Alternatives to some statistical conventions II: Visual exploration of data

The plot below allows risk to be displayed in relation to two variables. The contours on this plot are derived from a non-linear model fit to the data. What we see is that there are regions where slightly below average growth in the first year results in higher mortality than lower growth, even though the general trend is for higher birth weights and higher growth rates to result in lower mortality. This effect would be hard to detect in a conventional analysis of statistical significance.

Source: Barker, D. J. P. (1998). Mothers, Babies, and Health in Later Life. Edinburgh, Churchill

Advertisements

About Peter J. Taylor

Peter Taylor teaches and directs programs on critical thinking, reflective practice, and science-in-society at the University of Massachusetts Boston. He studies the complexity of environmental and health sciences in their social context as well as innovation in teaching, group process, and interdisciplinary collaboration (see bit.ly/pjtaylor). He is especially interested in conversations with others who are, in diverse ways, "troubled by heterogeneity" (bit.ly/tbhblog)
This entry was posted in general and tagged , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s