Calculate 95 percent confidence interval for the median
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 | *(Q) How can I calculate 95%CI for the median with SPSS? * (A) Answer by Marta Garcia-Granero 2002/03/11. *There are two ways. * Dummy data. INPUT PROGRAM. - VECTOR V(1). - LOOP #I = 1 TO 100. - COMPUTE V(1) = uniform(10). - END CASE. - END LOOP. - END FILE. END INPUT PROGRAM. execute. * Solution #1 (only with SPSS 11). TEMPORARY. COMPUTE constant=1. RATIO STATISTICS v1 WITH constant /PRINT = CIN(95) MEDIAN . * Solution #2 (any version of SPSS). DEFINE CIMED95(!POSITIONAL !TOKENS(1)). * Save original file for later. SAVE OUTFILE='\\cimed__.sav'. SORT CASES BY !1 (A). COMPUTE ID=$CASENUM. RANK VARIABLES=!1 (A) /N INTO n /PRINT=NO . * Lower bound position. COMPUTE l=TRUNC(((n+1)/2-(n<=284)*(0.9789*sqrt(n)))-(n>284)*(0.98*sqrt(n))). * Exception handling (two cases). IF n=17 l=5. IF n=67 l=26. * Upper bound position. COMPUTE u=n+1-l. * Getting lower and upper bound values. SELECT IF((id=l) or (id=u)). EXECUTE . * Report. STRING limits (A12). IF ($CASENUM=1) limits = '95% Lower:' . IF ($CASENUM=2) limits = '95% Upper:' . REPORT FORMAT=LIST AUTOMATIC ALIGN(CENTER) /VARIABLES=limits !1 /TITLE 'Approximate 95% CI for median'. * Restore original file. GET FILE='cimed__.sav'. !ENDDEFINE. * Macro call. cimed95 v1. |
Related pages
...