:<\/p>\n

\"ittest9\"<\/a><\/p>\n

Make sure to double-check and save your data. To conduct your analysis, pull down the Analyze<\/b> menu, choose Tests for One or Two\u00a0 Samples<\/b>, and then choose T Test for Independent Samples<\/b>. Here is the user dialog you will work with:<\/p>\n

\"ittest10\"<\/a><\/p>\n

Move both of your variables to the “Test Variables” window, check all the output options, and click Compute<\/strong>.<\/p>\n

Indicator Variable Approach<\/h4>\n

Enter the scores as one variable with 20 observations, and then create an indicator variable with two values to indicate which scores are from which sample.\u00a0 You can use a numeric or non-numeric indicator variable.\u00a0 Your screen should look something like this:<\/p>\n

\"ittest11\"<\/a><\/p>\n

In this case, your user dialog will look like this:<\/p>\n

\"ittest12\"<\/a><\/p>\n

Move your indicator variable to the “Grouping Variable” box, and the variable containing your data to the “Test Variable” box.\u00a0 Click all the options and then click Compute<\/strong>.<\/p>\n

Basic Output<\/h4>\n

\"ittest13\"<\/a><\/p>\n

Descriptive Statistics<\/b>. These basic statistics are described on the page for the explore procedure<\/a>.<\/p>\n

Levene’s Test for Inequality of Variance<\/b>. This is a test of the equal-variance assumption of the t<\/i> test. If this test reveals a significant difference in variance between your treatment groups (i.e., p < .05), then you would be advised to use the statistical result that assumes unequal variance.<\/p>\n

Assume Equal Variance<\/b>. This table lists the results of your t <\/i>test if you assume that the groups have equal variance (i.e., Levene’s test is not significant).<\/p>\n