Example: An Abridged Volunteerism Survey

To help illustrate these procedures, assume that you conduct a scaled-down version of your study on volunteerism. You construct a new questionnaire which asks just one question related to helping behavior. The questionnaire also contains an item that assesses participant sex, and another that determines each participant’s class in college (e.g., freshman, sophomore). See the questionnaire below:

Please indicate the extent to which you agree or
 disagree with the
following statement:

1.  I feel a personal responsibility to help needy
 people in my
    community. (please check your response below)

     (5) _____ Agree Strongly
     (4) _____ Agree Somewhat
     (3) _____ Neither Agree nor Disagree
     (2) _____ Disagree Somewhat
     (1) _____ Disagree Strongly

2.  Your sex (please check one):

     (F) _____ Female
     (M) _____ Male

3.  Your classification as a college student:

     (1) _____ Freshman
     (2) _____ Sophomore
     (3) _____ Junior
     (4) _____ Senior
     (5) _____ Other

Notice that this instrument is printed so that data entry will be relatively simple. With each variable, the value that will be entered appears to the left of the corresponding participant response. For example, with question 1 the value “5” appears to the left of “Agree Strongly.” This means that the number “5” will be entered for any participant checking that response. For participants checking “Disagree Strongly,” a “1” will be entered. Similarly, notice that, for question 2, the letter “F” appears to the left of “Female,” so an “F” will be entered for participants checking this response.

The following format is used when entering the data:

You administer the questionnaire to 14 students. The following is the entire SAS program used to analyze the data, including the raw data:

 1        DATA D1;
 2           INPUT   #1   @1   RESNEEDY    1.
 3                        @3   SEX        $1.
 4                        @5   CLASS       1.   ;
 5        DATALINES;
 6        5 F 1
 7        4 M 1
 8        5 F 1
 9          F 1
10        4 F 1
11        4 F 2
12        1 F 2
13        4 F 2
14        1 F 3
15        5 M
16        4 F 4
17        4 M 4
18        3 F
19        4 F 5
20        ;
21        RUN;
22
23        PROC MEANS   DATA=D1;
24           VAR RESNEEDY CLASS;
25        RUN;
26        PROC FREQ   DATA=D1;
27           TABLES SEX CLASS RESNEEDY;
28        RUN;
29        PROC PRINT   DATA=D1;
30           VAR RESNEEDY SEX CLASS;
31        RUN;

The data obtained from the first participant appears on line 6 of the preceding program. This participant has a value of “5” on the RESNEEDY variable (indicating that she checked “Agree Strongly”), has a value of “F” on the SEX variable (indicating that she is a female), and has a value of “1” on the CLASS variable (indicating that she is a freshman).

Notice that there are some missing data in this dataset. On line 9 in the program, you can see that this participant indicated that she was a female freshman, but did not answer question 1. That is why the corresponding space in column 1 is left blank. In addition, there appears to be missing data for the CLASS variable on lines 15 and 18. Unfortunately, missing data are common in questionnaire research.