Subjective frequency estimates for 2,938 monosyllabic words

Subjective frequency estimates for large sample of monosyllabic English wor ds were collected from 574 young adults (undergraduate students) and from a separate group of 1,590 adults of varying ages and educational backgrounds . Estimates from the latter group were collected via the internet. In addit ion, 90 healthy older adults provided estimates for a random sample of 480 of these words. All groups rated words with respect to the estimated freque ncy of encounters of each word on a 7-point scale, ranging from never encou ntered to encountered several times a day. The young and older groups also rated each word with respect to the frequency of encounters in different pe rceptual domains (e.g., reading, hearing, writing, or speaking). The result s of regression analyses indicated that objective log frequency and meaning fulness accounted for most of the variance in subjective frequency estimate s, whereas neighborhood size accounted for the least amount of variance in the ratings. The predictive power of log frequency and meaningfulness were dependent on the level of subjective frequency estimates. Meaningfulness wa s a better predictor of subjective frequency for uncommon words, whereas lo g frequency was a better predictor of subjective frequency for common words . Our discussion focuses on the utility of subjective frequency estimates c ompared with other estimates of familiarity. The raw subjective frequency d ata for all words are available at http://www.artsci.wustl.edu/similar to d balota/labpub.html.