SPSS I: Menu Basics Practice Exercises Target Software & Version: SPSS V Last Updated on January 17, 2007 Created by Jennifer Ortman

Transcription

1 SPSS I: Menu Basics Practice Exercises Target Software & Version: SPSS V Last Updated on January 17, 2007 Created by Jennifer Ortman PRACTICE EXERCISES Exercise A Obtain descriptive statistics (mean, median, mode, range, standard deviation, and distribution statistics skewness and kurtosis) and a histogram with normal curve for the variable, highest year of school completed (educ). Note: You don t need to create a frequency table. a. What is the total number of cases? What is the number of valid and missing cases? b. What are the mean, median, mode, range, standard deviation, and distribution statistics of this variable? c. Is the distribution skewed or normally distributed? Exercise B Use either Visual Bander or Recode to create two new variables. a. god Create a dummy variable of (0) Don t believe in God (1) Believe in God. b. numcong Create a categorical variable with 3 categories: (1) min (2) (3) 1001 max. god Please look at this card and tell me which statement comes closest to expressing what you believe about God. 1. I don't believe in God. 2. I don't know whether there is a God and I don't believe there is any way to find out. 3. I don't believe in a personal God, but I do believe in a Higher Power of some kind. 4. I find myself believing in God some of the time, but not at others. 5. While I have doubts, I feel that I do believe in God. 6. I know God really exists and I have no doubts about it. numcong About how many members does this congregation have? Exercise C Create a bar chart for respondent s highest degree (degree) by sex (sex). Make a title of the bar chart, change the font to Times New Roman, display data values (counts and percentages), and bring legend to the lower right of corner of the chart.

2 SPSS I: Menu Basics Practice Exercises Page 2 of 8 PRACTICE EXERCISE ANSWERS Exercise A 1. Run frequency. (Analyze > Descriptive Statistics > Frequencies) In order not to have a frequency table, you can delete the tick from Display frequency tables. 2. After selecting the variable, educ, click on Statistics and Charts in order to select statistics and charts that you want.

3 SPSS I: Menu Basics Practice Exercises Page 3 of 8 3. You will see statistics and histogram. Interpretation The total number of cases is 2817 (= 2808 (valid case) + 9 (missing case)). There are two measures of distribution skewness and kurtosis. When the distribution is normal, both skewness and kurtosis are equal to 0. Skewness describes the degree and direction of asymmetry. When the distribution is skewed to the left, the skewness statistics is negative (like this example skewness = -.134). Kurtosis describes whether the distribution is narrow and peaked or too wide and flat. When the distribution is narrow and peaked, the kurtosis statistics is positive (like this example kurtosis =.781). While you rarely get value of zero for skewness and kurtosis, you need to determine when you have to reject the hypothesis of a normal distribution. You can use the confidence interval: when the 95% confidence interval includes the value zero, then you cannot reject the hypothesis of a normal distribution. Thus, you need to calculate a 95% confidence interval (= skewness (or kurtosis) statistics ± 1.96 * (standard error of skewness (or kurtosis)). 95 % CI for Skewness = -.134± 1.96 *.046 (-0.224, 0.224) 95% CI for kurtosis =.781± 1.96 *.092 (0.601, 0.961) Based on the results, the distribution is not skewed but there is kurtosis (i.e., distribution is peaked and narrow).

4 SPSS I: Menu Basics Practice Exercises Page 4 of 8 Exercise B-a 1. Run frequency for god. (Analyze > Descriptive Statistics > Frequencies) 2. Transform the variable, god, using Recode. (Transform > Recode > Into Different Variables) See page for detailed steps. Exercise B-b 1. Run frequency for numcong. (Analyze > Descriptive Statistics > Frequencies) 2. Transform the variable, num,cong, using Visual Bander. (Transform > Visual Bander) 3. First type the new variable name in the Banded variable box. Next, You can create cutpoints manually. Type the cutpoints into the Value boxes. In this example, type 500 in the first box, and click enter. After you finished typing, click on Make Labels, you will see that SPSS has generated a value label for each category.

5 SPSS I: Menu Basics Practice Exercises Page 5 of 8 You will see the frequency table like below. As an alternative, you can use Recode for creating a new variable for numcong, instead of using Visual bander. As you can see, however, Visual bander preserves the missing value coding of the original variable for the banded variable. Exercise C Create a bar chart. (Graphs > Bar) 1. In this example, use Clustered since you want to define cluster by sex.

6 SPSS I: Menu Basics Practice Exercises Page 6 of 8 2. Select degree for Category Axis, sex for Define Clusters by. Click Titles and type title(s). 3. You will see the bar chart like this. 4. Double-click on the bar chart to bring up chart editor. (See page 28 for notes on Editing a Chart). 5. After you add the data value, double-click on any of the data values. In Properties, select the Data Value Labels tab. Select Percent and click on the arrow, and then click Apply and Close.

7 SPSS I: Menu Basics Practice Exercises Page 7 of You notice that the chart does not display all values due to space limitations. So, change the chart size by double-clicking on the chart. In the Chart Size tab under the Properties dialog box, set Height to be around After you change the font (See page 28 for notes on Editing a Chart), single-click on the legend area and you will see the purple border like below. Resize the legend box and bring it to the lower right hand corner of the chart. 8. At the end, your chart should look like this.

8 SPSS I: Menu Basics Practice Exercises Page 8 of 8