The shipley-hartford scale and the doppelt short form as estimators of wais IQ in a state hospital population

Abstract

PROBLEMThe senior author@) has reported the use of the Shipley-Hartford scale in estimating Wechsler-Bellevue scores in two VA populations. Further, several investigators (Himelstein @), Olin and Reznikoff @)) , have evaluated the predictive accuracy of Doppelt's(') short form WAIS in psychiatric settings. Because the conversion of Shipley scores to Wechsler equivalents among incoming patients at this hospital yielded consistent over-estimates when the tables developed in the VA settings were used, and in order to compare the efficiency of the Shipley scale with that of a validated "short cut" to WAIS estimation, the two methods were evaluated on a randomly selected sample of patients admitted to the Fergus Falls State Hospital. PROCEDURES AND SUBJECTSAll patients between the ages of 15 and 65 admitted to the hospital were screened for reading ability' and general (physical, psychiatric) amenability to psychological testing.Approximately 45a/, of all patients admitted during a consecutive 15 month period met the criteria established, and were given a routine battery of psychological tests, one of which was the Shipley-Hartford scale, within the first week of hospitalization. From that group of patients a random sample of 30 was selected serially, and each patient was given a full (eleven subtest) WAIS within one week of the time at which the group tests (including the Shipley) had been completed.The patients selected ranged in age from 16 to 62 (mean 40.3, sigma 13.5), and included 19 psychotics (12 schizophrenics), 4 neurotics, and 7 of miscellaneous diagnoses. RESULTSWAIS full scale I& ranged from 67 to 130 (mean 96.1, sigma 14.5), while Shipley total scores ranged from 13 to 70 (mean 37.0, sigma 17.6). Intercorrelations between the WAIS and Shipley subscales indicated that Shipley total score (uncorrected for omitted vocabulary items) was the best predictor of WAIS full scale I&. The product moment correlation coefficient obtained between the latter two variables was .90, and the standard error of estimate in predicting WAIS I& from Shipley total score was 6.33. Employing Wechsler's scheme of intellectual classification, it was found that the estimate of WAIS I& made from Shipley total score correctly classified eighteen (60%) of the thirty subjects.Doppelt short form I& estimates were obtained, and the correlation between such estimates and actual WAIS full scale I& was found to be .960. The standard error of estimate of I& from the Doppelt formula was 4.06, while the accuracy of classification of the latter method was 67y0. These figures are in substantial agreement with those reported by others in previous evaluations of the Doppelt formula.Though estimation of WAIS full scale TQ by the Doppelt short form method tends to yield somewhat better predictions than does estimation from the ShipleyHartford, the fact that scores on the latter scale are highly related to WAIS scores, and in addition, the fact that the scale requires no professional time for administration and scoring, justifies its use as a brief me...

Estimating WAIS IQ from the shipley institute of living scale using continuously adjusted age norms

“…It also correlates very highly with most standard intelligence tests, including the Wechsler-Bellevue Adult Intelligence Scale (Garfield & Fey, 1948;Grayson, 1951;Sines, 1958;Wright, 1946), the Army General Classification Test (Sines, 1958;Watson & Klett, 1968), the Revised Beta Examination (Mack, 1970;Watson & Klett, 1968), the Raven Progressive Matrices (Eisenthal & Hartford, 1971), the Slosson Intelligence Test (Martin, Blair, Sadowski, & Wheeler, 1981;Martin, Blair, Stokes, & Lester, 1977), the Quick Word Test (Martin, Blair, & Vicker, 1979a), Form B of the Wide Range Vocabulary Test (Martin et al, 1979a), and the California Short-Form Test of Mental Maturity (Martin, Blair, & Vicker, 1979b). Correlations between Shipley Total scores and Full Scale IQ scores on the Wechsler Adult Intelligence Scale (WAIS; Wechsler, 1955) range from .73 to .90 with a median correlation of .80 across 10 studies (Bartz, 1968;Bartz & Loy, 1970;Paulson & Lin, 1970b;Sines & Simmons, 1959;Stone & Ramer, 1965;Suinn, 1960;Wahler & Watson, 1962;Watson & Klett, 1968;Wiens & Banaka, 1960;Winkler, 1981).…”

“…One problem is that several of the earlier procedures did not provide adequate age correlations (Bartz & Loy, 1970;Sines & Simmons, 1959;Wiens & Banaka, 1960). As noted by Paulson and Lin (1970a), these uncorrected formulae are most appropriate for individuals who are similar in age to the samples on which the prediction formulae were derived.…”

Validity of the Doppelt Short Form WAIS in a Psychiatric Population

“…Since Doppelt fist published his Short Form of the WAIS (Doppelt, 1956) there have been other studies showing similar results (Fisher & Shotwell, 1959;Olin & Reznikoff, 1957;Sines h Simmons, 1959;Steme, 1957;Himmelstein, 1957a,b,c;Monroe, 1966) and one discordant study (Kramer & Francis, 1965). Duke (1967) suggests that this 'may be an unusual individual deviation since the other authors using a general psychiatric population were more in agreement with about one third of their patients being mis-classified instead of about one half'.…”

Estimating WAIS-R IQ from the shipley institute of living scale: A replication

“…The mean differences between estimated and obtained IQs were 1 1.32 (timed) and 8.63 (untimed), with correlations of .74 and .72, respectively, for the two conditions. Before the publication of Zachary et al, the authors had been using the WAIS norms of Sines and Simmons (1959) to estimate WAIS-R IQs. Analysis of our own data at that time showed that these estimates were approximately 14 IQ points too high.…”

“…Secondarily, the obtained estimates were to be compared with results derived from the norms of Sines and Simmons (1959).…”