Treating ordinal scales as interval scales: an attempt to resolve the controversy.

Ever since the Harvard psychologist S. S. Stevens (1946) advanced his ideas concerning the connections between measurement scales and statistical analysis, there has been a continuing controversy regarding the use of traditional descriptive and inferential statistics for ordinal-level variables. The exchange in Nursing Research between Armstrong and this writer (Armstrong, 1981; Knapp, 1984, Armstrong, 1984) is illustrative of the competing positions. In this commentary, an attempt will be made to sort out the dimensions of the controversy summarize some recent contributions that help to clarify the issues, and suggest a possible solution to the problem. The reader who is interested in the earlier history of conflicting views about scales and statistics is referred to the excellent and balanced review by Gardner (1975).