Describing Data
Figure 1335 –
Some figures may not display clearly when rendered as a PDF or printed.
Once a researcher has completed a study, he or she is faced with some major challenges: to analyze the data in order to publish, to add a line to the CV, to get promoted or tenured, to get more research grants, and to analyze more data.
There are two distinct steps in the process of analyzing data. The first step is to describe the data by using standard methods to determine the average value, the range of data around the average, and other characteristics. The objective of descriptive statistics is to communicate the results without attempting to generalize beyond the sample of individuals to any other group. This is an important first step in any analysis. For a reader to understand the basis for the conclusions of any study, an idea of what the data look like is necessary.
The second step in some, but not all, studies is to infer the likelihood that the observed results can be generalized to other samples of individuals. If we want to show that clam juice is an effective treatment for psoriasis, or that intelligence (IQ) is related to subsequent performance, we are attempting to make a general statement that goes beyond the particular individuals we have studied. The rub is that differences between groups can rarely be attributed simply to the experimental intervention; some people in the clam juice group may get worse, and some people in the placebo group may get better. The goal of inferential statistics is to determine the likelihood that these differences could have occurred by chance as a result of the combined effects of unforeseen variables not under the direct control of the experimenter. It is here the statistical heavy artillery is brought to bear. As a result, most damage to readers of journals is inflicted by inferential statistics. Most of this book is devoted to the methods of statistical inference. However, a good idea of what the data look like is a necessary prerequisite for complex statistical analysis, both for the experimenter and the reader, so let’s start there.5343
Content on this page was last changed on March 19, 2009.
© 2002 BC Decker Inc. Show Disclaimer
| 5343. | Norman GR, Streiner DL. PDQ Statistics . 3rd ed. Hamilton, Ontario: BC Decker Inc.; 2003. |
Next Page: Frequencies and Distributions »