Professional Documents
Culture Documents
Data
Presentation
Part II
Copyright (c) 2013 by The McGraw-Hill Companies. This material is solely for educational
use by licensed users of LearningStats. It may not be copied or resold for profit.
“Lord, Lord, how this
world is given to lying.”
600
480 464
500 430
408 419
400 366
300
200
100
0
2000 2001 2002 2003 2004 2005
500
480
464
450
430
419
408
400
366
350
2000 2001 2002 2003 2004 2005
Surgical incidents
Patient Falls
Nosocomial Infections
12 14 16 18 20 22 24 26
Source http://aolpf5.marketwatch.com/
Always Zero Origin?
U.S. Open Golf Winning Score, 1947-2002 U.S. Open Golf Winning Score, 1947-2002
350 305
300 300
250 295
200 290
Men Men
Women Women
150 285
100 280
50 275
0 270
47 51 955 959 963 967 971 975 979 983 987 991 995 999 47 51 55 59 63 67 71 75 79 83 87 91 95 99
19 19 1 1 1 1 1 1 1 1 1 1 1 1 19 19 19 19 19 19 19 19 19 19 19 19 19 19
Type of Scale
1,000
Billions of Dollars .
10
1990 1992 1994 1996 1998 2000 2002 2004
Unclear Definitions
Missing or unclear units of measurement
leave the reader in limbo. Is the variable
measured in dollars? Yen? Percent?
Company Growth
Undefined X and Y
600
500
variables can render a chart
400
useless. Are we showing
300 total revenue? Total profit?
200 Profit after taxes?
100
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Unlabeled Points
U.S. Hospital Closures, 1987-1997
100 88
76
For maximum clarity, 80 69
56 57
each bar should be 60 50
42
labeled with its 40
37 37 38
numerical value. 16
20
1987
1990
1991
1995
1996
1997
1988
1989
1992
1993
1994
Bar value labels may be omitted if the scale and axis tick
marks are clear (especially if a grid is used) or if you are
worried about distracting attention from the data.
Uneven Axis Ticks
140,000
120,000
80,000 Sable
Mountaineer
60,000
Cougar
40,000 Villager
20,000
0
1996 1997 2000 2001
Dramatic Title
36,000 35,735
34,791
34,000 33,625
32,000
2002 2003 2004
2005
62
Percent of U.S. Population
60
58
56
54
52
50
1950 1960 1970 1980 1990
130
Me n
110 Women
y = -0.2133x + 105.67 Linea r
(Wome n)
Linea r
100 (Me n)
1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000
Ye a r
Not specific
enough!
Authority Figures
Emergency Arrivals, 1980-1998
Advertisements 80000
portraying scientists,
20000
10000
0
doctors, or business 19
8 0 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19
leaders examining
scientific-looking charts.
Because the public
respects science, such
displays impart credibility
to self-serving
commercial claims.
3-D and Rotated Graphs
1995
1990
1980
0 1970
1960
1950
1936
800 721
700 642
600 566
501
500 398
400 336 311
300
200
100
0
1936 1950 1960 1970 1980 1990 1995
0.600
0.500
0.400
0.300
0.200
0.100
0.000
Distracting Pictures
Do For-Profit Hospitals Need a
Shot in the Arm?
1340
1330
Num ber of Hospitals
1320
1310
1300
1290
1280
1270
1260
1993 1994 1995 1996 1997 1998 1999
Data are from Modern Healthcare, Vol. 30, No. 3, January 17, 2000, p. 20,
made into a weird chart to demonstrate this problem.
Area Trick
40000
38529
37137
35000
1997 1998 1999 2000
Data are from Modern Healthcare, Vol. 30, No. 45, October 30, 2000, p. 16,
made into a deceptive chart here to illustrate the point.
Radar Chart
Example 3. Average Student Evaluation of Five Professors
On a Scale of 1 to 5 (5 = very good, 1 = very bad)
Jones Smith Green Brown
Sometimes called a
Grades Fairly 4.7 3.2 2.2 4.5 “spider chart.” Distorts
Organized 3.3 3.4 2.9 4.4 data by showing areas
Knows Subject 4.0 3.6 2.6 4.1
Explains Well 3.8 3.2 3.6 4.5 instead of true
Challenging 3.8 3.3 3.2 4.2 proportions.
Available 4.0 3.9 2.3 4.3
Enthusiastic 3.5 3.1 3.1 4.2
Grades Fairly
5.0 Ratings of Four Professors
4.0 Jones
Enthusiastic 3.0 Organized
Smith
2.0
Green
1.0
Brown
0.0
Available Knows Subject