{smcl} {* 2004-05-07}{...} {hline} help for {hi:vanelteren}{right:Version 1.0 2004-05-09} {hline} {title:van Elteren's test (generalized Wilcoxon-Mann-Whitney ranksum test)} {p 8 17 2} {cmd:vanelteren} [{it:depvar}] [{cmd:if} {it:exp}] [{cmd:in} {it:exp}] [{cmd:,} {cmd:by(}{it:groupvar}{cmd:)} {cmdab:st:rata:(}{it:varname}{cmd:)} ] {p 4 4 2} {cmd:by} {it:...}{cmd::} may be used with {cmd:vanelteren}; see help {help by}. {title:Description} {p 4 4 2} {cmd:vanelteren} performs van Elteren's generalization of the Wilcoxon-Mann-Whitney ranksum test, a modification that allows stratification or blocking on a variable. It can be thought of as a nonparametric analogue of a randomized-blocks analysis of variance (ANOVA), that is, with a treatment factor (which may only have two levels) and a blocking or stratification variable that accounts for some heterogeneity in response. {it:depvar} must be at least ordinal, and {it:groupvar} must be numeric. The stratification (blocking) variable may be either string or numeric. {cmd:vanelteren} calls {cmd:ranksum} for the within-stratum Wilcoxon statistics and weights them before combining them in an overall asymptotic test statistic. {title:Options} {p 4 8 2} {cmd:by(}{it:groupvar}{cmd:)} is not optional. It specifies the name of the grouping variable. This {cmd:by()} is separate from the {cmd:by} {it:...}{cmd::} prefix and may be combined with the prefix. {p 4 8 2} {cmdab:st:rata:(}{it:varname}{cmd:)} is likewise not optional. It specifies the name of the stratification or blocking variable, which may be an independent unit, for example, a subject ID analogous to that specified by {cmd:i(}{it:varname}{cmd:)} in {cmd:xtreg}. {title:Remarks} {p 4 4 2} For more information on van Elteren's test, see E. L. Lehmann (and H. J. M. D'Abrera), {it:Nonparametrics. Statistical Methods Based on Ranks} (San Francisco: Holden-Day, 1975), pp. 132-141. This cites the original{char -}{char -}P. H. van Elteren, On the combination of independent two-sample tests of Wilcoxon. {it:Bulletin of the International Statistical Institute} {bf:37}:351-61, 1960. {title:Examples} {p 4 8 2}{cmd:. vanelteren response, by(treatment) strata(clinic)} {p 4 8 2}{cmd:. vanelteren likertscore, by(intervention) st(packsaday)} {title:Author} {p 4 4 2} J. Coveney E-mail {browse "mailto:jcoveney@bigplanet.com":jcoveney@bigplanet.com} if you observe any problems. {title:Also see} {p 4 13 2} Manual: {hi:[R] signrank} {p 4 13 2} Online: help for {help signrank}