This paper proposes a simplified and formatted method for systematically assigning ranks within and between sample observations drawn from several populations: These populations may be measurements on as low as the ordinal scale and need not be continuous or even numeric. The method is formulated to intrinsically and structurally directly break any possibilities between observations both within and between sample observations and directly assign observations tied in one value their mean ranks without any additional calculations. The proposed method is illustrated with some sample data and shown to be relatively easier to use in practical applications than other existing methods that can possibly be used for the same purpose. KEYWORDS: Rank-order, Ties, observations, Ordinal Scale, Intrinsically Adjusted
展开▼
