The instructions given here and in all the data analysis will work with the SAS Learning Edition as well as with the SAS Enterprise Guide. The Summary Statistics task provides summary statistics including basic summary statistics. percentile summary statistics, and more advanced summary statistics including confidence intervals, statistics, coefficient of variation, and sums of squares. It also provides graphical displays including histograms and box and- whisker plots. The One-Way Frequencies task can be used to generate frequency tables as well as binomial and chi-square tests.

To select these tasks, click:

Describe>Summary Statistics

Describes-One- Way Frequencies

We give detailed steps for running frequencies on Familiarity with the Internet and plotting the histogram.

1. Select DESCRIBE on the SAS Learning Edition menu bar.

2. Click SUMMARY STATISTICS.

3. Move the variable “FAMILIARITY” to the Analysis variables role.

4. Click BASIC.

5. Select MEAN, STANDARD DEVIATION, VARIANCE, and RANGE.

6. Click PLOTS.

7. Click HISTOGRAM.

8. Click RUN

The major cross-tabulation task is called TABLE ANALYSIS. This task will display the cross classification table and provide cell counts, row and column percentages, the chi-square test for significance, and all the measures of strength of the association.

To select this task, click:

Describe>Table Analysis

We give detailed steps for running the cross-tabulation of sex and usage of the Internet given in Table 15.3,and calculating the chi-square, contingency coefficient, and Cramer’s V.

1. Select DESCRlBE from the SAS Learning Edition menu bar.

2. Click TABLE ANALYSIS.

3. Move the variables “IUSAGEGROUP” and “SEX” to the Table variables ·role.

4. Click TABLES.

5. Move “SEX” and then “IUSAGEGROUP”.

6. Click CELl.; StATISTICS.

7. Select COLUMN PERCENTAGES and CELL FREQUENCIES.

8. Click ASSOCIATION under TABLE STATISTICS.

9. Click CHI_SQUARE TESTS under TESTS OF ASSOCIATION.

10. Click RUN.

The major task for conducting parametric tests in SAS Learning Edition is T TEST. This task can be used to conduct tests on one sample or independent or paired samples. To select this task using SAS Learning Edition, click:

Analyze>ANOVA> Test

We give the detailed steps for running a one-sample test on the data of Table 15.1. We wanted to test the hypothesis that the mean familiarity ratings exceed 4.0. The null hypothesis is that the mean familiarity rating is less than or equal to 4.0.

1. Select ANALYZE from the SAS Learning Edition menu bar.

2. Click ANOVA and then TEST.

3. Click ONE SAMPLE.

4. Click TASK ROLES.

5. Move FAMILIARITY to the Analysis variables role.

6. Click ANALYSIS.

7. Enter “4” into the Null Hypothesis field.

8. Click RUN .

We give the detailed steps for running a two-independent-samples test on the data of Table 15.1. The null hypothesis is that the Internet usage for males and females is the same.

1. Select ANALYZE from the SAS Learning Edition menu bar.

2. Click ANOYA and then TEST.

3. Click TWO SAMPLE.

4. Click TASK ROLES.

5. Move “IUSAGE” to the Analysis variables role.

6. Move “SEX” to the Group by role.

7. Click RUN.

We give the detailed steps for running a paired samples test on the data of Table 15.1.The null hypothesis is that there is no difference in the attitude toward the Internet and attitude toward technology.

1. Select ANALYZE from the SAS Learning Edition menu bar.

2. Click ANOYA and then T TEST.

3. Click PAIRED.

4. Click TASK ROLES.

5. Move “IAlTITUDE” and ”TATTITUDE” to the Paired variables role.

6. Click RUN.

The non-parametric tests discussed in this chapter can be conducted as follows.

Non-parametric one sample test (Kolmogorov-Smirnov one sample test)

1. Select DESCRIBE from the SAS Learning Edition menu bar.

2. Click DISTRIBUTION ANALYSIS.

3. Click TASK ROLES.

4. Move “!USAGE” to the ANALYSIS variable role.

5. Click TABLES.

6. Check on TESTS FOR NORMALITY.

7. Click RUN .

Note that you check TEST FOR NORMALITY if you want a one-way test of distribution, If you want a non-parametric test for location, perform the same steps above except select TEST FOR LOCATION in Step 6. In addition to the (parametric) r-test, you will get the Sign Test and the Wilcoxon Signed Rank Test (also known as the Mann Whitney test).

Non-parametric two independent samples (Wilcoxon test which is also known as the Mann-Whitney test)

1. Select ANALYZE from the SAS Learning Edition menu bar.

2. Click ANOYA and then NONPARAMETRIC ONE-WAY ANOYA.

3. Click TASK ROLES.

4. Move “!USAGE” to the Dependent variable role.

5. Move “SEX” to the Independent variable role.

6. Click ANALYSIS.

7. Click WILCOXON.

8. Click RUN.

Non-parametric two paired samples (Wilcoxon matched-pairs signed-ranks test)

Create a “difference score” that is calculated as DIFF=IATTITUDE- TATTITUDE and then use one sample method to analyze DIFF:

1. Select DESCRIBE from the SAS Learning Edition menu bar.

2. Click DISTRIBUTION ANALYSIS.

3. Click TASK ROLES.

4. Move “DIFF’ te the ANALYSIS variable role.

5. Click TABLES.

6. Check on TESTS FOR NORMALITY

7. Click RUN.

**Project Research**

Basic Data Analysis

In the department store project, basic data analysis formed the foundation for conducting subsequent multivariate analysis Data analysis began by obtaining d frequency distribution and descriptive statistics for each variable. In addition to identifying possible problems with the data this information provided d good feel tor the data and insights into how specific variables should be treated in subsequent analysis. for example, should some variables be treated as categorical, and, If so how many categories should there be? Several two and three-variable cross-tabulations were also conducted to identify associations in the data. The effects of variables with two categories on the metric dependent variables of interest were examined by means of tests and other hypothesis-testing procedures.

**Project Activities**

1. Run a frequency distribution tor each familiarity variable and the overall familiarity)- score with all accompanying descriptive statistics.

2. Recede the overall familiarity score 3~ follows: 32 or less = 1; 33 to 37 = 2; 38 to 43 = 3; 4-4 to 60 = 4.

Run cross-tab, of the receded overall familiarity score with demographic variables as recoded in.

3. Test the null hypothesis that the average overall familiarity score is less than or equal to 30.

4. Do a parametric and a corresponding non-parametric test to determine whether the married and not married (recoded marital status) differ in their overall familiarity score.

5. Do a parametric and a corresponding non-parametric test to determine whether the respondents differ in their familiarity with Neiman Marcus and JCPenney.