01974cam a22003492u 4500001000800000005001700008007001400025008004100039035002000080037001900100040002700119088001400146091000800160100001900168245012500187260006800312300001000380520069300390530008001083533008101163650003201244650003301276650002101309650002901330650002301359650002501382653004401407700002901451710006501480856003701545984004201582518908120181011150944.0he u||024||||080220s1964 xxu ||| b ||| | eng d 9(ericd)ED016251 aED016251bERIC aericdbengcericddMvI aCRP-S-010 amfm1 aTate, Merle W.10aTables for comparing related-sample percentages and for the median testh[microform] /cMerle W. Tate and Sara M. Brown. a[Washington, D.C.] :bDistributed by ERIC Clearinghouse,c1964. a40 p. aTo study the exact distributional differences between small, related-sample percentages and the accuracy of large-sample tests, some six hundred sampling distributions were constructed by randomization. Exact probability figures in the .204-.005 range were tabulated. The table of significance values extends to samples yielding matrices varying in size from tree columns and 12 rows to six columns and five rows, including all combinations of row totals. The two-sample problem was treated as a special case of the sign test. The use and reliability of the q-test is discussed. The use and accuracy of the median test in the two-way classification of quantitative data is described. (ps) aMay also be available online. Address as at 14/8/18: uhttps://eric.ed.gov/ aMicrofiche.b[Washington D.C.]:cERIC Clearinghouseemicrofiches : positive.07aExperimental Groups.2ericd17aResearch Methodology.2ericd17aSampling.2ericd17aStatistical Data.2ericd07aStatistics.2ericd07aTables (Data)2ericd0 aMEDIAN TESTaRELATED SAMPLE PERCENTAGES1 aBrown, Sara M.,eauthor.2 aPennsylvania Univ., Philadelphia. Government Studies Center.41uhttps://eric.ed.gov/?id=ED016251 aANLcmc 2253 ED016251d77000000028798