Abstract
Many research reports display summary data on study subjects. Seemingly simple tables of data present the mean and standard deviation for each variable. However, this method of presentation is sometimes less than helpful and at other times if is plain wrong. Ordinal data should not be summarized using means and standard deviations and even interval data are often best not summarised in this way Medians and percentiles are usually more informative. This article, the first in a series, outlines some pitfalls in describing study subjects and advises how to avoid them.
Original language | English |
---|---|
Pages (from-to) | 154-155 |
Number of pages | 2 |
Journal | The British Journal of Hospital Medicine |
Volume | 59 |
Issue number | 2 |
Publication status | Published - Feb 1998 |