- The proportion of “no preference” responses for the remaining fifteen pairs was a fraction under 50%.
- I am advised that the design of this test poses serious problems for a meaningful statistical analysis; the figures are therefore given without any attempt at further interpretation.
- It will be noted than an effort was made to get the respondents to evaluate items is terms of the test material rather than in comparison with established lexical items. As for the unfamiliarity of the items on our list, we can at least say that none of them are listed in the NCD.
- We are not considering nasty as an obvious antonym of nice; the point is of course arguable.
- X2 = 21.85, with 4 degrees of freedom; x2.001 = 18.46. For the Friedman test, cf. Sidney Siegel, Nonparametric Statistics for the Behavioral Sciences (New York, 1956), pp. 166–73.
WORD
Volume 20, 1964 - Issue sup1: Monograph No.5