Making Sense of Your Data

By the end of this lesson, you will be able to look at a set of data and describe its important features, patterns, mistakes, and limits.

Imagine your class measures the high temperature outside your school every day for two weeks in September in Texas. You end up with fourteen numbers. Just staring at a long list of numbers does not tell you much, so scientists use descriptive statistics, which are simple calculations that summarize a data set. The mean, or average, tells you the typical temperature. The range, which is the highest value minus the lowest value, tells you how spread out your numbers are. If your mean is 93 degrees and your range is only 6 degrees, you know the weather was hot and fairly steady.

Next, look for patterns. Maybe the temperature climbed a little each day, then dropped sharply the day a cold front blew through. A pattern is any repeating or trending shape in the data, and patterns are what let you make predictions about what might happen next.

Then hunt for sources of error, which are things that make a measurement less accurate. Suppose one day a student read the thermometer while it sat in direct sunlight, and that reading was 104 degrees while every other day was near 93. That unusual value should make you suspicious, not excited. Maybe the thermometer was misread, or maybe the timing was different. Honest scientists report possible errors instead of hiding them.

Finally, recognize limitations, meaning what your data cannot tell you. Fourteen days of temperatures from one schoolyard cannot tell you the climate of all of Texas, and it cannot tell you what next September will be like. More data, collected more carefully over a longer time, would let you say much more.

Try it: If your fourteen temperatures included one reading of 40 degrees in September, what error might explain it?