Letters

Allocating census data to general practice populations

EDITOR,--Glen Scrivener and David C E F Lloyd conclude that the linkage of general practice data and census data is too inaccurate to use,¹ But this conclusion is not warranted by their findings. Whether such linkage is too inaccurate depends on the use in question and is a personal judgment; it cannot be determined by the correlation coefficient. Furthermore, by using correlation the authors fail to display the limits of the agreement and gain maximum benefit from their work.

Correlation measures the degree of linear association and not agreement.² The agreement of the two proportions of people aged over 64 is better shown by the difference between the two proportions plotted against the mean of those proportions. The 95% confidence interval for this can be calculated and gives a useful view of the accuracy of the agreement. In this study it is not surprising . . . [Full text of this article]

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

Relevant Article

Allocating census data to general practice populations: implications for study of prescribing variation at practice level

Glen Scrivener and David C E F Lloyd
BMJ 1995 311: 163-165. [Abstract] [Full Text]