Moving beyond Kučera and Francis: A critical evaluation of current word frequency norms and the introduction of a new and improved word frequency measure for American English

it has been well established that word frequency is a very important variable in cognitive processing. Highfrequency words are perceived and produced more quickly and more efficiently than low-frequency words (e.g., Balota & Chumbley, 1984; Jescheniak & Levelt, 1994; Monsell, Doyle, & Haggard, 1989; Rayner & Duffy, 1986). At the same time, high-frequency words are easier to recall but more difficult to recognize in episodic memory tasks (e.g., Glanzer & Bowles, 1976; Yonelinas, 2002). To investigate the word frequency effect, psychologists need estimates of how often words occur in a language. Howes and Solomon (1951), for instance, made use of Thorndike and Lorge’s (1944; hereafter, TL) list of words as counted in books. Subsequently, Ku era and Francis’s (1967; hereafter, KF) frequency norms became the measure of preference and formed the basis of over 40 years of psycholinguistic and memory research in the U.S. The latter may be surprising, because the KF list was based on a corpus of 1.014 million words only, whereas TL was based on a corpus of 18 million words. The reasons why KF became more popular may have been that the texts were more recent (from 1961 vs. the 1920s and 1930s) and were entirely based on adult reading material, whereas TL also contained children’s books. Differences in availability may have played a role as well, in addition to a snowball effect (once KF was used in a number of key articles, it became the measure of choice for the group of researchers working on that topic).