SBIR-STTR Award

The aim of the proposed research is to determine whether an individual has acquired an adult reading vocabulary and to develop effectiveness measures of growth towards adult competence in vocabulary. Adult standards are defined by previously identified adult corpora.In Phase I, word frequencies will be calculated and checked against published counts for these corpora. Other corpora, defined by materials written for students from early grades on, will be analyzed. Counts will be conditioned on prose difficulty, because variability in book difficulty defined by publishers' grade level designations is exceedingly large. Phase I word frequency counts will be twice the size of the largest published school corpora.Vocabulary measures are developed in Phase II. Stratified samples from the frequency bands of the adult corpus will be drawn to test for lexical meanings. Latent trait models and TASA's calibration designs will be used to obtain sample-free estimates of vocabulary item difficulty Phase II research should show that several factors account for difficulty of vocabulary test items, including the frequency of the word in the adult corpus, the relative frequency of the word in corpora that form the development scale, and the relative frequency of the word's lexical meaning.National Institute of Child and Human Development (NICHD)

The aim of this research is to determine whether an individual has acquired an adult reading vocabulary and to develop an effectiveness measure of growth toward adult vocabulary literacy. The adult corpus is defined in terms of the difficulty of informational materials written for a general adult audience. Grade-level corpora are defined in terms of the difficulties of series textbooks designated for each grade.All these corpora were identified during Phase I, when the largest (over 16,000,000 tokens) systematic word frequency count ever done was conducted. Vocabulary measures will be developed in Phase II by drawing stratified samples from the frequency bands of the adult corpus and testing for lexical meanings. Latent trait models and calibration designs will be used to obtain estimates of vocabulary item difficulty.Phase II research should show that several factors account for difficulty of vocabulary test items, including the frequency of the word selected in the adult corpus, the relative frequency of the word in corpora that form the developmental scale, and the relative frequency of the word's lexical meaning. Semantic analyses conducted for test development will form the basis for semantic word counts.Awardee's statement of the potential commercial applications of the research:The Word Frequency Volume should serve as the standard reference for years to come. The effectiveness measure of vocabulary is the first of its kind. At the very least, Touchstone Applied Science Associates, Inc., will incorporate it into a battery with the Degrees of Reading Power Tests.National Institute of Child Health and Human Development (NICHD)