Activity #1 (Group or Individual)

Have students do the following activity, individually or in small groups:
Make a list of variables pertaining to your classmates (e.g., height, eye color, and major). Develop your list so that it includes a range of nominal, ordinal, and interval/ratio level variables. Next, define your response categories. Remember, these categories must be both exhaustive and mutually exclusive. Thus, for example, eye color could be categorized according to the following: blue, brown, green, hazel, and other. Survey each person in your class and proceed to compute all appropriate measures of central tendency. Finally, write a brief report about what you observed and either present or circulate your report to the class.

Activity #2 (Group or Individual)

Have students do the following activity, individually or in small groups:
Take the data collected from Activity #1. Begin by entering the data into a new SPSS datasheet. Save your work. Next, use SPSS to compute the various measures of central tendency computed by hand in the previous group exercise. Are these answers the same as those you calculated by hand? They should be. If not, revisit your work from the previous group exercise. Find any problem spots and redo your work as needed.

Activity #3 (Group or Individual)

Have students come up with examples of variables that might be more likely to have skewed data than others (e.g., income and age). Students should come up with 3–5 examples of variables with potentially skewed data. Next, have students draw what they expect the shape of the distribution to look like and explain why. Their drawings should be labeled as positively skewed or negatively skewed. Have students present their variables and drawings to the class.

Activity #4 (Group or Individual)

Have students create one question for each level of measurement (e.g., for nominal level data you could ask, What is your favorite sports team out of the following list the Royals, Yankees, or Astros?). Then have all of the class answer the questions with their personal data. Have them calculate all of the measures of central tendency. Then have the students determine and explain which measure(s) of central tendency would best represent their dataset. Also have them discuss if any measures of central tendency would not be appropriate for their particular level of measurement. Have students bring all of this work back to the class as a whole for discussion.