Transcript Slide 1
Making Graphs The Basics … Graphical Displays Should: • induce the viewer to think about the substance rather than about the methodology, graphic design, the technology of the graphic production, or something else • avoid distorting what the data have to say • present many numbers in a small space Continued... The Basics … Graphical Displays Should: (2) • make large data sets coherent • encourage the eye to compare different pieces of data • serve a clear purpose • be closely integrated with the statistical and verbal descriptions of a data set. Lie Factor • Lie Factor = size of effect shown in graphic size of effect in data • Greater than 1.05% or less than .95% indicates substantial distortion, far beyond minor inaccuracies in plotting. NYT: Fuil economy “graph” The eye perceives area, not height Maps: just bad graphs Maps: just bad graphs Maps: just bad graphs Maps: just bad graphs Maps: just bad graphs Maps: just bad graphs Maps: just bad graphs Chartjunk • What is it? Anything that doesn’t NEED to be included in the chart. • To clean-up chartjunk, watch your data-ink ratio. “Data-ink is the non-erasable core of a graphic, the non-redundant ink arranged in response to a variation in the numbers represented.” Data-ink ratio = data-ink total ink used to print the graphic Some cool historical graphs 1. Planetary orbits, 10th or 11th century William Playfair (1759-1823) Inventor of: • Line graph • Bar graph • Pie chart Trade balance of England Imports and exports of Scotland Playfair: area of countries (circles), population (left line seg.) and tax revenue (right line seg.). U.S. age pyramids, 1874 Minard's Napoleon's March to Moscow Tufte principles: • • • • • Show Data Focus on Content instead of graphic production Avoid Distorting what Data has to say Make Large Data Sets Coherent Encourage Eye to Compare Different Pieces of Data • Reveal Data at several Levels of Detail • Closely integrate Statistical and Verbal Descriptions • Line Graph – x-axis requires quantitative variable – Variables have contiguous values – familiar/conventional ordering among ordinals • Bar Graph – comparison of relative point values • Scatter Plot – convey overall impression of relationship between two variables • Pie Chart – Emphasizing differences in proportion among a few numbers Bar charts • Best for comparing different things during the same time period • Neither the bars nor the axis should be interrupted • Axis should usually include zero (some exceptions) • Avoid 3-D effects, can be misleading Line graphs • Best for showing change over time • Can indicate trends • Use a different color and symbol for each line • Avoid too many lines • When to use log scale Labeling: Title 80 Labeling: lines 70 White Height/width should be about 3:4 (same as oldfashioned TV Percent with children 60 50 Black 40 30 20 10 0 1850 1870 1890 1910 1930 Census year 1950 1970 1990 Percent of the Labor Force Employed in Agriculture, United States, 1800-2000 80 70 60 Percent 50 40 30 20 10 0 1800 1820 1840 1860 1880 1900 Year 1920 1940 1960 1980 2000 Figure 1: Percent of elders in intergenerational families Argentina 70 Brazil Chile 60 Colombia Costa Rica Percent 50 Ecuador 40 Kenya Mexico 30 20 Philippines Too many lines! Romania Rwanda 10 Vietnam South Africa 0 1970 1975 1980 1985 1990 1995 2000 Uganda Venezuela Married Female Labor Force Participation in Latin America (age 18 to 65) 50 45 40 Brazil Percent in Labor Force 35 30 Colombia 25 Venezuela 20 15 Chile 10 Mexico Costa Rica Ecuador 5 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Married Female Labor Force Participation: Latin America and U.S. (age 18 to 65) 70 60 Percent in Labor Force 50 40 United States 30 20 Latin America 10 0 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 Married Female Labor Force Participation: Latin America and U.S. (age 18 to 65) 70 United States 60 Percent in Labor Force 50 Brazil 40 Compare Latin America to U.S. 40 years ago Colombia 30 Venezuela 20 Ecuador Chile Costa Rica 10 0 1920 Mexico 1930 1940 1950 1960 1970 1980 1990 2000 2010 Married Female Labor Force Participation: Mexican-born Women, 1970-2000 70 60 Mexican-born Women in United States Percent in Labor Force 50 40 30 Women in Mexico 20 10 0 1970 1975 1980 1985 1990 1995 2000 Males 1963 1973 1984 2000 Females Persons age 16 to 65. United United United United United 1962 1968 1975 1982 1990 States 1960 States 1970 States 1980 States 1990 States 2000 France France France France France South Africa 1996 South Africa 2001 Kenya 1989 Kenya 1999 Vietnam 1989 Vietnam 1999 China 1982 Venezuela 1971 Venezuela 1981 Venezuela 1990 Mexico 1970 Mexico 1990 Mexico 2000 Ecuador 1962 Ecuador 1974 Ecuador 1982 Ecuador 1990 Ecuador 2001 Rica Rica Rica Rica 1964 1973 1985 1993 Colombia Colombia Colombia Colombia Costa Costa Costa Costa 1960 1970 1982 1992 2002 Chile Chile Chile Chile Chile Brazil 1960 Brazil 1970 Brazil 1980 Brazil 1991 Brazil 2000 Percent of Working-Age Population Working-Age Population in the Labor Force, by Sex 100 90 80 70 60 50 40 30 20 10 0 Persons with Completed Secondary Education: National Populations Versus Migrants to the United States 100 90 80 70 Percent 60 50 40 30 20 10 0 Brazil Chile Costa Rica Ecuador In home country, ca. 2000 Mexico Vietnam Migrants to U.S. 1995-2000 Kenya South Africa Population Residing with an Elderly Person 30 20 15 10 5 Brazil Colombia Mexico Kenya Elderly persons (age 65+) S Africa China Vietnam France Non-elderly residing with an elderly person 2000 1990 1980 1970 1960 1990 1982 1975 1968 1962 1999 1989 1982 2001 1996 1999 1989 2000 1990 1970 1993 1985 1973 2000 1991 1980 1970 0 1960 Percent of total population 25 United States coreside more than average Percent deviation in intergenerational coresidence of each occupational group from nonfarm average: Younger generation 60.0 Low status Mid status High status 40.0 20.0 coreside less than average 0.0 -20.0 -40.0 -60.0 1850 1860 1870 1880 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 IPUMS Graph from “A Century of Women in Science and Engineering,” History Day project by Abby Norling- Ruggles, age 12 Percent Female; Scientists and Engineers 40 35 30 Percent Female Scientists 25 20 15 Engineers 10 5 0 1900 1910 1920 1930 1940 1950 1960 Year 1970 1980 1990 2000 2005