Using frequencies from the earlier Thorndike-Lorge and the later Kučera-Francis frequency counts, a lognormal distribution model is applied to judge shifts in the frequencies of occurrence of trait adjectives from a like-ableness scale. In the time between frequency counts, the frequencies of the adjectives have shifted an average of approximately .68 words per million toward higher frequencies of occurrence. The amount of shift would probably not vitiate the generalizability of results based upon the Thorndike-Lorge count.
References
1.
AndersonN. H.Likeableness ratings of 555 personality-trait words. Journal of Personality and Social Psychology, 1968, 9, 272–279.
2.
BowenJ. H.The lognormal frequency model and adjective trait names. Psychological Reports, 1971, 29, 473–474. (a)
3.
BowenJ. H.The stability of the printed frequencies of occurrence of 420 English conceptual nouns. Psychonomic Science, 1971, 23, 156–158. (b)
4.
CarrollJ. B.On sampling from a lognormal model of word-frequency distribution. In KučeraH.FrancisW. N. (Eds.), Computational analysis of present-day American English. Providence: Brown Univer., 1967. Pp. 406–413.
5.
KučeraH.FrancisW. N.Computational analysis of present-day American English. Providence: Brown Univer., 1967.
6.
ThorndikeE. L.LorgeI.The teacher's word book of 30,000 words. New York: Columbia Univer., 1944.