This Article  • References  • Full text PDF  • Reply to article  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citing articles on HighWire  • Contact me when this article is cited  Related Content  • Similar articles in this journal  Social Bookmarking   What's this?

Helen Chmura Kraemer, PhD
Department of Psychiatry and Behavioral Sciences Stanford University School of Medicine Stanford, CA 94305

Arch Gen Psychiatry. 1987;44(2):192-193.

	Since this article does not have an abstract, we have provided the first 150 words of the full text PDF and any section headings.

To the Editor.—

A "Peanuts" cartoon has Charlie Brown exhorting his baseball team to greater efforts, saying "All right. We've heard the report from our statistician... Both our hitting and our fielding averages were down this year... So you all know what we have to do next season." In chorus, the team responds "Get a new statistician!!!" The article by Spitznagel and Helzer¹ proposing a solution to the base rate problem in the k statistic has much the same flavor. The k statistic indicates that the validity and reliability of diagnostic tests is population-specific and tends to be low in extreme-risk populations (low or high base rate or prevalence). What to do about it? Get a new statistic!

It is, of course, a fact that k is population-specific.² However, the situation is even more problematic than has been suggested, for two different clinical populations having similar base rates . . . [Full Text PDF of this Article]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?