CORRELATIONS BETWEEN O’CONNOR’S GENERAL SCALE SCORE
Compiled by Michael Shand, 8/26/93
Anderson, J.D. (1974). SATs and English vocabulary compared. Statistical Bulletin #1974-12. Boston: Human Engineering Laboratories. Examined SAT Verbal and General Scale Scores for 91 males and females aged 16-25; found correlation of .76.
Bowker, R. (1975). Prediction of SAT verbal scores from general scale scores. Statistical Bulletin #1975-30. Boston: Human Engineering Laboratories. Found correlation of .85 between SAT-Verbal score and General Scale score for 100 students aged 16-18. Standard error for SAT score predicted by GSS was 52.
Bowker, R. (1976). Vocabulary as a prediction of your SAT-verbal score. Human Engineering Laboratory, Bulletin Number 123. Shows high positive correlation between JOCRF vocabulary score and SAT-verbal. Argues importance of studying vocabulary in high school in order to achieve college success.
Bowker, R. (1977). Comparison of the vocabulary knowledge of high and low verbal-SAT students. Tech. Rep. #854. Boston: Human Engineering Laboratories. Examined word knowledge patterns of 20 high-scorers on SAT-Verbal and 28 low-scorers in an attempt to determine which words are known to the high scorers but not to the low scorers. Notes that there are some words which are known to almost all in the high SAT group, but to few in the low SAT group. “These are the words which are doing the best job of separating the two groups…. These words should certainly be known by anyone hoping to score over 600 on the SATs” (3). “Other words are known to some of the high group but none of the low group” and hence might be suitable for distinguishing within the high group. “Laboratory theory would hold that these words should not be studied by a student who wishes to raise his score from 400 to 600 since they would beyond his level of difficulty” (3). “What this study suggests, then, is the possibility of arranging words, not by their theoretical General Scale value, but also by their equivalent SAT-Verbal value…. Students…would have a better idea of which words to study in order to raise [their] score…. Development of a list of these SAT equivalents could be a valuable tool in helping students raise their SAT-Verbal scores” (4). Contains table of 150 words on Worksample 95, form AD showing number of students in high SAT group and number of students in low SAT group who got each word correct.
Bowker, R. (1977). Scholastic aptitude tests, vocabulary and aptitudes: A preliminary study. Tech. Rep. #861. Boston: Human Engineering Laboratories. Found correlations between .81 and .85 between Human Engineering Laboratory’s vocabulary tests and the SAT-Verbal, and a .47 correlation between vocabulary and SAT-Math for 100 Laboratory examinees and two groups (N=87 and N=76) of high school seniors. All samples were above national averages, perhaps limiting generalizability of results. “The strength of the relationship between vocabulary scores and SAT-Math scores supports the Laboratory’s contention that vocabulary is important in all school subjects” (7).
Bowker, R. (1981). English vocabulary manual. New York: Johnson O’Connor Research Foundation. “English vocabulary level has been shown to be strongly related to educational success. In addition, it is related to the level of occupation attained. It is highly correlated with measures of reading ability and intelligence” (p.1). Correlations with other standard vocabulary measures: +.64 with Nelson-Denny Vocabulary; +.76 with SAT-Verbal; +.85 with Educational Records Bureau Comprehensive Testing Program, Vocabulary Subtest (level 5). Contains list of average vocabulary scores, by percentile, for a large number of occupations. “It seems clear that most intelligence tests are primarily measures of verbal ability” (p. 15). Vocabulary correlates highly with reading comprehension. “Various factor analyses have consistently shown that knowledge of word meanings is the dominant factor in reading comprehension” (p. 16). “Vocabulary level is a useful predictor of academic ability, even for courses like Chemistry that do not emphasize language usage” (p. 16).
Bowker, R. (1989). Vocabulary as a predictor of your SAT-Verbal score. Bulletin #142. Chicago: Johnson O’Connor Research Foundation. Notes the importance of vocabulary for academic success, that vocabulary can be improved at a faster than normal rate through study, and that this is particularly important for low vocabulary students, especially at a time when such improvement is most likely to prove most fruitful (the years prior to college entrance). Notes high correlation between score on O’Connor test and SAT-Verbal score, and includes chart to convert between Foundation’s VSS (Vocabulary Scale Score) and SAT-Verbal. Gave Foundation test to approximately 100 high school students who took the SAT a month later. Notes that there is a set of words that fall within a certain range of levels of difficulty which are known to high SAT verbal scorers which are not known to low SAT scorers and suggests learning these words in order of difficulty. “What is important…is that the words…are not rarefied, specialized terms of interest only to a linguist, but words that are in constant use in our language. The average student in the low SAT group probably recognizes most of them, but has only a vague, confused idea of their meanings.”
Human Engineering Laboratories. (1980). Validity of English vocabulary tests. Statistical Bulletin #1980-34. Boston: Human Engineering Laboratories. Discusses content, convergent, and criterion validity of JOCRF vocabulary tests. Regarding content validity, notes that Foundation tests measure reading vocabulary, measure breadth rather than depth of knowledge, and (in the case of polysemous words) test for a single meaning. Regarding convergent validity, notes correlations, based on other Foundation studies, of .91 with Nelson-Denny vocabulary, .79 with SAT, and .71 with Educational Records Bureau Comprehensive Testing Program, Vocabulary Subtest (level 5), suggesting that “these tests are all measuring the same trait, despite considerable differences in approach” (2). Examines criterion validity for occupations, and notes that Foundation samples may not be representative of occupation populations, but notes close agreement with U.S. Employment Service data for various occupations. In general the data suggest “that vocabulary level is an important factor in the level of occupation attained” (8). Notes high correlation between vocabulary knowledge and intelligence test scores. “It seems clear that most intelligence tests are primarily measures of verbal ability” (8). Notes high correlations between Foundation vocabulary test scores and reading measures and between vocabulary scores and academic success in various school subjects. “Vocabulary level is a useful predictor of academic ability, even for courses like Chemistry that do not emphasize language usage” (9).
Mathiasen, R. (1984). Predicting college achievement: A research view. College Student Learning, 18, 380-386. Vocabulary correlates highly with SAT Verbal and with ACT and these tests correlate with college achievement.
Richek, M.A. (1988). A bonanza of golden words. Houghton Mifflin/English Basics, Jan., 1988, p. 6. “A person’s vocabulary is the single best predictor of his or her reading comprehension and performance on standardized verbal tests such as the SAT. John B. Carroll…described the verbal SAT as chiefly a measure of ‘advanced vocabulary knowledge’. Correlational analyses over several decades have also shown that meaning vocabulary is the single most important predictor of a person’s intelligence quotient.” “Many vocabulary researchers feel that words encapsulate concepts, and that to know a word is to understand the concepts they represent.” “The first learning of a word provides a base for building further concepts.” Describes techniques for setting up a rich association environment for learning new words.
Trembly, D. (1966). Laws of learning general and specialized vocabularies. In Proceedings of the Meeting of the American Psychological Association. (Paper presented at the American Psychological Association annual convention, New York.) “Every word in the English language appears to have its own order of difficulty. Along the scale of difficulty, each person knows all words up to the frontier of his own knowledge and very few beyond. Slightly familiar words, at one’s frontier, are most efficiently learned. The sequential nature of the learning process also applies to specialized bodies of knowledge” (abs). The paper presents O’Connor’s basic principles. Trembly found a strong correlation between verbal ability and grade-point averages for 380 college freshmen. The paper provides samples of “frontier words” for students at various SAT-Verbal levels. “How can the college students understand what he reads or hears if he does not know the meaning of words like basic, or prediction, or diameter, or vital? Would anyone expect the average twelve year old, or fourteen year old to succeed in college?” (6).
Weizman, R. (1982). The prediction of college achievement by the SAT and the high school record. Journal of Educational Measurement, 19, 179-191. Vocabulary correlates highly with SAT Verbal, which correlates with college achievement.
Wilder, R. (1989). A multiple choice of SAT cram courses. U.S. News & World Report, Feb 27, 65-68. SAT prep courses are a $100 million a year business. Courses cost $65-$600. 50+ programs. Average gains of 80-160 points, with math scores increasing more than verbal. SAT used by 2,600 colleges. Coaching does work. Practice effect accounts for 40 point lessons. Kaplan Educ. Cntr. course has 45 hrs class + 30 hrs homework; $495. Claims 150 point average gain, but probably 50-100 more accurate. Princeton Review yields 110 points gain for similar hours and $600. Only 14% of those taking school prep courses improved their scores.