The Details: This exercise involved the use of two datasets that I created. The first dataset contained "fake" data about 10 elementary school students, their ages, sex, and test scores. I used this dataset to show the students basic tools in SPSS and to get them comfortable thinking about how to interpret and visualize descriptive statistics. I purposefully designed the dataset so that the correlations were strong (see dataset1.sav and Figure 1), and reminded students that we do not always get such "good" results when examining legitimate datasets.

The second dataset contained actual data from the 2004 Connecticut Mastery Test (CMT) for sixth graders at the district level. I included a number of variables:

• Each district's Educational Reference Group (ERG), which is an approximation of the wealth of that district
• Each district's average class size at multiple grade levels
• The percentage of students that reached CMT goal range in that district

Finally, I created a detailed handout (see handout.doc and Figure 2) to guide students through SPSS and some basic data exploration. I encouraged students to work in pairs, so that one student could do the actual "pointing and clicking" while the other could follow along with the handout.

The Downloads: dataset1 (.sav), dataset2 (.sav), handout (.doc)

The Goals: From my standpoint, there were a number of pedagogical goals for this exercise. I wanted students:

• To understand what a dataset looks like in spreadsheet format and to know the difference between a case and a variable
• To be aware that software packages like SPSS exist and are freely availabe at Trinity for their use
• To be able to read a scatterplot
• To know why social scientists are concerned about "controlling" for variables, since we controlled for ERG when measuring the effect of class size on test scores

Figure 1:

Figure 2: