[EMAIL PROTECTED] (F. Goldhammer) wrote in message news:<[EMAIL PROTECTED]>... > Hi all, > > I want to calculate the reliability of the following quotient: > quotient=MEAN(item7, item8 ... item18 )/MEAN(item1, ... item6). The 18 > items form a time series. > Analogous to the testhalf-reliability I have devided the 18 items in > odd an even items. I have calculated the quotient separately with the > odd and the even items. At last I have interpreted the correlation > between the quotient_odd and the quotient_even (+ correction: > Spearman-Brown-Formula ) as the reliability of the quotient. > > Is that correct or is there a better solution? > > Many Thanks in advance > > Frank Goldhammer
my first reaction is to calculate cronbach's alpha coefficient. this is the mean of all possible split-half correlations. BTW, the quotient of means is somewhat intuitive, but not what the alpha calculation is expecting. there you use the actual 18 item scores for each observational unit; no summary statistic is calculated. so i take it you are expecting between units variation and wanting relatively less within units variation. the alpha coefficient falls into the broad area of internal consistency measures. it sounds like what you are seeking. hope this helps. JJD . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
