Review Exercises

1
Review Exercises

• Pagano and Gauvreau (2nd Edition)
• Chapter 2 Problems
• 2, 9, 10 and 14

2
Review Exercise 2.5. 2. How do ordinal data
differ from nominal data?

• The categories have an order.

3
Review Exercise 2.5. 9. The table below
categorized 10,614,000 office visits to
cardiovascular disease specialists in the US by
duration of visit. A duration of 0 minutes
implies that the patient did not have
face-to-face contact with the specialist.
Duration (min) Visits (1000s)
0 390
1-5 227
6-10 1023
11-15 3390
16-30 4431
31-60 968
61 185
Total 10,614

• The statement is mad that office visits to
  cardiovascular disease specialists are most often
  between 16 and 30 minutes long.
• Do you agree? (see Excel file
  Chapter2-problems-pagano.xls

4
Review Exercise 2.5. 9. Evaluate Rate of visits
Duration (min) Width Visits (1000s) Rate of Visit (per min)
0 NA 390 NA
1-5 5 227
6-10 5 1023
11-15 5 3390
16-30 15 4431
31-60 30 968
61 ? 185
Total 10,614

• Are office visits most often between 16 and 30
  minutes long?
• No- based on rates, 11-15 min has highest rate
  of visits. (see Excel)
• Use RATES when interval widths are not the same
  for classes.

5
Review Exercise 2.5. 10. Construct a Bar Chart
(ejs10b540p03.sas)

• DATA d
• INPUT Year cases
• LABEL cases"Number of Cases"
• CARDS
• 1983 122
• etc
• RUN
• PROC GCHART
• VBAR year / DISCRETE SUMVARcases
• TITLE1 "Figures 1. Number of Pediatric AIDS
  cases reported in the US by Year"
• FOOTNOTE1 "Source prg"
• RUN

6
Review Exercise 2.5. 14. The relative
frequencies of blood lead concentrations for two
groups of workers in Canada- one examined in 1979
and the other in 1987- are given below. Construct
a Percent Freq Polygon. (SEE ejs10b540p04.sas)
Blood Lead 1979 1987lt20 11.5 37.820-29 12.1 14
.7etc
Determine Endpoints (where a class ends, and
Midpoints (middle of class interval. These are
needed for plots.

• DATA d
• INPUT endpoint midpoint yr1979 yr1987
• LABEL endpoint"Blood Lead (ug/dl)ENDPOINT"
• midpoint"Blood Lead (ug/dl)MIDPOINT"
• yr1979"Percent"
• CARDS
• 20 10 11.5 37.8
• etc
• RUN
• SYMBOL INTERPOLJOIN
• PROC GPLOT
• PLOT yr1979midpoint yr1987midpoint /OVERLAY
• TITLE1 "Figure 1. Initial Plot of Percent with
  Blood Lead for Canadian Workers by Year"
• FOOTNOTE1 "Source prg"
• RUN

7
Review Exercise 2.5. 14. Construct a Percent
Freq Polygon (ejs10b540p04.sas)

• DATA d1
• INPUT endpoint midpoint yr1979 yr1987
• CARDS
• 0 0 0 0
• 20 10 11.5 37.8
• etc
• 95 85 9.4 0.4
• 100 100 0 0
• RUN
• SYMBOL INTERPOLJOIN
• PROC GPLOT
• PLOT yr1979midpoint yr1987midpoint /OVERLAY
• TITLE1 "Figure 2. Percent with Blood Lead for
  Canadian Workers by Year"
• FOOTNOTE1 "Source prg"

8
Review Exercise 2.5. 14. Construct a Percent
Freq Polygon (ejs10b540p04.sas) Express as rates
(per 10 years)

• DATA d2
• SET d1
• p1979yr1979
• p1987yr1987
• IF endpoint20 THEN DO
• p1979yr1979/2
• p1987yr1987/2
• END
• RUN

9
Review Exercise 2.5. 14. Construct a Percent
Freq Polygon (ejs10b540p04.sas) Express as rates
(per 10 years)

• DATA d2
• SET d1
• p1979yr1979
• p1987yr1987
• IF endpoint20 THEN DO
• p1979yr1979/2
• p1987yr1987/2
• END
• RUN

10
Review Exercise 2.5. 14. Construct a Cum
Percent Distribution (ejs10b540p04.sas)

• Create cumulative percents
• DATA d2
• SET d1
• RETAIN cyr1979 cyr1987
• IF _N_1 THEN DO
• cyr19790
• cyr19870
• END
• cyr1979cyr1979yr1979
• cyr1987cyr1987yr1987
• LABEL cyr1979"Cum Percent 1979"
• cyr1987"Cum Percent 1987"
• RUN
• SYMBOL INTERPOLJOIN
• LEGEND1 labelnone value(h2 fontswiss '1979'
  '1987')
• POSITION(bottom right inside)
  modeprotect cborderblack