Do linguistically-naive native speakers' scores on a grammaticality judgment test cluster about a single point, or do the scores vary over a large spectrum? To answer this question, a researcher gathers data on the educational level (in years of public education) of the native speakers.  Also, the researcher gathers data on speakers' judgments of 25 sentences. The scores show speakers' average (in percent). 

Perform descriptive statistics on the data at "j:\ling\msdos\otherdata\dsstudy3.txt."

Data file: dsstudy3.sav
Variables: EDUC, JUDG

JUDG    EDUC
12         1
16.5       1
18         1
17         1
46.5       1
45         1
15         1
60         2
(data continue in dsstudy3.txt) 

Answer the following questions:

1. What kind of data are each of the two variables (nominal, ordinal, or interval)?
    
    
2. Provide three figures to indicate the central tendency of the EDUC data.
    
    
3. Provide three figures to indicate the central tendency of the JUDG data.
    
    
4. Provide two figures to indicate the dispersion of the EDUC data.
    
    
5. Provide two figures to indicate the dispersion of the JUDG data.
    
    
6. Do the JUDG data have any problems with kurtosis or skewness? If so, explain with data.
    
    
    
7. Approximately 68% of the scores will fall within one standard deviation of the mean.
   a. State the 68% limits of variation (upper and lower) for these data.
   b. What crucial assumption about the distribution of the data does this calculation require?