Say it with pictures - displaying experimental data

Science World, Sept 16, 1994 by Chana Freiman

Your class's science experiments are almost complete. The test tubes are simmering down, the motorized paper airplanes are nearing Canada, and the incubated duck eggs are starting to hatch. Now all you're left with are the mysterious "results"--reams of dizzying data. How can you find out what all those numbers mean?

Simple, say scientists. Put them on display. "Often the most effective way to describe, explore, and summarize a set of numbers--even a very large set--is to look at pictures of those numbers," says Edward R. Tufte, a professor at Yale University.

You can create pictures of numbers, Tufte says, by representing them with points, lines, bars, words, symbols, shading, and color. Just put these "data markers" on maps, graphs, charts, or diagrams, and watch your findings become clear.

Take the example of Dr. John Snow, a British physician who was confronted with a deadly epidemic in 1854. That year, some 500 people in central London died from cholera, a water-borne bacterial disease. No one knew how to stop the epidemic. Snow had an idea: He decided to plot the location of each death on a street map and try to zero in on the source of the problem (see map, p. 17).

[CHART OMITTED]

He plotted a dot for each cholera death and an x for each of the neighborhood's water pumps--possible areas of contamination. To everyone's amazement, Snow found that most of the deaths clustered near a single pump on Broad Street. Apparently, bacteria had contaminated the pump's water; people drinking it fell ill. Solution: Remove the handle of the pump, making it unusable, thereby saving hundreds of lives.

Snow's map provides a quick, clear analysis of data. Line graphs can accomplish the same goal. They're best used when the variable you're studying (your independent variable) has continuous values, such as minutes of time, degrees of temperature, or levels of chemical concentration. Even though you may only mark data points every 10 minutes, say, or every 10 degrees, values do exist between those intervals. By drawing a line to connect the data points, you can easily spot trends and estimate those in-between values.

CLEARING THE AIR

For instance, check out the line graph showing the carbon dioxide (C[O.sub.2]) level in Earth's atmosphere, above. Notice that time, the independent variable, is plotted along the horizontal (x) axis. The vertical (y) axis has the values for the dependent variable, C[O.sub.2] concentration. Scientists plotted a data point for the C[O.sub.2] level every month between 1958 and 1986, and then connected the dots. The finished graph depicts the rise in C[O.sub.2] over time.

C[O.sub.2], a "greenhouse" gas generated by burning fossil fuels such as oil and coal, traps heat in Earth's atmosphere. Some scientists warn that if C[O.sub.2] levels continue to rise, Earth's temperature could too. How might living on a warmer planet affect your life?

SPACE-AGE DATA

The zigzag or curve of a line graph can be an effective, attentiongrabbing tool--if your independent variable is continuous. But what if you're studying a discrete variable--a variable that comes in different kinds (e.g., cars or animal species) with no continuous values in between? For that kind of data, bar graphs are the way to go.

In the example below, scientists counted the different kinds of objects (satellites, leftover rocket parts, and other space junk) orbiting Earth in 1961--when the space program took off--and in 1993. The independent variable (types of orbiting objects) again appears along the x axis; the dependent variable (number) along the y axis. But in this case, instead of plotting data points, researchers drew separate bars from each x "value" to the corresponding value on the y axis.

To drive home the message that space is becoming more crowded--and more dangerous--scientists have turned to other vivid and creative methods of data display. For example, check out the sequence of diagrams that accompanies the bar graph below. The "globes" show how the debris from a single exploded satellite has traveled far beyond the satellite's original orbit, eventually circling Earth. Scientists say that even a single speck of this flying metal could have the impact of a hand grenade if it collided with a spacecraft. Would these pictures make you think twice before blasting off through space-junk orbits?

As you can see, simple visuals can not only make scientific info clear and exciting, but can also help spark solutions to some major problems. With a little creativity, pencils, markers, and paper, you too can turn your data into displays that make you and your viewers think--and perhaps take action.