vital and health statistics; series 11, no. 128 (9/73)
TRANSCRIPT
Data from theNATIONAL HEALTH SURVEY
.
Series 11Number 128
IntellectualDevelopmentof YouthsAs Measuredby a Short Formof the
WechslerUnitedStates
IntelligenceScale
A discussionof results from the Vocabulary and Block Designsubtestsofthe WISC administered to a national probability sample of noninstitu-tionalized youths 12 through 17 yearsof age.Description of the derivationof percentile equivalents, normalized scaled scores, and a short-formestimate of Full Scale Scores.Analysis of variations in raw scores,scaledscores, and estimates of Full Scale Scores associatedwith age, sex, andgradeplacement.
DHEW Publication No. (HRA) 74-1610
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFAREPublic Health Service
Health Resources AdministrationNational Center for Health Statistics
Rockville, Md. September 1973
Vital and Health Statistics-Series 1 l-No. 128
For sale by the Superintendent of Documents, U.S. Government Prfntfng Office, IVashfngton, D.C. 20402- Price 8(Jcents
NATIONAL CENTER FOR HEALTH STATISTICS
EDWARD B, PERRIN, Ph.D., Acting Director
PHILIP S, LAWRENCE, SC.D., Deputy Director
GAIL F. FISHER, Assktant Director for Health Statistics Development
WALT R, SIMMONS, M.A., Assistant Director for Research and Scientific Development
JOHN J. HANLON, M.D., Medical Advisor
JAMES E. KELLY, D.D.S., Dental Advisor
EDWARD E. MINTY, Executive Officer
ALICE HAYWOOD, Information Officer
DIVISION OF HEALTH EXAMINATION STATISTICS
ARTHUR J. McDOWELL, Director
GARRIE J. LOSEE, Deputy Director
PETER V. V. HAMILL, M.D., Medical Advisor, Children and Youth Program
HENRY W. MILLER, ChieJ Operations and Quality Control Branch
LINCOLN I. OLIVER, Chiej PsychoZogica2 Statistics Branch
HAROLD J. DUPUY, Ph.D., PsychoZogica2Advkor
COOPERATION OF THE BUREAU OF THE CENSUS
In accordance with specifications established by the National
Health Survey, the Bureau of the Census, under a contractualagreement, participated in the design and selection of the -sample,
and carried out the first stage of the field interviewing and certainparts of the statistical processing.
Vital and Health Statistics-Series 1 l-No. 128
DHEW-Publication No. (HRA) 74-1610
Library of Congress Catalog Card Number 72-600141
CONTENTS
Page
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .SourceoftheData . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PsychologicalTestBattery . . . . . . . . . . . . . . . . . . . . . . . . . . .
TheWechslerIntelligenceScale . . . . . . . . . . . . . . . . . . . . . . . .Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ShortFormsoftheWISC . . . . . . . . . . . . . . . . . . . . . . . . . .
TestingProcedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Verbal Development–Vocabulary Raw Scores . . . . . . . . . . . . . . .Performance–Block Design Raw Scores . . . . . . . . . . . . . . . . . . .Comparison With the WISC Standardization Group . . . . , . . . . . . . .Normalized Scale Scores . . . . . . . . . . . . . . . . . . . . . . . . . . .Construction ofa Short Form... . . . . . . . . . . . . . . . . . . . . .Percentile Equivalents . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Listof Detailed Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Appendix I. Technical Notes... . . . . . . . . . . . . . . . . . . . . . .The Survey Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Extent of Missing Test Results and Imputation Procedures Used . . . . . .Sampling and Measurement Error . . . . . . . . . . . . . . . . . . . . . .Hypothesis Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Small Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Quality Control for the Psychological Test Battery . . . . . . . . . . . . .
111
2
334
5
6688
121213
13
16
18
20
3434353637383940
Data not
Category
Quantity
SYMBOLS
available---------------------------------------- ---
not applicable ------------------------------- . . .
zero ---------------------------------------------- -
Quantity more than Obut less than 0.05 ----- 0.0
Figure does not meet standards ofreliability or precision ------------------------------ *
INTELLECTUAL DEVELOPMENT OF YOUTHS
AS MEASURED BY A SHORT FORM OF THE
WECHSLER INTELLIGENCE SCALE
James Scanlon, Division of Health Examination Statistics
INTRODUCTION
General
This report presents information on aspects ofthe intellectual development of youths 12through 17 years of age in the noninstitutionalpopulation of the United States, as measured bytwo subtests of the Wechsler Intelligence Scalefor Children (WISC) used in the Health Exam-ination Survey of 1966-70. The two WISCsubtests used, for reasons explained below, werethe Vocabulary and Block Design subtests. Tworeports describing WISC test results of a previousHealth Examination Survey of children 6through 11 years of age have been published. 132
In this report on WISC findings from thesurvey of youths, analysis is limited to variationsin test scores associated with age, sex, and gradeplacement. Construction of normalized scaledscores and of a short-form estimate of WISC FullScale Scores for use in the analysis of other datacollected in the survey is also described. Subse-quent reports will explore the relationshipsbetween WISC scores and the behavioral, bio-medical, socioeconomic, and other psychometricdata gathered in the survey.
Source of the Data
Test results analyzed in this report wereobtained from the Health Examination Survey, amajor program of the National Center for Health
Statistics (NCfiS), authorized under the Na-tional Health Survey Act of 1956 and subse-quent amendments as a continuing Public HealthService activity.3 Complementing other NCHSprograms aimed at assessing the healtk status ofthe American people, the Health ExaminationSurvey collects and analyzes data gathered bydirect physical examinations, tests, and measure-ments performed on probability samples of theU.S. population. The Survey is conducted as aseries of separate, cross-sectional programs re-ferred to as “cycles.” Each cycle is limited tosome specific segment of the U.S. populationand to certain aspects of the health of thatsegment of the population. Since 1960, threeseparate surveys, or cycles, have been com-pleted. Cycle I was concerned with adults aged18 through 79 years in the noninstitutionalpopulation of the United States and was com-pl:ted in 1962.4~5 Cycle 11 was a survey ofchildren aged 6 through 11 years, completed in1965;6~7 and Cycle III, which was completed in1970,8 focused on youths 12 through 17 yearsof age. For the survey of youths, on which thisreport is based, a probability sample of theNation’s noninstitutionalized youths 12 through17 years of age was selected and examined. Fieldsurvey operations began in March 1966 and were~ompleted in March 1970. Of the 7,514 youthsselected for the sample, 6,768 were examined, aresponse rate of 90 percent. Because of thesample design, adjustment for nonresponse, andweighting procedures used, examination results
1
may be considered representative of the approx-imately 23 million noninstitutionalized youthsin the United States 12 through 17 years of ageat the time of the surveys Sampling errorsassociated with estimates shown in this reportare presented in appendix I.
Each youth in the sample was administered a3-hour, single-visit examination in a mobileexamination center specially constructed for thesurvey. The examination focused primarily onfactors related to biological and psychologicalaspects of growth and development. It includedexaminations by a physician and a dentist, testsadministered by a psychologist, and a variety ofadditional tests and measurements performed bytechnicians. To insure that the survey wouldprovide comparable data on growth and develop-ment and on health ‘characteristics throughoutthe continuum of childhood and adolescence,many of the tests and measurements performedon the youths are the same as those, that werecarried out on children aged 6 through 11 yearsin the previous cycle, with some modificationsfor the difference in ages. In addition, inforrna-fion relating specifically to adolescent healthand behavior was collected.
To supplement data obtained from the exami-nation, several questionnaires were employed.Among these were a household questionnaireadministered by a U.S. Bureau of the Censusinterviewer to obtain demographic and socio-economic information, two medical histories ofthe sample youth, one completed by a parentand another by the youth, and a health behaviorquestionnaire completed by the youth at theexamination center. For those adolescents inschool, information on grade placement, teach-ers’ ratings of behavior and adjustment, anddetails of any health problems known to theteacher were requested from the school at-tended.
The grade placement of each sample youthwas obtained from the questionnaire sent to theschool each youth attended. For youths onsummer vacation, the grade placement recordedwas the grade the youth would enter in the faILIf the school questionnaire was not available fora sample youth, grade placement or the fact ofhaving left school was obtained from thepsychological test record forms.
Birth certificates were obtained for verifica-tion of the youth’s age and for other facts
relating to his or her birth. The age recorded foreach youth was his age at last birthday as of thedate of examination and was confirmed bycomparison with the date of birth entered onthe birth certificate. The age criterion forinclusion in the sample was defined in terms ofage at the time of the first interview. Since theexamination usually took place 2 to 4 weeksafter the interview, some of the youths whowere 17 years of age at the time of interviewbecame 18 by the time of the examination.There were 58 such cases. In the adjustment andweighting procedures and in this analysis, theseyouths were included in the 17-year-old group.
All information was collected under condi-tions of confidentiality. More detailed informa-tion on the survey plan, sample design, exam-ination content, and operation of the survey ofyouths is presented in appendix I and in aprevious report.8
THE PSYCHOLOGICAL TEST BATTERY
The psychological test battery included inboth the survey of chiIdren and the survey ofadolescents was the result of consultation withchild psychologists from the academic commu-nity and the National Institute of Mental Health.By using essentially the same battery of tests foryouths as that used in the children’s survey, itwas believed possible to assesscertain aspects ofintellectual and, to some extent, emotionalgrowth and development, on a comparable basisthroughout childhood and adolescence.
The 70-rninute battery contained measures ofintellectual development (both verbal and per-formance), school achievement, literacy (bothreading and writing), and aspects of personalitydevelopment. The Vocabulary and Block Designsubtests of the WISC and a modified version ofthe Goodenough-Harris Drawing Test were theprincipal measures of intellectual developmentused. Two subtests of the Wide Range Achieve-ment Test (wRAT) were included to provideestimates of school achievement in the basicskills of oral reading and arithmetic computa-tion. Inclusion of five cards of the ThematicApperception Test (TAT) represented an at-tempt to measure some aspects of personality, aswell as oral speech and communication. Finally,to assess the level of illiteracy among youths intie Nation, a brief test of literacy was developed
2
for the survey under a contract with theEducational Testing Service of Princeton, NewJersey, and administered to each youth in thesample. The development of this test has beendescribed in an earlier report.g
A study evaluating the test battery as admin-istered in the survey of children, which did notinclude the Brief Test of Literacy but wasotherwise almost identical to the test batteryemployed in the present study, was conductedon a contractual basis by Dr. S. B. Sells of theInstitute of Behavioral Research, Texas ChristianLhiversity.10 That study provided a review ofthe literature pertaining to research and evalua-tion of the battery components, recommenda-tions concerning the types of inferences thatcould appropriately be made about the testresults, and recommendations relating to addi-tional research considered necessary to makeproper use of the data collected.
An additional contractual study relatingspecifically to the WISC was completed by JaneMercer and Joyce Smith of the University ofCkAforniaatRiverside.11 This analysis evaluatedthe use of the Vocabulary and Block Designsubtests as a basis for estimating Full ScaleScores among children from different socio-economic levels and ethnic groups. It alsoexamined the amount and direction of errorlikely to occur if these subtests were used toestimate rates of retarded intellectual develop-ment in those populations.
THE WECHSLER INTELLIGENCE SCALE
Background
The WISC, published in 1949, extended thewell-known Wechsler scales for adolescents andadults into the age range of 5 through 15years. 10Y12 Since its publication, the WISC has~cen the subject of extensive investigation andhas been used extensively in schools and clinicsw an individual measure of general intellectualdevelopment. 10
Much has been written about the concept ofintelligence, and attempts to define it have been~egion.19 For the purposes of this report, it isnecessary to focus on the concept of intelligence~mplicit in the Wechsler Scales. That concept,
embodied in the WISC, assumes an aggregate or$obal capacity of the individual to act purpose-fully, to think rationally, and to deal effectivelywith his environment. 12 ~14J15 While intelli-gence is operationally defined and assessed interms of performance on a composite of abilitytasks, such as tests of verbal and spatial abilityand tests of abstract and arithmetic reasoning, itis also considered to be part of the broaderconstruct, personality. 1Z The thinking under-lying the WISC is that intelligence cannot beseparated from the rest of the personality, and adeliberate attempt was made in the developmentof the test to take into account other personal-ity factors, such as persistence, drive, and energylevel, which were assumed to contribute to thetotal effective intelligence of the individual. Thiseffort is reflected both in the composition of theWISC and ‘in the equal weights assigned to the~eparate subtests.1Z~1q No attempt was made inthe WISC to construct a series of tests thatmeasure “primary abilities” or to order theminto a hierarchy of relative importance.
The WISC consists of 12 subtests (six Verbaland six Performance), two of which are tieatedw supplementary and not routinely used. Animportant innovation in the Wechsler scales wasthe use of the deviation intelligence quotient.This measure is considered a superior alternativeto the mental age concept and evaluates testperformance on the basis of distributions ofscores of representative samples of persons ofcomparable chronological age. In the standard-ization of the WISC, Wechsler kept the standarddeviation of intelligence quotients constant fromyear to year, with the result that when tested atdifferent ages an individual’s obtained deviationIQ would not vary unless his actual test perform-ance varied when compared with that of hispeers. Raw scores for each subtest are convertedto scaled scores. These are normalized standardscores which have a mean of 10 and a standarddeviation of 3 for each age level. The sum of thefive scaled scores for the Verbal Series consti-tutes the Verbil Scale Score (VS); and similarlythe Performance Scale Score (PS) is the sum ofthe scaled scores for the five subtests of theperformance Series. The Full Scale Score (FS) isthe sum of the Verbal Scale and the Perform-Wce Scale Scores. Deviation intelligence quo-tients are provided in the WISC manual for VS,J%, and FS. Deviation IQ scales at each age
3
interval have a mean of 100 and’ a standarddeviation of 15.
6hort Forms of the WISC
Because time limitations must inevitably beimposed on a multidisciplinary endeavor such asthe Health Examination Survey, no single factor,whether physiological, dental, physical, orpsychological, could be evaluated as thoroughlyas desired. Thus, compromises were necessary inplanning the test battery. Time limitations didnot permit administration of the full WISC inaddition to the other tests and measurementsSelected for the appraisal of growth and develop-ment of adolescents. The test battery adoptedprovided for the administration of two subtestsof the WISC, the Vocabulary and Block Designsubtests. These were selected to serve as ashort-form test from which estimates of FullScale WISC IQ’s could be made. In addition toproviding a general index of intellectual develop-ment, the two subtests can be interpretedseparately as measures of verbzd developmentand the performance (nonverbal) aspects ofintellectual development. Some discussion isappropriate, therefore, concerning the validityof a composite based on these subtests for theestimation of Full Scale Scores in the context ofthe Health Examination Survey.
A number of investigators have assessed theefficiency and accuracy of various combinationsof two or more subtests of the WISC inestimating Full Scale Scores. 16-3 z Many ofthese studies have been based on small samplesof exceptional children such as the mentallyretarded, the physically disabled, or childrenreferred to child guidance clinics or social serviceagencies.
Correlations between a linear Vocabulary andBlock Design combination and Full Scale Scoresin these special populations have been reportedby a number of investigators. Finley andThompsonl G obtained a correlation of 0.68amen 309 mentally retarded children; Green-mun 17 obtained one of 0.92 among 632 refer-rals to the Child Guidance Clinic at TexasChristian University; Simpson and Bridgesl 8found the correlation to be 0.87 for 120children and youths referred to the Division ofChild Guidance of the Oklahoma City Public
Schools; Wight and Sandryl 9 obtained 0.91among 83 children hospitalized for a physicaldisability, and Mumpower20 found the correla-tion to be 0.95 among 50 children and youthsreferred for evaluation to the Special EducationDepartment at the University of SouthwesternLouisiana.
C)ne study, by Silverstein,2 1 dealt with afairly large sample of predominantly normalchildren. He determined the correlations withFull Scale Scores of all possible short forms oftwo, three, four, and five subtests using theW-MC standardization data for 200 children atthree age levels: 7% years, 10% years, and 13?4years. The best dyad predictor at age 10?4 yearswas the Vocabulary and Block Design combina-tion, while at ages 7?4 and 13?4 this dyad wasamong the better predictors. Differential weight-ing of the subtest scores did not result inappreciably higher predictive validities for FullScale IQ.
A second large-scale study, mentioned above,was completed by Mercer and Smith undercontract with the Health Examination Survey. 11The authors used correlation and regressionanalysis of data from 1,310 children attendingpublic elementary schools in Riverside, Cali-fornia, during the school year 1967-68 toevaluate the use of the Vocabulary and BlockDesign subtests of the WISC for predicting FullScale Scores of children from different socio-economic and ethnic groups. The sample con-sisted of children aged 6-11 years from threegroups; 505 were English-speaking Caucasians(called “Angles” in the study), 487 were ofMexican-American heritage; and 318 wereNegro. Of the total sample, 1,270 were enrolledin regular classes and 40 were enrolled in classesfor the educable mentz=dly retarded. WhenVocabulary and Block Design scaled scores wereused to predict Full Scale, Verbal, and Perform-ance IQ’s, itwas found that estimates were notmaterially improved by categorizing the separateethnic groups according to socioeconomic statusor age. Multiple correlation coefficients wereapproximately the same for all three ethnicgroups, and the authors concluded that Vocabu-lary md Block Design provided an essentiallyequivalent short form for predicting Full ScaleScores for the three ethnic groups. The multiplecorrelation coefficients were 0.87 for Anglo
children, 0.83 for Negro children, and 0.85 forthose of Mexican-American heritage; the stand-ard errors of estimate were 7.5, 6.8, and 6.7,respectively.
A score representing the simple sum of thetwo subtest scaled scores yielded a correlation of0.88 with Full Scale Scores for all ethnic groupscombined. Vocabulary and Block Design werethe optimal dyad for Anglo and Mexican-American children. Vocabulary and ObjectAssembly were the optimal dyad for Negrochildren, but the difference in accuracy ofprediction was minimal (R = 0.835 versus 0.827,or a 0.008 difference). As expected, VocabularyScale Scores in this study correlated quite highlywith Verbal Scale Scores for all three ethnicgroups (0.82 for Anglo children, 0.80 for Negrochildren, and 0.82 for Mexican-American chil-dren). Similarly, Block Design Scale Scores werefound to correlate 0.76 with Performance IQamong Anglo children, 0.70 among Negro chil-dren, and 0.71 among Mexican-American chil-dren. The authors concluded that no other dyadwould have produced better overall predictionsof Full Scale IQ for the tlyee ethnic groups intheir sample, and that the Full Scale IQ for thatsample could be predicted with about the sameaccuracy from the unweighed sum of the twoscaled scores as from differentially weightedscaled scores.
~~te has confirmed Wechsler’s data indicatingthat Vocabulary and Block Design are the mostreliable subtests of the WISC battery. 12$33J3agen and Cohen in the United States and Gaultin Australia have reported that both of thesesubtests are highly loaded on the so-called“general” factor of intelligence obtained infactor analytic studies of the WfSC over theentire age range of 5 through 15 years.3 4-36
Hence the weight of all available evidencesupports the conclusion that the two subtestsselected for use in the survey constitute the bestdyad available for estimating Full Scale IQ andthat the validity of this dyad for the estimationof Full Scale Scores is high. While this dyad isvalid and appropriate for the analysis of groupstatistics in a research setting such as the HealthExamination Survey, its use as the sole instru-ment in the comprehensive assessment of intel-lectual functioning in clinical settings is obvi-ously not advocated.11
TESTING PROCEDURES
The Vocabulary and Block Design subtests ofthe WISC were the second and third procedures@ the 70-minute psychological- test battery,following administration of the Wide RangeAchievement Test. The testing was accomplishedin a small, adequately lighted, climate-controlled, sound-treated room in the mobileexamination center. The psychologists whoadministered these tests had at least a master’sdegree and had experience in administering tests.There were two psychologists on the examiningteam at all times. The examiners were selected,trained in the field-testing procedures, andsupervised by the Office of the PsychologicalAdvisor to the Health Examination Survey,under the direction of an experienced Ph.D.-level psychologist. In the initial training andensuing supervision of the examiners, strongemphasis was placed on uniform methods of testadministration, scoring, and recording of data.During the course of the survey of youths, atotal of 12 examiners participated in administra-tion of the battery. Quality-control proceduresemployed in administration of the psychologicaltest battery are described in appendix I.
The standard WISC Record Form was usedfor recording responses and scores. The subtestswere administered as specified in the WISCManual except that, in most cases, each youthwas started with the sixth word on the Vocabu-lary subtest.
The Vocabulary subtest of the WISC consistsof 40 words arranged in order of increasingdifficulty. The youth was asked what each wordmeans. The response to each word was scored662,99,<~,), or “O,” except words 1 through 5which were scored “2” or “O.” The maximumraw score was 80. The test was discontinuedafter five consecutive failures. Since each youthnormally started at word 6, he was automati-cally credited with a score of 2 for words 1through 5 if 2-point responses were given forwords 6 through 10. If no credit or a l-pointresponse was given for any of words 6 through10, the examiner went back to word 5 andworked backward until the criterion of fiveconsecutive 2-point responses was met, or all ofwords 1 through 5 had been administered.
The Block Design subtest of the WISC re-quires the examinee to reproduce designs con-structed of small colored wooden cubes withinprescribed time limits. There are 10 such de-signs, increasing in complexity and requiringeither four or nine cubes. Administration andscoring followed procedures outlined in the testmanual. The maximum score was 55 points.
44 further word on the sample design and itspossible effect on WISC findings in this report isappropriate. Because of both operational consid-erations and the desire to obtain longitudinaldata, over 30 percent of the youths comprisingthe Cycle III sample had also participated in thesurvey of children completed in 1965. Since theshortest test-retest period in the two surveys wasabout 2Y2 years, it was assumed that WISCresults for the retested youths were not appreci-ably influenced by any practice effect.
FINDINGS
Verbal Development–Vocabulary Raw Scores
Varz”ations with age.–As mentioned above,the Vocabulary subtest of the WISC is a measureof verbal aspects of intellectual development,correlating highly with both the Verbal Scaleand Full Scale Scores of the WfSC. Results ofthis test from the Health Examination Surveyindicate, not surprisingly, that verbal develop-ment is positively correlated with age within theage group of 12 through 17 years (table 1 andfigure 1). Mean Vocabulary raw scores increasedfrom 36.5 points among 1Z-year-oIds to 45.2points among 17-year-olds. The highest possibleraw score was 80 points. Examination of un-soothed and smoothed mean Vocabulary rawscores for 4-month age groups reveals a similar
50
rVOCABULARY
10Ir)~
12 13 14 15 16 17 18
AGE IN YEARS
50
40
10
0
Boys
--- Girls
BLOCK DESIGN
I I I I I I I
12 13 14 15 16 17 18
AGE IN YEARS
Figure 1. Mean raw scores on the Vocabulary and Block Desigri subtests of the WISC, by sex and age: United States, 1966-70.
6
trend of higher mean raw scores with age, but ata decelerating rate (table 2 and figure 2). Thecurves were smoothed using a 3-point movingaverage technique.
The variability in raw scores attained by theyouths, as indicated by the standard deviationsof the Vocabulary subtest raw score distribu-tions at each 1-year age level, tended to increasevery slightly with age (table 1 and figure 3). Anapproximate index of relative variation, thecoefficient of variation (ratio of the standarddeviation to the mean) remained fairly constantfor the Vocabulary subtest scores over all agegroups. The two highest raw scores found in thisnational sample, scores of 76 and 73, wereattained by two 16-year-olds. Thus, it appearsthat the subtest had an adequate “ceiling” forthe age group tested in the survey.
Sex differences in scores. –Boys, in general,tended to perform slightly better than girls onthe Vocabulary subtest (table 1 and figure 1). Atevery l-year age level except the oldest, themean score for boys exceeded that for girls. Thedifferences ranged from 1 to 3 raw score points,but were large enough to be statistically signifi-cant, at the 5-percent confidence level only forthe 12-, 13-, and 15-year-olds. At age 17, themeans for boys and girls were identical.(Hypothesis testing procedures are described inappendix I.)
Grade placement.–As expected, meanvocabulary raw scores tended to increase withsuccessive grade levels (table 1 and figure 4).This grade differentiation was true for all agegroups combined and for each l-year age group.The mean vocabulary raw score among those
.
.; -> ,’~-?~o-,-c \o-/
20 20
10 – 10
0 I I 1 I I I I 1 I I 1 I t 1 I ! 1 I o I ? 1 I 1 1 I 1 1 I 1 r I 1 1 I 1 1 I
12 13 14 16 16 17 1s 12 13 14 15 16 17 18
AGE IN YEARS AGE IN YEARS
Figure 2. Unsoothed and smoothed mean Vocabulary and Block Dasign raw scores on the WISC, by age in 4-month age intervals:
United Statas, 1966-70.
Block Design
Vocabulary
I I I I I I I
12 13 14 15 16 17 18
AGE IN YEARS
Figure3. Standard deviations of raw scores on the Vocabulary
and Block Design subtests of the WISC, by age: United
States, 1966-70.
relatively few youths, mostly 12- and 13-year-olds, whose grade placement at the time oftesting was less than 5th grade was 18.9, whilethe mean for youths in the 12th grade at thetime of testing was 49.0. High school graduateswith no additional formal schooling had a meanof 47.5, and youths who were attending schoolbeyond high school had a mean of 54.1. Youthswho had left school before graduating (mostly16- and 17-year-olds) attained a mean of 32.1.Generally, in each grade younger individualsobtained higher mean raw scores than did olderones (table 1).
Consistent with the sex differences in meanscores within each age group reported above,boys attained higher mean scores than did girlsat almost every grade level, including youths inspecial ungraded classes for the mentally re-tarded and physically handicapped, and thosewho left school before graduating.
performance-Block Design Raw Scores
Variations with age.–The Block Design sub-test is an approximate measure of the perform-ance or nonverbal aspects of intellectualdevelopment. Results for this subtest alsoindicate a positive association with age in the 12-through 17-year-old population studied, with atendency to level off. in the oldest groups(figures 1 and 2). Mean raw scores on the Block
Design subtest12-year-olds to3). The highest
increased from 24.0 among33.0 among 17-year-olds (tablepossible ra; sc&e on this test
was 55 and was attained by about 0.4 percent ofthe sample youths.
Variability in Block Design raw scores washigher at every age level and for both sexes thanthat observed for the Vocabulary subtest (tables1, 2, and 3 and fi~re 3). The larger BlockDesign standard deviations, however, are prob-ably a result of the O versus 4 scoring of mostitems on this subtest, not counting bonus points.No age trends in standard deviations weieapparent, and the relative variation of BlockDesign raw scores tended .to decrease somewhatwith increasing age.
Sex differences in scores.–Throughout theage range tested, boys on the average scoredhigher than did girls on the Block Design subtest(table 3). The differences were statisticallysignificant at all age levels and ranged fromabout two to four raw score points.
Grade @cement. -On the Block Design sub-test, as on the Vocabulary subtest, mean rawscores tended to increase with successive gradelevels among both boys and girls (table 3). Thiswas true for each l-year age group. At everygrade level, boys attained higher scores thangirls, including youths in special placement andthose who left school before graduating. The sexdifferences in mean scores were large enough tobe statistically significant at the 5-percent con-fidence level for youths in the 6th through the12th grades and for those in special placement.There was a tendency for younger children ineach grade to perform somewhat better than didolder children on this test, up to about the 10thgrade (table 3).
Comparison With the WISCStandardization Group
The WISC was standardized in 1949 on asample of 100 boys and 100 “rlsat each year of
Yage from 5 through 15.12 J3 The standardiza-tion. sample also included 55 children, most ofthem institutionalized, who were in the chrono-logical age range 5 through 15 years anddetermined by staff psychologists of theirrespective institutions or schools as having IQ’sunder 70 but not below 50. With the exception
8
= B.,,
60
r
n Girls
50
I
VOCABULARY
40
t
.,~ Lessthan 5 ‘6 7 Bcc 5za
9
40
30
1:
20
10
0 kl.-ll
BLOCK DESIGN
L$SSthan 55’
Ii_t6 ‘ .7
10 11
‘B “9 “10 ’11
GRADE
12 High Higher Special Leftschool than placement school
graduate 12 beforegraduation
12 High Higher Special Leftschool than placement school
graduate 12 beforegraduation
Figure 4, Mean raw scores on the Vocabulary and Block Design subtests of the WISC, by grade in school and sex: United Statas,1966-70.
9
of the known low IQ group, most of whom weretested within 2 months of their midyear, eachchild was tested within 1?42months of hismidyear. Only white children were included inthe standardization group.
Within each year of age and for the totalgroup, the standardization sample was selectedso that it would be representative of the totalU.S. population, according to data from the1940 U.S. Census of Population, with regard tothe following three variables:
1.
2.
3.
The proportion of the U.S. populationliving in each of four geographic regions ofthe country-( 1) the region comprised ofthe New England and Middle AtlanticStates, (2) the region comprised of theNorth Central States, (3) the regioncomprised of the South Atlantic andSouth Central States, and (4) the regioncomprised of the Mountain and PacificStates.
The proportion of the 1940 U.S. popula-tion living in rural and urban areas.
The proportion of the total adult em-ployeb &hite male population in nineoccupational classification groups.
The Midwest group was reduced in size andthe Western group was made slightly larger thanthe U.S. proportions in 1940 to allow forwartime and postwar population shifts duringthe 1940’s. Standardization testing took place in85 communities. The sample was drawn fromschool populations, except for the majority ofthe 55 children of known low IQ’s, most ofwhom were institutionalized.
Except for a slight bias for urban and small-town areas as opposed to rural areas, and asidefrom the fact that the standardization wasaccomplished on a native white population only,the standardization sample for the WISC hasbeen regarded as good.1 O
The present study was based on findings from6,768 examined youths who may be consideredrepresentative of the estimated 23 million non-institutionalized youths 12-17 years old in theUnited States as of the midpoint of the survey.Testing took place at 40 locations between 1966and 1970. The sample design used in the ,survey
was a multistage, stratified probability sample ofhouseholds in land-based segments. A moredettied description of the sampling design ispresented in appendix I and in previous publi-cations.6-8 The sampling frame insured therepresentativeness of the sample with respect todegree of urbanization, region, and the rate ofpopulation change from 1950 to 1960, the latterbeing an indirect indicator of the economiccondition of the area. Data used in this reportfor each sample youth were inflated in theestimation process to characterize the largeruniverse of which the sample youths are repre-sentative and to include an adjustment for thenonresponse group. This made the final sampleestimates of the populations agree exactly withindependent controls from the Bureau of theCensus for the U.S. noninstitutionalized popula-tion as of the approximate midpoint of thesurvey, with regard to age in years, race, and sexfor the study population. Thus, findings fromthe present study are based on from approxi-mately 960 to 1,200 youths at each year of age12 through 17, and may be considered repre-sentative of the total noninstitutionalized U.S.population of 12- through 17-year-olds withregard to age, race, sex, region, population size,and rate of population change between 1950and 1960.
Comparisons of raw score means and standarddeviations on the Vocabulary and Block Designsubtests for the 1949 WISC standardizationsample and U.S. estimates from the nationalHealth Examination Survey are shown in figures5 and 6. WISC standardization sample meansand standard deviations were estimated from theconversion tables in the WISC Manual. For every4-month age group in the range 12 through 15years, estirnat;d l@3C standardization means forthe Vocabulary subtest exceeded unsoothedmeans from the Health Examination Survey.The differences ranged from about 3 to 6 rawscore points. For the Block Design subtest also,estimated WISC standardization group meansranged from about 2 to 8 raw score pointshigher than corresponding Health ExaminationSurvey findings.
Using data from the previous survey of 6- to1l-year-olds, it is possible to compare Vocabu-lary and Block Design subtest characteristicsfrom the original Wechsler standardization
10
1949 WISC Vocabulary1111111118 HES WISC Vocabulary-mm 1949 WISC Block DesignrHm/mx. HES WISC Block Design
< ,,*,,ln181111nln1tn1:a11n8s11n11,11111*
~ ,“--”
1 1 I , 1 I , I , I12 13 14 15 16
AGEINYEARS
Figure 5, Mean raw scores on the Vocabulary and Block Designsubtests of the WISC for the WI-SC standardization group and
for U.S. estimatas from the Health Examination Survay, at 4-
month age intervals.
~ 1949 W ISC Vocabularynnaa~~~a~n~~ HES WISC Vocabulary
---- 1949 WISC Block Design
~ZZZ.’Z HES WISC Block Design
V7*7*-==-’’--*’ MPH~ mm,,,,,,,,,,,llslllmlI1lB1lBB#8,m,8n8tl88#,,,*1N11llllln1,1
I I I r I I ? I t , t I12 13 14 15 16
AGE IN YEARS
Figure6. Standard deviations of rawscore distributions on the
Vocabulary and Block Design subtests of the WISC for theWISC standardization group and for U.S. estimates from the
Health Examination Survey, at 4-month age intervals.
sample and the national Health ExaminationSurveys in more detail. The comparisons aremade for ages 7%, 10?42,and 13?4, ages which thetest developers felt were probably most repre-sentative of the age range for which the WISCwas designed.l Z At the 7%-year age level, themean Vocabulary subtest raw score was lower inthe Cycle II sample of the Health ExaminationSurvey than that estimated from the originalWISC sample, and the variability was greater.Both differences were statistically significant.1At age 10?4 years, the mean and variability ofthe distribution of Vocabulary raw scores wereboth slightly, but not significantly, greater in theHealth Examination Survey sample than in theWISC sample.1 Among the 13%-year-olds, theHealth Examination Survey mean Vocabularyraw score was significantly lower than theoriginal standardization figure, while the varia-bility was slightly, but not significantly greater.
For the Block Design subtest, mean rawscores in the Health Examination Survey ex-ceeded those found in the WISC standardizationsample at ages 7% and 10Y2and were lower thanthose found in the WISC group at age 13%. Thedifferences were statistically significant for the7?4- and 13%year age groups. Variability of thedistribution of Block Design raw scores wasslightly greater in the Health ExaminationSurvey sample than the WISC sample at age 7%,significantly smaller than the WISC group at age10Y2, and slightly, but not significantly, smallerthan the WISC sample at age 18%.
The correlations between scaled scores on theVocabulary and Block Design subtests in thenational Health Examination Survey are in fairlyclose agreement with those of the originalstandardization sample. At age 7%, the PearsonProduct-Moment coefficient of correlationfound in the Health Examination Survey,0.38 + 0.027, was slightly higher than the stand-ardization sample figure of 0.33, while at age10% it was slightly lower, 0.50 * 0.022, com-pared with 0.54 in the WISC group. At age 18%,the correlation coefficient found in the HealthExamination Survey was 0.54 * 0.159, some-what higher than that in the standardizationsample, which was 0.42.
Boys tended to score higher than girls on theverbal and performance measures in both theoriginal WISC standardization group, which
11
employed all 12 subtests of the WISC battery,and in the present study, which was limited toone verbal subtest and one performance subtest.Seashore, Wesman, and Doppelt found that, onthe average, boys in the standardization groupdid slightly better than girls at each year of agefrom 12 through 15.37 The differences found inthat study were small, but consistent, occurringprimarily in the older age groups. On the totalVerbal Scale, boys excelled girls by more than 3points at age 8 and ages 10 through 15. On thefull Performance Scale, the differences favoredboys by more than 3 points at ages 8 and 10,while the girls did better at ages 5, 6, 7, and 9.36
In the Health Examination Survey, whichincluded 16- and 17-year-olds, boys also tendedto do slightly better on the tests than did girls.On the Vocabulary subtest, average scores forboys exceeded those for girls by 1 to 3 raw scorepoints at every l-year age level except theoldest, when the means were identical. Thedifferences were statistically significant, how-ever, only among the 12-, 13-, and 15-year-olds.On the Block Design subtest, boys, on theaverage, outscored girls at every age level from12 through 17. The differences were statisticallysignificant at all age levels and ranged fromabout 2 to about 4 raw score points.
Normalized Scaled Scores
To express scores in a form that indicates ayouth’s performance on a test relative to his agegroup, to make such scores comparable amongdifferent age groups, and to construct a commonmetric for the comparison of two or more testsor subtests of different length, raw scores mustbe converted to some common measure. In theoriginal WISC standardization process, normali-zed standard scores with a mean of 10 and astandard deviation of 3 were constructed foreach of the 12 subtests within each age level.The WISC Manual provides such scores for each4-month age interval from age 5 through 15years.
In the present study raw scores from the totalU.S. sample were similarly converted to normali-zed standard scores with a mean of 10 and astandard deviation of 3 for each subtest withineach 4-month age interval.38 As in the WISCstandardization process, minor irregularities in
the progression of scaled score equivalents fromage to age were smoothed. It was assumed thatthese irregularities were due to sampling variabil-ity. Because there were few instances of suchminor deviations, and these were at the extremesof the raw score distribution, the scaled scoresare essentially a direct translation from rawscores to a normalized distribution of scaledscores with ‘a mean of 10 points and a standarddeviation of 3 points. Tables 6 and 7 present thescaled score equivalent of each raw score fromthe present study.
Mean scaled scores on the Vocabulary andBlock Desigyi subtests are shown according toage in years, sex, and grade placement in tables 4and 5 and parallel the findings on raw scoresreported above. Mean scores on each subtestwere generally slightly higher for boys than forgirls and tended to increase with successive gradelevels. Younger children in each grade tended toattain higher scaled scores than did older chil-dren in the same grade. Among the relativelysmall number of youths in the sample who leftschool before graduating, both boys and girlsobtained higher mean scores on the BlockDesign subtest than on the Vocabulary subtest;the differences were statistically significant forboth boys and girls.
Construction of a Short Form
To carry out analyses of the interrelationshipsamong intellectual development and othersurvey variables, some approximate index ofgeneral intellectual development is required,incorporating both the Vocabulary and BIockDesign subtests. For this purpose it was decidedto employ a linear transformation of the simplesum of the scaled scores on the two subtests intoa new distribution of standard scores with amean of 100 points and a standard deviation of15. This method weights the two subtestsequally and makes use of the recent representa-tive national norms provided for each subtest bythe survey. The resulting standard scores are ofthe same order of magnitude as the WISC FullScale IQ; and as constructed in the presentstudy, can range from 43 to 157. Such anindex is obviously not as reliable as one based onall 10 subtests. It is, however, a reasonableestimate of intellectual development, suitable
12
for the research purposes of the survey.1 o~l 1The basis for converting the sum of the scaledscores into index scores in this study is shown intable 8.
Variations in mean index scores associatedwith age, sex, and grade placement are presentedin table 9. With the two subtest results com-bined, boys again tended to outscore girls, themean differences being statistically significantfor each l-year age group except for the17-year-olds.
Variations in mean scores with grade aresimilar to those observed among the separatesubtest scaled scores. Mean scores tended toincrease with successive grade levels. Youths inthe normal, or modal, grade placement for theirage obtained mean scores slightly above 100, theaverage score. Youths who were accelerated ingrade placement tended to obtain above-averagescores, while older youths in each grade ob-tained lower than average scores.
Pwxmtile Equivalents
Another index of a youth’s relative standingin his age group with regard to a particularcharacteristic is the percentile equivalent. Per-centile equivalents, as defined in this report,represent the percentage of individuals in thesaynple falling below particular raw scores. Table10 presents selected percentile equivalents ofraw scores for both subtests, based on resultsobtained in the Health Examination Survey. Thepercentile equivalents were computed at l-yearage intervals from 12 through 17 for both sexescombined and for boys and girls separately andwere smoothed at the extremes of the distribu-tions. Thus, a 14-year-old obtaining a raw scoreof 48 on the Vocabulary subtest in this nationalsample surpassed 75 percent of the 14-year-oldstested. Unsoothed cumulative percent distribu-tions of raw scores for each subtest, plotted onnormal probabilityy paper, are shown in figure 7.Cumulative percentages derived from a normaldistribution would yield a straight line curve onsuch a graph. Thus it appears that Vocabularysubtest raw scores are approximately normallydistributed within each age group in this study.The cumulative percent distribution of BlockDesign raw scores is much more irregular,particularly at the lower raw score levels.
DISCUSSION
Though the roots of testing itself are lost inantiquity, the major development shaping thecontemporary testing of general intellectual levelhad its effective beginning with the developmentof the Binet-Simon Scale in 1905. Binet and hiscoworkers were seeking to measure generalintelligence in an educational setting in order toidentify children in Paris schools who wouldrequire special educational services.s%qo Thefirst such tests were considered successful be-cause they differentiated between childrenknown to do well in school and those known toperform poorly in the classroom. Most later testsof intellectual level were patterned after andvalidated following the Binet model, thus con-tinuing the influence of the scholastic aptitudeorientation on intelligence testings 9 These testswe thus indicators of both previous opportunityto learn and of ability to learn something new ina scholastic setting.
Despite the fact that Wechsler developed theWISC in protest against the measurement con-cept of mental age implicit in the Stanfordrevision of the’ Binet Scales and despite severalimportant differences between the WISC and theStanford-Binet, a number of WISC validationstudies have used the Stanford-Binet as a crit-erion, Correlations between WISC Full ScaleScores and Stanford-Binet IQ’s tend to be high,generally in the mid-80’s, with some as high asthe reliability coefficients of the tests.1‘J3 g
A number of investigators have found con-current validity coefficients between WISCscores and achievement tests or other academiccriteria clustering around 0.60.39 As pointedout above, the Vocabulary and Block Designsubtests of the WISC are the most reliable of theentire Scale, and a short form consisting of thisdyad appears to be about the best available,given the constraints and research purposes ofthe Health Examination Survey.
This study makes available for the first timeVocabulary and Block Design subtest findingsfor a national probability sample of the noninsti-tutionalized population of youths 12 through 17years of age with proportionate representationfrom W races. The sample is several times largerthan that used in the original test standardiza-tion and hence should provide an even more
13
,,,,,,,,,,,,,,--------------.—-..
RAWSCORE
12 wars13 years14Wars15 Ymrl16 was17 yam
99
99
95
W
80
70
80
e3
40
33
2s
10
s
2
1
0 10 20 30 40 54 00
RAW SCORE
Figure 7. Cumulative percent distributions of WISC Vocabulary and Block Dasign rawscoresbya9e, plotted on normal probabilitypaper: United States, 1966-70.
stable base for standardization of the twosubtests than did the original standardizationsample.
The finding from the present study thatHealth Examination Survey raw score meanswere somewhat lower than those estimated fromthe WISC standardization group in the age range12-15 years is interesting. Though this study canprovide no definitive explanation, one likelyreason “is the difference in the two sampledesigns. As mentioned earlier, Wechsler’s stand-ardization was carried out in the late 1940’s on asample of white persons only, although it did
include some children who were institutionaLized. Health Examination Survey findings arebased on a national probability sample designedto be representative of the total, noninstitution-al population aged 12 through 17 years in theUnited States during the period 1966-70, withrespect to age, sex, race, region, population size,Wd other socioeconomic variables.
A second finding from the present study thatboys tend to outscore girls on at least these twosubtests of the WISC is consistent with otheranalyses, including the Health ExaminationSurvey of children 6 through 11 years of age,l
14
Mercer and Smith,l 1 and Seashore et al., in theiroriginal standardization study of the WISC.37Regarding these differences in scores betweenboys and girls, Seashore and his coworkersconcluded that the most reasonable assumptionwas that boys and girls are equal in levels ofmental ability, but either the test items chosenturned out to be slightly biased in favor of boysor the 1949 standardization sample of boys wassomehow biased; or that both biases may beresponsible for the differences. This conclusionwas based on the fact that Terrnan and Merrillfound similar results in their 1937 revision ofthe Stanford-Binet Scales and could also find nodefinitive explanation from their data.q 1 Find-ings from the present study and the previoussurvey of children, however, are based on largesamples that are closely representative of thetotal U.S. noninstitutionalized population of6-1 7-year-olds. In the light of the results of thesetwo HES investigations, the hypothesis of bias inthe standardization sample selection of childrenand youths does not explain the sex differencesin scores. Although Seashore’s data did not showtest results separately for the two subtests usedhere, it is likely that the persistent differencebetween the scores of boys and girls on thesesubtests would have been found in the originalstudy because of the high correlation betweenthese two subtests and the total Verbal andPerformance Scales of the WISC.
Another implication of the WISC results~btained in Cycles II and 111can be drawn fromthe variation in mean indexes of verbal and
erformance level over the age range 6-17 years.E sing raw score data from both cycles it ispossible to construct approximate ~oup growth
curves for verbal development, as measured by~hc Vocabulary subtest, and for nonverbalaspects of intellectual development, as assessedby the Block Design subtest. These curves arepot true growth curves because the approachhere was essentially cross sectional rather thanlongitudinal. Also, an interval of about 3 calen-dar years separated the midpoints of the twosurveys, and as mentioned earlier, some of the
children examined in Cycle II had becomeeligible for and were selected for inclusion assample youths in Cycle III, thus introducing aquasi-longitudinal aspect to these two sequentialcross-sectional surveys. Because the time intervalbetween the two surveys was small, it is likelythat test scores from the two surveys arecomparable and can be treated, for the mostpart, as resulting from a cross-sectional design.In addition, the fact that the age range encom-passed in the two surveys was relatively smalltends to minimize one of the major difficultiesin the analysis of growth data from cross-sectional designs, namely, the differential effectthat major cultural changes may have on thepopulation characteristics of some but not all
?ge groups studied.13 ,42
Smoothed Vocabulary and Block Design rawscore means for each 4-month age interval fromages 6 through 17 are shown in figure 8. Thecurves were smoothed using a 3-point movingaverage technique. The curves of both Vocabu-lary and Block Design raw score means illustratea rapid rate of increase in the younger groupsand a somewhat decelerating rate of increase inthe older age ranges. Unsoothed means foreach l-year age group from 6 through 17 yearsof age appear in figure 9 and generally ihstratethe same trend. Not.bly absent from thesecurves and from curves of changes in smoothedraw score means from age to age is the “adoles-cent growth spurt,” a phenomenon typical ofmany physicaI aspects of growth and develop-ment.4s Recent longitudinal studies of intellec-tual growth suggest that intelligence continuesto increase gradually after the teens, that manyof the more complex intellectual skills reachtheir peak as Iate as age 50, and that only itemsthat depend on speed and intensity of response
show the decline in the mid-twenties longconsidered characteristic of intellectual develop-ment. 13
A more detaiIed discussion of sex differencesin scores on the Wechsler scales and of changesin scores over age has been published.15
15
50
r
o I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1 1
6 7 8 9 10 17 12 13 14 15 16 17 18
AGE IN YEARS
Figure 8. Smoothed mean raw scores on the Vocabulary and 810ck Design subtests of the WISC, by age in 4-month intervals: Cycles
I I and 11I of the Health Examination Survey.
SUMMARY
This report has presented findings on twosubtests of the WISC, Vocabulary and BlockDesign, administered to a national probabilitysample of youths in the Health ExaminationSurvey of 1966-70. In the survey, a probabilitysample of 7,514 youths was selected to representthe 23 million noninstitutionalized youths, 12through 17 years of age, in the United States. Ofthe 7,514 youths selected in the sample, 6,768,or 90 percent, puticipated in the survey. Be-cause of the sample design, adjustment fornonresponse, and weighting procedures used inthe survey, findings for these youths may be
considered to be representative of the totalnoninstitution,&zed population of 12- through17-year-olds in the United States with respect toage, sex, race, region, tid other socioeconomiccharacteristics.
Variations in raw scores and scaled scoresassociated with age, sex, and grade placementhave been discussed in this report. Percentileequivalents have been constructed, and thederivation of normalized scaled scores and ashort-form estimate of WISC Full Scale Scoresfor subsequent use in analysis of survey datahave been described.
Findings on the two subtests have beencompared with data estimated from Wechsler’s
16
50
40
30
20 –
10 –
I
o I I I I I I I I I I I I I I
6 7 8 9 10 11 12 13 14 15 16 17 18
AGE IN YEARS
Figure 9. Mean raw scores on the Vocabulary and Block Design subtests of the WI SC, by age: Cycles II and III of tha HealthExamination Survey.
original standardization sample, and variations inmean subtest raw scores over the age range 6-17years have also been discussed.
Selected findings include the following rela-tionships: 3.
1.
2.
Mean raw scores on both subtests tended ~to increase with age, with a tendency to “level off in the older age groups.
For every 4-month age group in the 5.12-1 5-year age range, Health ExaminationSurvey mean raw scores on each subtestwere lower than those found in the
original standardization sample, whichconsisted of white youths only but in-cluded some children and youths ininstitutions.
Mean raw scores for boys generally ex-ceeded those for girls.
Scaled Scores and estimated Full ScaleScores exhibited a tendency to increasewith successive grade levels.
Youths who had left school before gradu-ating tended to score significantly higheron the nonverbal test than on the verbaltest.
17
lNational Center for Health Statistics: Intellectual develop-
ment of children as measured by the Weclisler Intelligence Scrde,
United States. Vital and Health Statistics. PHS Pub. No.
1000-Series 1I-No. 107. Public Health Service. Washington. U.S.
Government Printing Office, Aug. 1971.2National Center for Health Statistics: Intellectual develop-
ment of children by demographic and socioeconomic factors,
United States. Vital and Health Statistics. PHS Pub. No.
1000-Series 1 l-No. 110. Public Health Service. Washington. U.S.Government Printing Office, Dec. 1971.
3National Center for Health Statistics: Origin, Pro8r~$ and
operation of the U.S. National Health Survey. Vital and Health
Statistics. PHS Pub. No. 1000-Series l-No. 1. Public Health
Service. Washington. U.S. Government Printing Office, Apr.1965.
4National Center for Health Statistics: Plan and isdtial
program of the Health Examination Survey. Vital and Health
Statistics. PHS Pub, No. 1000-Series l-No. 4. Public Health
Service. Washington. U.S. Government Printing Office, July1965.
5National Center for Health Statistics: Cycle I of the Health
Examination Survey: Sample and response, United States,1960-1962. Vital and Health Statistics. PHS Pub. No, 1000-
Series 1 l-No. 1. Public Herdth Service. Washington. U.S, Gover-nment Printing Office, Apr. 1964.
6National Center for Health Statistics: Plan, operation, and
response results of a program of children’s examinations. Vital
and Health Stat&tics. PHS Pub. No. 1000-Senes 1-No. 5. PublicHeaIth Service. Washington. U.S. Government Printing Office,Oct. 1967.
7National Center for Health Statistics: Sample design andestimation procedures for a National Health Examination Surveyof children. Vital and Health Statistics. PHS Pub. No. 1000-Series 2-No. 43. Public Health Service. Washington. U.S. Gover-nmentPrinting Office, Aug. 1971.
8National Center for Health Statistics: Plan and operation ofa Health Examination Survey of U.S. youths, 12-17 years of age.Vital and Health Statistics. PHS Pub. No. 1000-Series 1-No. 8.Public Health Service. Washington. U.S. Government PrintingOffice, Sept. 1969.
‘National Center for Health Statistics: Development of theBrief Test of Literacy. Vital and Health Statistics. PHS Pub. No.1000-Series 2-No. 27. Public Health Service. Washington. U.S.Government Printing Office, Mar. 1968.
10National Center for Health Statistics: Evaluation ofpsychological measures used in the Health Examination Surveyof children ages 6-11. Vital and Health Statistics. PHS Pub. No.1000-Series 2-No. 15, Public Health Service. Washington. U.S.Government Printing Office, Mar. 1966.
1lNational Center for Health Statistics: Subtest estimates of
tie WISC Full Scale IQ’s for children. Vital and Health Statistics.
Series 2, No. 47. DHEW Pub. No. (HSM) 72-1047. Washington.
U.S. Government Printing Office, Mar. 1972.12Wechsler, D.: Wechsler Intelligence Scale for Childreq
ManuaL New York. The Psychological Corporation, 1949.13McCandless, B. R.: Adolescents: Behavior and Develop-
ment. Hlnsdale, Illinois. The Dryden Press Inc., 1970,
1%echsler, D... The Measurement and Appraisal of Adult
Intelligence (4th cd.). Baltimore, Williams and Wilkkss, 195S.15Mat~~~o, J. D.: Wechslerk Measurement and APPras3al of
Adult Intelligence (5th cd.). Baltimore. The Williams and WilkinsCorn any, 1972.
l%Firdey, C. J., and Thompson, J.: An abbreviated WechslerIntelligence Scsde for children for use with the educable mentallyretarded. Am.J.Ment.Deficiency 63:473480, 1958.
17Greenmun, R.: Abbreviated Forms of the Wechsler Intel-
Isgence Scale for Children. Unpublished master’s thesis. TexasChristian University, Aug. 1965.
18Simpson, W. H., and Bridges, C. C., Jr.: A short form ofthe Wecbsler Intelligence Scale for Children, J. Clin.Psychol.
k15(4 ‘424, Oct. 1959.
1 Wight, B., and Sandry, M.: A short form of the WechslerIntelligence Scrde for Chddren. J. Clin.PsychoL 18(2):166, Apr.
1962.
20Mumpower, D. L.: The fallacy of the short form.
J.Clin.PsychoL 20(1):11 l-l13,Jan. 1964.21 Silverstein, A. B.: Validity of WISC short forms at three
age levels. J. Consult.PWchoL 31:635-636, 1967.
22Enburg, R., Rowley, V. N., and Stone, B.: Short forms ofthe WISC for use with emotionally disturbed children. J. Clin.PsychoL 17(3) :280-284, July 1961.
23McKerracher, D. W., and Watson, R. A.: Validation of ashort form WfSC with clinic children. Bnt.J.Educ.PsychoL
38:205-208,1968.
24Nickols, J., and Nickels, M.:Brief forms of the WISC forresearch. J. Clin.PsychoL 19(4):425, Oct. 1963.
25Schwartz, L., and Levitt, E. E.: Short forms of theWechsler Intelligence Scale for Children in the educable, non-
institutiomlized mentaIly retarded. J.Educ.P~choL 51(4):187-
1904&?~mpson, J. M., and Firdey, C. J.: An abbreviated WISCfor use with gifted elementary school children. California
J.Educ.Res. 14(4):167-177, Sept. 1963.
27Yalowitz, J. M., and Armstrong, R. G.: Validity of shortforms of the Wechsler Intelligence Scale for Children. J. Clin.PsychoL 11 (3):275-277, July 1955.
28Silverstein, A. B.: Estimating Full Scale IQ’s from WISCshort forms. PsychoLRep. 20: 1264,June 1967, part 2.
18
29Silverstein. A. B.: A short form of Weehsler’s wales for
screening purpo~es. Psychol.Rep. 21:842, 1967.30Silverstein, A. B.: A short short form of the WISC and
WAIS for saeening purposes. Ps-ycholRep. 20:682, Apr. 1967.31 Silverstein, A. B.: Reappraisal of the v~ldity of a short
form of Wechsler’s scales. PsychoLRep. 26(2):559-561, Apr.1970.
32Silverstein, A. B.: Reappraisal of the wdklity of WAIS,WISC, imd WPPSI short forms. J. Consult. Wlin.PsychoL
J34(1 :12-14, Feb. 1970.
3 Hke, L.: Analysis of Reliability and Validity of theWechsler Intelligence Scale for Children. Unpublished doctoraldissertation. Western Reserve University, 1953.
34Hagen, E. P.: A Factor Analysis of the Wechsler Intel-ligence Scale for Children. Unpublished doctomf dissertation.Columbia University, 1952.
35Cohen, J.: The factoriaf structure of the WISC at ages 7-6,10.6, and 13-6. J. Con.m/t.Psychol. 23(4):285-299, Aug. 1959.
36Gault, U.: Factorial patterns of the Wechsler IntelligenceScales. Aust../.Psychol. 6(1):85-89, June 1954.
37Seashore, H., Wesman, A., and Doppelt, J.: The standard-ization of the Wechsler Intelligence Scale for Children.J, Consult.PsychoL 14(2) :99-1 10, Apr. 1950.
38Thorndike, IL L. (cd.): Educational Measurement, ed. 2.
Washington. American Councif on Education, 1971.39Anastasi, A.; Psychological Testing, ed. 3. New York. The
Macmillan Co., 1968.40Anastasi, A.: Individual Differences. New York. John
Wiley & Sons, Inc., 1965.41 Terman, L. M., and Merrill, M. A.: Measuring Intelligence.
Boston. Houghton M]fflin Co., 1937.42Sehaie, K. W., and Strother, C. R: A cross-sequential
study of age changes in cognitive behavior. PsychoLBull.
70(6 :671-680, Dec. 1968.!34 Tanner, J. M.: Growth at Adolescence, ed. 2. Oxford.
Blackwell Scientific Publications, 1962.44Nation~ Center for Heafth Statistics: Replication: An
approach to tie anzIYsis of data from complex surveys. vital andHealth Statistics. PHS Pub. No. 1000-Series 2-No. 14. PublicHealth Service. Washington. U.S. Government Printing Office,Apr. 1966.
45 Nationsd Center for Health Statistics Pseudoreplication:Further evaluation and application of the balanced half-sampletechnique. Vital and Health Statzktics. PHS Pub. No. 1000-Series2-No. 31. Public Health Service. Washington. U.S. GovernmentPrinting Office, Jan. 1969.
19
LIST OF DETAILED TABLES
Page
Table 1. Mean and standard deviation (SD) of raw scores on the Vocabulary subtast of the WISC, by sex and age and meanscores, by grade, sax, and age for noninstitutionalized youths: United Statas, 1966-70 . . . . . . . . . . . . . . .
2. Mean, standard deviation (SD), coefficient of variation (CV), median, and standard error of the mean (SE) of rawscores on tha Vocabulary and Block Dasign subtasts of the WISC for noninstitutionalized youths, by 4-month age
intervals: United Stetas,1966-70. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Mean and standard deviation (SD) of rawscores onthe Block Design sub@stof the WiSC, bysexend age, and meanscores, bygrade, sex, andagefor noninstitutionalized youths: Unitad States, 1966-70 , . . . . . . . . . . . . . .
4. Mean scaled scores on the Vocabulary subtast of the WISC for noninstitutionalized youths, by grade, sex, andage: UnitedStatas,1966-70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Mean scaled scores on the Block Design subtest of tha WISC for noninstitutionalizad youths, bygreda, sex, and
age: UnitadStates,1966-70. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. %ledscore equivalents forrawscores onthe Vocabulary submstof the WISCfor noninstitutionalized youths, by4-month ageintervals: United States, 1966-70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7. Scaled score equivalents forrawscores onthe Block Design sub@stof the WISCfor noninstitutionalizti youths, by4inonth ageintervals: United Statas,1966-70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8. Conversion table for sum of scaled scorason the Vocabulary end 810ck Design subtasts intoestimates of Full ScaleScorasamong noninstitutionalized youths: United Statas,1966-70 . . . . . . . . , . . . . . . , . . . . . . . . .
9. Mean estimated Full Scale scores on the Vocabulary and Block Design short form of the WISC fornoninstitutionalized youths, bygrade, sex, and age: Unitad Statas,1966-70 . . . . . . . . . . . , . . . . . . . . .
10. Smoothed percentile equivalents of raw scores on the Vocabulary and Block Design subtests of the WISC fornoninstitutionalized youths, byage andsex: United Statas,1966-70 . . . . . . . . . . . . . . . . . . . . . . . .
21
22
23
24
25
26
28
30
31
32
20
Tciblel. Mum and standard deviation (SD) of raw scoreson the Vocabulary subtestof the WISC. by sex and aqe, and mean scores, bygrade, sex, andage, for
noninstitutionalized youths: United States, 1966-70
Present made in school ITotal
SD
High Higher
LowerSpecial
school than
than 5th 6th 7th 8th 9th 10th llth 12th graduate 12thplacement
5th
Left
before
graduating
32.1
Scx and age
Mean
—
41.2—
36.5
38.8
40.9
42.2
44.4
45.2
42.2
Mean raw scoreBoth SC)(8S
12.17 years . . . 41.0 47.5—
. . .
. . .
. . ..
●
47.4
50.1
. . .
. . .
. . ..
. . .
50.1
46.2
11.24_
9.4410.76
10,87
11,20
11.27
11.39
11.26
18.9_
19.1●
.
.
●
. . .
20.5
25,0 32.1=
33.4
27.0
29.0.
. . .
. . .
33.8
37.0 39.7 44.0-
. . ..
46.3
44.9
42.7
37.2
45.1
47.0 49.0 54.1_
. . .
. . .
. . .
. . .●
54.1
54.6
. . .
. . .
. . .
. . .●
54.9
53.7
21.4. .
12years . . . . . . . . . .
13years . . . . . . . . . .
14yeara . . . . . . . . . .
15ymws. . . . . . . . . .
16years . . . . . . . . . .
17yk3ars . . . . . . . . . .
26.6
24.2.
“
.
. . .
25.0
38.2
36.8
31.8
28.8●
,
38.5
41.2
41.1
38.1
31.0
35.1.
41.2
●
42.8
43.4
3B.9
32.2
26.9
42.2
21.6
19.4
20.5.
●
●
22.2
. . ...
48.9
47.5
44.9
48.0
. . .
. . .
. . .
51.0
48.8
45.1
46.0
. . .
. . .●
●
49.6
48.9
50.2
. . .
. . .●
.
52.0
49.8
48.0
. . .
. . ...
32.2
33.0
33.6
. . .
. . .●
●
30.8
35.8
30.7
&
12-17 y0ars , . .
12yom.i . . . . . . . .13years . . . . . . . . . .
14years . . . . . . . . . .
15years . . . . . . . . . .
16yews . . . . . . . . . .
17 years . . . . . . . . . .
37.8
40.3
42.0
43.5
45.1
45.2
40.2
9.75
10.96
11.01
11.12
11.34
11.49
11,12
20,5.
●
●
. . .
25,4
25.o●
. . .
. . .
. . .
25.0
35.0
29.6
29.6●
. . .
. . .
29.9
39.9
38.7
33.2
31.3●
.
35.4
43.2
43.0
39.7
32.8
37.2●
38.1
.44.1
45.5
40.3
33.6
26.8
39.8
22.4
20.0
21.1●
●
●
19.6
. . .●
48.6
47.1
42.7
37.5
42.6
GA
12.17 y0ars . . . 17.3
12yei3rs . . . . . . . . . .
13vears . . . . . . . . . .14v0ws . . . . . . . . . .
15ycars . . . . . . . . . .
lGycars . . . . . . . . . .
17vcars . . . . . . . . . .
35.2
37.3
39.7
40.9
43.7
4.5,3
0.59
6.93
10.34
10.60
11.13
11.14
11.29
18.3.
●
-..
. . .
. . .
3.82
28.4.
. . .●
*
. . .
2.05
—
31.5
23.3●
●
. . .
. . .
1.01
36.6
34.4
29.6.
●
.
0.66
39.7
39.4
36.0
28.2●
. . .
0.54
20.1
18.0.
●
●
●
1.21
. . .41.7
41.7
36.9
29.8●
0.63
. . ..
44.3
43.0
42.8
36.5
0.64
. . ...
47.3
46.3
44.6
0.43
. . .
. . .
. . ..
47.5
48.1
0.53
. . .
. . .
. . ..●
46.0
1.91
. . .
. . .
. . .
. . ..
53.5
1.27
. . .
. . .●
●
33.4
29.9
0.84Stand,wd errar, total . .
21
Table2. Mean, standard deviation (SD), coefficient of variation (CV), median, andstandard error of themean (SE) ofraw scoreson
the Vocabulary and Block Design subtests of the WISC for noninstitutionalized youths, by 4-month age intervals: United States,1966-70
Age interval
12 years:0-3months . . . . . . . . . .4-7months . . . . . . . . . .8-n months . . . . . . . . .
13 years:0-3months . . . . . . . . . .4-7months . . . . . . . . . .8-n months . . . . . . . . .
14 years:O-3 months . . . . . . . . . .4-7 months . . . . . . . . . .8-n months . . . . . . . . .
15 years:O-3 months . . . . . . . . . .4-7 months . . . . . . . . . .8-n months . . . . . . . . .
16 years:O-3 months . . . . . . . . . .4-7 months . . . . . . . . . .8-n months . . . . . . . . .
17 years:O-3 months . . . . . . . . . .
4-7months . . . . . . . . . .8-n months . . . . . . . . .
-
n
318440432
414403390
414379409
370381365
404328360
336294327
N
1,0811,4661,456
1,3561,3231,271
1,3371,1701,334
1,2201,2871,244
1,4081,0291,188
1,2321,0611,214
Vocabulary subtest Block Design subtest
Mean SD Cv Median SE Mean SD Cv Median SE
Raw score
35.636.137.6
38.638.239.8
40.440.941.3
41.842.342.5
43.944.444.9
45.044.845.9
8.98
9.479.65
10.9810.6110.61
10.6411.1210.87
11.1211.3311.15
10.5511.9711.44
11.6310.9311.51
0.250.260.26
0.280.280.27
0.260.270.26
0.270.270.26
0.240.270.25
0.260.240.25
36,8
36.438.0
39.038.540.8
41.340.542.2
42.142.442.8
45.045.645.4
46.046.246.4
0.550.750.40
0.650.840.87
0.861.320.69
0.670.801.11
0.890.820.80
0.930.830.98
23.823.524.8
26.025.927.2
28.629.329.9
29.030.129.5
31.532.131.9
33.032.533.4
12.8812.6613.04
13.0213.7513.51
13.5113.9013.53
13.0314.0113.62
13.6613.7414.64
13.8613.6713.70
0.540.660.53
0.500.530.50
0.470.470.46
0.450.470.46
0.430.430.46
0.420.420.41
23.523.125.1
27.326.527.9
29.229.830.6
29.430.530.8
33.333.534.5
34.934.435.7
0.620.940.70
0.751.010,68
0.630.990.74
0.660.870.83
0.950.760.94
0.830.820.72
NOTE.–n = sample size; N = estimated number of youths in population in thousands.
22
Table 3. Mean and standard deviation (SD) of raw scores on the Stock Design subtest of the W ISC, by sex and age, and mean scores, by grade, sex, and age for
Total
ncminstituticmal ized youths: United States, 1966-70—
Present grade in school
Lower
than 5th 6th 7th 8th 9th 10th llth 12th
5th
High
school
graduate
HigherSpecial
Left
than before
12thplacement
graduatingSD
13.S8
%x and age
Mean
2S.9
Both sexes
12.17 years ., .
12yaors . . . . . . . . . .
13yews . . . . . . . . . .
14y00rs . . . . . . . . . .
15years . . . . . . . . . .
lG years . . . . . . . . . .
17yw3r s . .,,....,.
Boys—
12-17 years . . .
12yoars . . . . . . . . . .
13yews . . . . . . ...!
14years . . . . . . . . . .
15ycar5 . . . . . . . . . .
16ycars . . . . . . . . . .
17years . . . . . . . . . .
Girls—
12-17 yeara , , .
12yem3. . . . . . . . . .
13y0ms . . . . . . . . . .
14years . . . . . . . . . .
15wwrs. . . . . . . . . .
16ycws . . . . . . . . . .
17ycws . . . . . . . . . .
Standard error, total . .
Mean raw score
29.4_
.
31.0
31.4
27.1
23.2
19.s
31.0
.
32.1
33.8
29.4
24.5
19.6
27.7
. . .
30.0
29.4
23.9
20.8.
0.42
31.4.—
. . .●
34.5
31.9
29.4
29.7
33.7
. . .●
36.7
34.6
31.7
30.8
29.2
. . .●
32.6
28.5
27.0
2?.3
0.48
34.210.3 13.0 20.3 35.1 37.1—
. . .
. . .
. . ..
.
37.2
38.1
. . .
. . .
. . .,
. . .
37.7
36.6
. . .
. . .
. . ..
●
37.0
1.59
22.224.6-
25.7
24.4
19.2
18.7●
.
25.8
27.2
26.0
19.6
20.0●
●
23.3—
24.2
22.4
18.5●
●
●
0.63
27.4.
28.6
28.0
27.1
20.9
23.7.
29.3—
33.0
29.6
29.4
22.6
24.8●
25.4
25.2
26.6
24.1
18.3●
. . .
0.40
40.2_
. . .
. . .
. . .
. . .●
39.9
43.4
. . .
. . .
. . .
. . .●
42.8
37.5
. . .
. . .
. . .
. . ..
37.5
1.61
12.0_ . _ .
24.0
26.3
29.3
28.6
31.8
33.0
30.4
12.87
13.44
13.85
13.58
14.01
13.75
13,73
9.2.
●
,
*
. . .
14.5
21.0
17.5
20.4.
. . .
. . .
22.3.
22.8
20.4
24.4.
. . .
. . .
17.8—
18.9
13.6.
.
. . .
. . .
1.08
9.6
11.8
10.8●
●
●
13.8
14,5
10,9●
●
●
. . .
13.5—
14.8
10.5●
. . .
. . .
. . .
12.3—
14.0●
. . ..
.
. . .
1.72
. . ..●
33.2
35.0
33.0
36.2
. . .
. . .
. . .
33.6
37.8
34.1
32.3
. . .●
●
32.8
32.3
31.9
0.52
..-
. . ...
35.4
35.0
37.3
. . .
. . .●
●
39.4
36.8
33.1—
. . .
. . .
. . ..
31.8
33.4
0.46
..-
. . ...
21.4
22.8
24.7
25.6
27.7
30.8
31.3
34.2
33.9
27.2
12.82
13.32
13.86
13.21
13.44
13.64
13.84
13.7*
*
●
. . .
. . .
6.1
11.014.7
.
.
●
●
8.3
. . .
. . ..●
22.7
26.2
19.8
22,4
25.0
27.6
27.7
29.4
32.0
0.36
12.71
13.42
13.24
13.70
14.17
13.80
6.8.
*
. . ..
. . .
2.84
6.7
6.2.
.
●
.
0.91
. . .
. . .●
●
20.2
18.9
1.24
23
Table 4. Mean Scaled Scores on the Vocabulary subtest of the WISC for noninstitutionalized youths, by grade, sex, and age: United States, 1966-70
Sex and age
Present grade in school
Total Lowerthan 5th 6th 7th 8th 9th 10th llth 12th5th
Both sexes Mean Scaled Score
12-17 vears . . . . . 10.0 4,5 6.5 9.9 10.1 9.9 10.9 12.4_
. . .
. . .
. . .
. . .●
12.3
12.6
5.0 6.88.7 10.3 10.7 10.5. . . —12 years . . . . . . . . . . .13 years . . . . . . . . . . .14years . . . . . . . . . . .16 years . . . . . . . . . . .16 years . . . . . . . . . . .17years . . . . . . . . . . .
10.010.010.010.010.010.0
10.3
4.9●
●
●
●
-..
4.8
7.28.0
●
●
●
.-.
6.6
9.26.86.7
*
10.6
9.57.56.5
●
●
10.3
11.210.69.37.0
●
●
10.5
●
10.910.79.16.95.6
10.2
5.74.04.5
*●
*
5.2
. . .*
11.310.79.68.0
10.6
. . .●
●
11.810,89.9
11.0
. . .
. . .●
●
11.310.8
11.3
-... . .. . .
*●
10.4
11.2
. . .●
●
696.9
7.1
. . .
9.1
Boys—
12-17 years . . . . .
12yeara . . . . . . . . . . .13years . . . . . . . . . . .14years . . . . . . . . . . .16years . . . . . . . . . . .16years . . . . . . . . . . .17years . . . . . . . . . . .
10.410.410.310.410.210.0
9.7
5.5●
●
. . .
. . .
. . .
4.2
6.86.2
●
. . .
. . .
. . .
6.4
9.77.66.9
*
. . .-..
8.0—
B.65.9
●
.
. . .
. . .
0.30
11.010.07.97.2
●
●
9.5
10.18.86.9
●
●
*
0.19
—
11.811.29.77.5
●
●
9.7—
10.710.18.76.2
●
.-.
0.14
—
●
11.211.29.57.35.6
9.6
..-
10.610.28.66.2
.
0,17
—
5.94.84.6
*●
●
4.6
..-●
12.011.39.68.1
10.0
. . .
. . .
. . .12.411.2
9.9
10.4
.-.●
*
11.310.49.9
0.13
. . . .-.. . .. . .
●
✎ ✎ ✎
11.2
10.0
. . .
. . .-... . .
●
12.6
12.2
. . .
. . .
. . .
. . .●
12.1
0.33
. . .
. . ..*
6.57.6
6.5
●
☛
12.011.1
10.6
Girls—
12-17 vears , . . . .
12years . . . . . . . . . . .13years . . . . . . . . . . .14years . . . . . . . . . . .15years . . . . . . . . . . .16years . . . . . . . . . . .17vears . . . . . . . . . . .
9.69.69.69.69.8
10.0
0.16
4,6●
●
.-.
. . .
. . .
1.18
7.6●
. . .●
●
-..
0.62
5.44.2
●
●
●
●
0.37
-...
10.810.29.67.9
0.16
. . .
. . .. . .. . .
*●
7.36.1
0.21
. . .
. . .*
10.810.6
0.14
*.
10,0
0.53Standard error, total . . .
24
Table5, Mean Scalad Scorason tha Block Design subtest oftie WISCfor noninstitutionalized youths, bygrade, sex, and age: United States, 1966-70
IIHigherSpecial
Left
than before
12thplacement
graduating
Present grada in school
High
Lower school
than 5th 6th 7th 8th 9th 10th Ilth 12th graduate
5th
Sex and age Total
Both sexes Mean Scaled Score
10.2 10.5 10.5 10.6
. . . . . . . . . ---● ● --- ---
11.1 ● ‘ ---
10.4 10.7 ‘ ●
9.5 10.6 10.7 ●
9.4 10.0 10.5 10.9
10.7 11.0 11.0 11.0
-.. -.. .-. . . .
● ..- -.. . . .
11.6 --- ● ---
11.0 10.8 ●*
10.0 11.3 11.6 ---
9.7 10.3 10.B 11.0
9.7 10.0 10.1 10.7
. . . ..- --- . . .
● ● --- -..
10.6 ● --- ---
9.9 10.6 ‘ ●
9.0 10.0 9.9 .
B.8 9.8 10.1 10.8
0.10 0.12 0.11 0.31
8.012.17 years . . . . . 10.0 5.8 9.1.
9.4
8.2
B.2●
. . .
. . .
9.6—
9.8
8.8
9.0●
. . .
. . .
B.6—
8.9
7.3●
●
. . .
. . .
0.25
—
9.9 10.1 10.0 11.5_
. . .
. . .
. . .
. . ..
11.4
12.2
6.27.3.
IZyears . . . . . . . . . . .13 years.,.,...,,..
14 years . . . . . . . . . . .
16ycars . . . . . . . . . . .
16years . . . . . . . . . . .
17years, , ...,,,,..
10.010.010.010.010.010.0
10.4
5.8●
●
●
*
. . .
7.0
7.B
6.6.
●
●
. . .
7.5
10.4
9.6
7.9
7.4*
●
10.1
10.9
10.3
9.6
8.0
B.2●
10.4
6.6
5.9
5.1.
.
●
6.6
●
10.B10.5
9.4
8.3
7.5
10.3
. . .
..-,.
7.8
B.O
8.5
Boys—
12-17 years . . . . .
12years . . . . . . . . . . .
13years . . . . . . . . . . .
14years . . . . . . . . . . .
15years . . . . . . . . . . .
16years . . . . . . . . . . .
17yeers . . . . . . . . . . .
10.4
10.2
10.4
10.3
10.5
10.3
9.6
7.3●
●
●
. . .
. . .
4.6
8.0
6.6●
. . .
---
.-.
6.9
10.7
9.9
8.0
7.9●
●
9.6
11.9
10.7
10.1
8.5
8.4●
9.7
●
11.2
11.0
9.8
B.5
7.5
9.7
7.0
6.6●
*
●
●
5.3
. . .
. . .
. . .
.-.●
11.9
11.0
-... . .
●
✎
8.2
B.7
7.5
Girls—
12-17years . . . . .
5.6
4.6●
.
*
,
0.24
12years . . . . . . . . . . .13years . . . . . . . . . . .
14years . . . . . . . . . . .
15years . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . ..m..
9.6
9.6
9.7
9.5
9.4
9.8
0.08
5,1*
●
. . .●
. . .
1.06
7.6●
..-,
.
. . .
D.46
10.19.1
7.8.
.
●
0.14
10.1
10.0
9.0
7.3●
. . .
0.09
—
. . .10.5
10.1
B.7
7.9●
0.09
—
.-.
. . .
. . .
. . ..
11.0
0.41
.-.
. . .●
✎
7.5
7.3
0.32Standard error, total
25
Table6. Scaled Score equivalents forraw scoreson the Vocabulary subtestof the WISCfor noninstutionalized youths, by 4-month
Scaled Scores
0 . . . . . . . . . . . . . . . . . . . . . . .1 .......................2 . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . .6 . . . . . . . . . . . . . . . . . . . . . . .7 . . . . . . . . . . . . . . . . . . . . . . .8 . . . . . . . . . . . . . . . . . . . . . . .9 . . . . . . . . . . . . . . . . . . . . . . .lo . . . . . . . . . . . . . . . . . . . . . . .11 . . . . . . . . . . . . . . . . . . . . . . .
12 . . . . . . . . . . . . . . . . . . . . . . .13 . . . . . . . . . . . . . . . . . . . . . . .14 . . . . . . . . . . . . . . . . . . . . . . .15 . . . . . . . . . . . . . . . . . . . . . . .16 . . . . . . . . . . . . . . . . . . . . . . .
17 . . . . . . . . . . . . . . . . . . . . . . .18 . . . . . . . . . . . . . . . . . . . . . . .19 . . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . .
ageintervais: United States, 1966-70
I
Raw scoras
o-56-78-1o
11-1415-1819-2021-2425-2728-3132-3435-3839-40
41-4243-4546-4849-5051-54
55-5657-5960-6162-80
0-67-89-11
12-1415-1819-2122-2425-2728-3132-3435-3839-41
42-4445-4647-4950-5253-5657-59
60-6162-6364-80
0-67-89-11
12-1415-1819-2122-2425-2829-3334-3637-3940-41
42-4546-4849-5253-5556-5859-60
61-6263-6465-80
0-78-9
10-1112-1415-1819-2122-2425-2829-3334-3637-404143
44-4849-5051-5354-5657-59
60-6263-6465-6667-80
0-78-9
10-1213-1415-1819-2122-2526-2930-3334-3738-4041-43
44-4849-5051-5354-5657-60
61-6364-6566-6768-80
0-78-9
10-12
13-1415-1819-2223-2728-3132-3536-3940-4243-45
46-4849-5152-5455-5758-60
61-6364-6566-6768-80
14 years
=
O-89-1o
11-1314-1516-1819-2223-2728-3233-3536-39404243-45
46-4849-5253-5657-5960-61
62-6465-6667-6869-80
0-89-1o
11-1314-1516-1920-2223-2829-3233-3536-3940-4243-46
47-5152-5455-5758-5960-62
63-6566-6768-6970-80
0-89-1o
11-1314-1516-1920-2425-2829-3233-3536-4041-4344-4647-5152-5455-5758-6061-63
64-6566-6768-6970-80
26
Table6. Scaled Score equivalents forrawscoras on the Vocabulary subtest of the WISC for noninstitutionalized vouths, bv 4-monthageintetvals: United States, 1966-70–Con.
Scaled Scores
o . . . . . . . . . . . . . . . . . . . . . . .1 .......................2 ,,, ,. ..,. . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . ,,, ...,. . . . . . . .6 . . . . . . . . . . . . . . . . . . . . . . .7 . . . . . . . . . . . . . . . . . . . . . . .8 ... ,,... . . . . . . . . . . . . . . .9 . . . . . . . . . . . . . . . . . . . . . . .lo, . . . . . . . . . . . . . . . . . . . . . .11, . . . . . . . . . . . . . . . . . . . . . .12 . . . . . . . . . . . . . . . . . . . . . . .13 . . . . . . . . . . . . . . . . . . . . . . .14, . . . . . . . . . . . . . . . . . . . . . .15, . . . . . . . . . . . . . . . . . . . . . .16 . . . . . . . . . . . . . . . . . . . . . . .17 . . . . . . . . . . . . . . . . . . . . . . .18 . . . . . . . . . . . . . . . . . . . . . . .
19 . . . . . . . . . . . . . . . . . . . . . . .20 ., . ! . . . . . . . . . . . . . . . . . . .
15 yeers
=l=J=
o-910-1112-1314-1617-2122-2425-2829-3233-3536-4041-4344-4849-5253-5556-5859-6162-6364-6667-68
6970-80
0-910-1213-1516-1819-2122-2425-2829-3233-3637-4041-&l45-4849-5253-5657-5859-6162-6364-6667-6869-7071-80
0-910-1213-1516-1819-2122-25
26-2930-3233-3637-4041-4445-4849-5253-5657-5859-6162-6364-6667-6869-7071-80
-
Raw scores
0-1011-1213-1516-1819-2122-2627-3031-3435-3839-4344-46474950-5253-57
5859-6162-6465-6667-6869-7071-80
0-1o11-1213-1516-1819-2122-2627-3031-3435-3839-4344-4748-5152-5556-5859-6061-63
6465-5667-68
69-7071-80
0-1011-1213-1516-1819-2122-2627-3132-3435-3839-4344-4748-5152-5556-5859-6061-6364-6566-6768-69
7071-80
0-1o11-1213-1516-1819-2122-2627-3132-3435-3839-4344-4849-5152-5556-5859-60
61-6364-6566-6768-69707172-80
0-1o11-1213-1516-1819-2122-2627-3132-3536-4041-4445-4849-5152-5556-5859-6061-6364-6667-6869-70
7172-80
8-11months
o-1o11-1213-1516-1920-2223-2627-3132-3536-4142-4546-4849-5152-5566-5960-6162-6465-6667-6869-70
7172-80
27
Table7. %aled*ore equivalents forrawscores ontie Block Design subtest of the WISCfor noninstitutionalized youths, by4-monthage intervals: United States, 1966-70
Scaled Scores
o . . . . . . . . . . . . . . . . . . . . . . .1 ,....,...............#.2 . . . . . . . . . . . . . . . . . . . . . . .
3 . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . .6 . . . . . . . . . . . . . . . . . . . . . . .7 . . . . . . . . . . . . . . . . . . . . . . .8 . . . . . . . . ,. . . . . . . . . . . . . .9 . . . . . . . . . . . . . . . . . . . . . . .lo . . . . . . . . . . . . . . . . . . . . . . .11 . . . . . . . . . . . . . . . . . . . . . . .12 . . . . . . . . . . . . . . . . . . . . . . .13 . . . . . . . . . . . . . . . . . . . . . . .14 . . . . . . . . . . . . . . . . . . . . . . .15 . . . . . . . . . . . . . . . . . . . . . . .16 . . . . . . . . . . . . . . . . . . . . . . .17 . . . . . . . . . . . . . . . . . . . . . . .18 . . . . . . . . . . . . . . . . . . . . . . .19 . . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . .
12 years
=1=
o-12-3
45
6-89-1o
11-1314-2021-2526-3132-3738-4041-4243-4546-49
5051-52
5354-55
0-12-3
45
6-910
11-1314-2122-2526-3132-3738-4041-4344-4546-4950-51
5253
54-55
8-11months
o-12-3
456
7-910-11
12-1415-2223-2728-3233-3738-4142-4546-4748-5051-62
535455
13 years
o-3 4-7 I 8-11months months months
o-12-3
456
7-910-11
12-1819-2223-2930-3435-3639-4243-45464748-5051-52
535465
Rawscores
0-1
2-3
456
7-910-11
12-1819-2324-2930-3536-3940-44454748-49
5051-52
535455
0-1
2-3
45
6-78-10
1112-1920-2526-3031-3536-404144454748-5051-52
535455
0-3months
0-12-3
45
6-78-1o
11-1314-2122-2627-3132-3738-4142-4546-4748-5051-52
535455
14 years
4-7months
o-12-3
45
6-89-1o
11-1314-2122-2627-3233-3839-434446474849-51
52535456
8-11months
0-12.3
45
6-89-11
12-1415-2223-2728-3334-38394344-46474950-51
52535455
28
Table7, Scaled Score equivalents forrawwores onthe Bl~k Design subtest of the WISCfor noninstiWtionalized youths, by4-month
aga intervals: Unitad States, 1966-70-Con.
Sealed Scores
o . . . . . . . . . . . . . . . . . . . . . . .l ........................2 . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . .6 ., ...,.. . . . . . . . . . . . . . . .7 ,,, .,.,. ... .,.,, m.,.,.a . ..!.... . . . . . . . . . . . . . . .9 ,., .,.,. ... ,.,., . . . . . . .10 . . . . . . . . . . . . . . . . . . . . . . .11 . . . . . . . . . . . . . . . . . . . . . . .12 . . . . . . . . . . . . . . . . . . . . . . .13 . . . . . . . . . . . . . . . . . . . . . . .14 . . . . . . . . . . . . . . . . . . . . . . .15 ,, ..,,,, . . . . . . . . ,,,..,,16 . . . . . . . . . . . . . . . . . . . . . . .17, . . . . . . . . . . . . . . . . . . . . . .18 . . . . . . . . . . . . . .. . . . . . . . . .19, . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . .
15 yearsI
16 yearsI
17 years
o-3 4-7months months
o-12-3
45
6-910-1112-1415-2223-2728-3334-3839-4344-4647-4950-51
52535455
0-12-3
45
6-910-1112-1415-2223-2728-3334-3839-4344-4748-50
515253.5455
8-11months
o1
2-345
6-910-1112-1516-2223-2829-3334-3839-4344-4748-50
5152535455
0-3 4-.7 8-11months months months
o1
2-345
6-910-1112-1819-2425-3031-3637-404144454748-50
515253
54-55
Raw scores
o1
2-345
6-910-1112-19
20-2627-3132-3637-41424445-4748-50
515253
54-55
01
2-345
6-9l@ll12-1920-2627-3132-373841424546-4849-5051-52
535455
0-3 I 4-7months months
o1
2-345
6-910-1213-19
20-2627-3132-3738-424345464849-5051-52
535455
01-2
345
6-910-1213-20
21-2627-3132-37384243-45484849-5051-52
535455
8-11months
o-12345
6-910-1213-2021-2728-3334-3639424346474849-5051-52
535455
29
Table8. Conversion table for sum of Scaled Scoreson the Vocabulary and Block Design subtests into estimates of Full Scale Scores
among noninstitutionalized youths: United States,l B66-70
Sum of Scaled Scores
00 . . . . . . . . . . . . . . . . . . . . . . .!
01 . . . . . . . . . . . . . . . . . . . . . . . .02 . . . . . . . . . . . . . . . . . . . . . . . .
03 . . . . . . . . . . . . . . . . . . . . . . . .
04 . . . . . . . . . . . . . . . . . . . . . . . .
05 . . . . . . . . . . . . . . . . . . . . . . . .
m . . . . . . . . . . . . . . . . . . . . . . . .07 . . . . . . . . . . . . . . . . . . . . . . . .
08 . . . . . . . . . . . . . . . . . . . . . . . .
09 . . . . . . . . . . . . . . . . . . . . . . . .
lo . . . . . . . . . . . . . . . . . . . . . . . .11 ........................12 . . . . . . . . . . . . . . . . . . . . . . . .
13 . . . . . . . . . . . . . . . . . . . . . . . .
14 . . . . . . . . . . . . . . . . . . . . . . .,
15 . . . . . . . . . . . . . . . . . . . . . . . .
16 . . . . . . . . . . . . . . . . . . . . . . . .
17 . . . . . . . . . . . . . . . . . . . . . . .,
18 . . . . . . . . . . . . . . . . . . . . . . . ,
19 . . . . . . . . . . . . . . . . . . . . . . . .
20 . . . . . . . . . . . . . . . . . . . . . . . .
Short-form
estimate of
Full Scale Score
43
46
49
51
54
57
60
63
66
69
71
74
77
BO
83
86
89
91
94
97
100
Sum of Scaled Scores
21, . . . . . . . . . . . . . . . . . . . . . . .
22 . . . . . . . . . . . . . . . . . . . . . . . .
23 . . . . . . . . . . . . . . . . . . . . . . .,
24 . . . . . . . . . . . . . . . . . . . . . . . .
25 . . . . . . . . . . . . . . . . . . . . . . . .
26 . . . . . . . . . . . . . . . . . . . . . . . .
27 . . . . . . . . . . . . . . . . . . . . . . . .
2B . . . . . . . . . . . . . . . . . . . . . . . .
29 . . . . . . . . . . . . . . . . . . . . . . . .
30 . . . . . . . . . . . . . . . . . . . . . . . .
31 . . . . . . . . . . . . . . . . . . . . . . . .
32 . . . . . . . . . . . . . . . . . . . . . . . .
33 . . . . . . . . . . . . . . . . . . . . . . . .
34 . . . . . . . . . . . . . . . . . . . . . . . .
35 . . . . . . . . . . . . . . . . . . . . . . . .
36 . . . . . . . . . . . . . . . . . . . . . . . .
37 . . . . . . . . . . . . . . . . . . . . . . . .
38 . . . . . . . . . . . . . . . . . . . . . . . .
39 . . . . . . . . . . . . . . . . . . . . . . . .
40 . . . . . . . . . . . . . . . . . . . . . . . .
Short-form
estimate of
Full Scale Score
103
106
109
111
114
117
120
123
126
129
131
134
137
140
743
146
149
151
154
157
30
Tablc9. Mwm cstlmated Full Scale Scores on the Vocabulary and Block Design short form of the WISC for nmrinstituti.mralized youths, by grade, sex. and
age: United States, 1966.70
IHigh Higher
school than
graduate 12th
II Present grade in school
SpecialLeft
beforeplacement
graduating8th 9th 10th llth 12th
Mean estimated Full Scale ScweBoth S@XGS
12-17wws . . . . 100.5 99.8 101.4 103.4 04.2100.0_
100.1
99.9
100.1
99.7
100.0
100.0
101.9
72.4.
73.5●
●
●
.
. . .
76.9
79.6●
●
. . .
. . .
. . .
67.9
82.3 93.7 111.2-
. . .
. . .
. . .
. . ..
110.7
113.9
. . .
. . .
. . .
. . .●
113.2
108.9
85.199.4 103.7—
. . .
. . .
. . .●
.
103.7
1M.5
. . .
. . .
. . ..
. . .
106.2
102.4
74.9_ .12yenrs . . . . . . . . . .
13ycars . . . . . . . . . .
14yeiws . . . . . . . . . .
15 yews . . . . . . . . . .
16ycm’s . . . . . . . . . .
17ymrs . . . . . . . . . .
55.7
78.9●
●
●
. . .
83.3—
85.2
79.3●
. . .
. . .
. . .
31.0
!)6.1
B5.8
85.2.
. . .
. . .
96.4
102.7
97.4
87.0
82.8.
“
101.2
105.1
99.9
88.4
86.0.
.
97.4
1C6.O
102.8
97.0
85.7
86.6●
102.7
110.6
105.4
99.5
88.5
90.7.
98.2
.104.9
103.3
95.8
86.3
80.2
101.5—
●
I 08.7
106.3
98.2
88.0
80.4
98.1
78.0
73.1
70.0.
.
●
76.6
. . .●
106.9
103.2
97.6
92.4
103.7
. . ..
110.3
108.5
99.0
93.4
99.2
. . .●
●
107.1
104.2
99.8
105.5
. . .
. . .
. . .
109.0
107.1
100.6
101.5
. . .
. . ...
05.8
03.7
06.5—
. . .
. . .●
●
10.2
05.5
D2.1
..-
. . ...
85.0
85.5
87.5
~
12-17 years . . . .—
12ytws . . . . . . . . . .
13vwm . . . . . . . . . .
14yms . . . . . . . . . .
15ywws . . . . . . . . . .
16ywrrs . . . . . . . . . .
17ytws . . . . . . . . . .
102.3
102.0
101.9
101.9
102.2
100.7
96.0
98.6
89.6
88.0●
...
79.9
75.7
70.5*
,
+
71.3
. . .
. . .●
●
84.9
89.3
82.8
Girls—
12-17 yews . . . . 30.3
12yws . . . . . . . . . .
13ycm . . . . . . . . . .
14y0:rs . . . . . . . . . .15yhirs . . . . . . . . . .
lGywirs . . . . . . . . . .
17vtws . . . . . . . . . .
97.8
97.7
98.197.5
97.9
99.4
0.65
70.2●
.
...
.-.
...
6.00
B6.4●
....
.
...
3.00
32.9
30.4●
.
...
...
1.47
100.4
94.2
84.8.
.
.
0.91
—
102.5
100.3
93.581.5
.
...
0.55
74.5
67.8.
.
.
.
1.76
. . .103.4
100.8
92.4
83.3.
0.71
. . ..
104.0100.2
96.2
90.3
0.66
. . ...
105.6
101.3
99.1
0.63
. . .
. . .
. . ..
01.9
32.1
0.62
. . .
. . .
. . ...
102.4
2.17
. . .
. . .
. . .
. . ..
108.6
1.88
. . .
. . ...
85.1
81.1
1.17St.mrJwd urror, total . .
31
Table IO. Smoothed percentile equivalents of raw scores on tha Vocabulary and Block Design subtests of the WISC fornoninstitutionalized youths, byage and sex: United States, 1866-70
Sex and percentile
Both sexes
88 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
75 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70 . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . ..$65 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Boys
99 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 . . . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . . . . . .50 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
45 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Age in years
12 13 14 15 16 17
Vocabulary raw score
585554
5352484644
43424140383736343332302826242019181614
59
5756555450484644
43424i403B
37
36
34333129
615957
5655535048
47
454342413938373533323027
242019181714
62
6059585754525049
4745434241
393836343331
646160595B555250
484745444341403837363432302722201917
14
646260605956545250
4847454442
4139
38373534
656362
6160575453
51
494745444241393836343330
272423211916
646362616058565453
5049474544
4241
39373534
666463
6261595654
53514948474544424039373532
2925242220
17
6665646362595856
54
5251494746
4443
41393835
676665
6463595755
53525049484645434240373533
302524222017
67
6665646359585654
5251494846
444341403835
12 13 14 15 16 17
Block Design raw score
514947
4645423937
34323028262422201713121110
665544
524948474643413936
3332302826
242220
171312
515049
4847454240
37353331292725232017131210
96655
4
5250494948464341
38
3734323029
272522191412
535251
5050474543
41393734323028262422181411
106665
4
545252515047464443
4740383533
302825232016
535252
5150474543
41
39373533302826242220141210
6665
4
545252525149464443
4140383533
302927252219
545353
5252494745
43414038363432302825221812106665
4
54
54535353504947
45
4342403936
343230282522
545353
5252504846
44434139373533312926232015
1166665
5454535353514947
45
4342413937
3532
30282522
32
Table lO. Smoothed percentile equivalents of raw scores on the Vocabulary and Block Design subtests of the WISC fornoninstitutionalized youths, byage and sex: United States, 1866-70-Con.
Sex and percentile
Boys–Con.
15 ., . ., .,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 . . . . . . . . . .,, !..... . . . . . . . . . . . . . . . . . . . . . .5 .,, ,,... .,, ,..., . . . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 ... ..,,, ,,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 ., ..,,,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Girls
99, , .,..,,,.,,,,.,,,..,,. . . . . . . . . . . . . . . . . .98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..?...88 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
45 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 .,,,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 ,. ..,,,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
—12
—
27
2521201917
14
5453515049464443424139383736
35333231292825231918171614—
Age in years
EIEIzIIVocabulary raw score
29 32 32 3326 28 29 2921 22 25 2520 20 24 2419 19 23 2318 18 21 2114 14 17 17
5856555453504846454342413938
37363433312826231919171614—
6360585756535048474644434241
39373634333128252120181614
—
6663616058555351494745444341
39383635333229252321191814
—
6663626161575553515048474645
43424039363432282423201916
—
—17
—
333026242321
17
6766646362595755535251494846
45444241383533302423221916
—
12 13 14 15 16 17
Block Design raw score
11 12 12 13 18 1810 10 10 12 12 1208 m 06 10 10 1008 06 (X 06 w 08m 05 w 06 05 0505 05 06 05 08 0804 04 04 04 05 06
5047464544403834
323028262422
21171312111008IX05050504
w—
5049484847444138
363431292725
23211813121110IX0605050504—
5351505049
464341
383634323028
2625222016121110M05050504—
53
52505049464341403835333028
2625222016131110M05050504—
5353525150464443413937363331
2927242218 ~131110 I
1
0605050504
54
53525252
494746444139373534
3230282523181311mw060605—
33
APPENDIX I
TECHNICAL NOTES
The Survey Design
The sample designs for the first three pro-grams, or Cycles I-III, of the Health Examina-tion Survey have been essentially similar, in thateach has been a multistage, stratified probabilitysample of clusters of households in land-basedsegments. The successive elements for this sam-ple design are primary sampling unit (PSU),census enumeration district (ED), segment (acluster of households), household, eligibleyouths, and finally, the sample youth.
The 40 sample areas and the segments utilizedin the design of Cycle III were the same as thosein Cycle II. Previous reports describe in detailthe sample design used for Cycle II and inaddition discuss the problems and considerationsgiven to other types of sampling frames andwhether or not to control the selection ofsiblings.6~7
Requirements and limitations placed on thedesign for Cycle III, similar to those for thedesign for Cycle II, were that:
1.
2.
3.
24
The target population be defined as thecivilian noninstitutional population of theUnited States, including Alaska andHawaii, in the age range of 12 through 17years, with the special exclusion of chil-dren residing on reservation lands of theAmerican Indians. The latter exclusion wasadopted as a result of operational prob-lems encountered on these lands in CycleI.
The time period of data collection belimited to about 3 years for each cycle andthe length of the individual examinationwithin the specially constructed mobileexamination center be between 2 and 3hours.
Ancillary data be collecteddesigned household, medical
e
on speciallyhistory, and
4.
5.
school questiormaires, and from copies ofbirth certificates.
Examination objectives be related primar-ily to factors of physical and intellectualgrowth and development.
The sample be sufficiently lame to yield. .- .reliable findings within broad geographicregions and population density groups aswell as within age, sex, and limited socio-economic groups for the total sample.
The sample was drawn jointly with the US.Bureau of the Census, beginning with the 1960Decennial Census list of addresses and the nearly1,900 PSU’S into which the entire United Stateswas divided. Each PSU “is either a standardmetropolitan statistical area (SMSA), a county,or a group of two or three contiguous counties.These PSU’S were grouped into 40 strata, witheach stratum having an average size of about 4.5million persons. Stratification was accomplishedso as to maximize the degree of homogeneitywithin strata with regard to the population sizeof the PSU’S, degTee of urbanization, geographicproximity, and degree of industrialization. The40 strata were then classified into four broadgeographic regions of 10 strata each and thenwithin each region, cross-classified by four popu-lation density classes and classes of rate ofpopulation change from 1950 to 1960. Using amodified Goodman-Kish controlled-selectiontechnique, one PSU was drawn from each of the40 strata.
Generally, within each PSU, 20 census enu-meration districts were selected, with the proba-bility of selection of a particular ED propor-tional to its population in the age group 5-9years in the 1960 Census, which by 1966approximated the target population for CycleIII. A similar method was used for selecting onesegment (a smaller cluster of households) in each
ED. Because of the approximately 3-year time
interval between Cycle II and Cycle III, theCycle III sampling frame was updated for newconstruction and to compensate for segmentswhere housing was partially or totally demol-ished to make room for highway construction orurban redevelopment. Each of the resulting 20segments within a PSU was either a boundedarea or a cluster of households (or addresses).All youths in the appropriate age range whoresided at the address visited were eligible
youths, i.e., eligible for inclusion in the sample.Operational considerations made it necessary toreduce the number of prospective examinees atany one location to a maximum of 200. Whenthe number of eligible youths in a particularlocation exceeded this number, the “excess”eligible youths were deleted from the samplethrough a systematic sampling technique.Youths who were not selected as sample youthsin the Cyclp III sample, but who were previouslyexamined in Cycle II, were scheduled for exam-ination if time permitted and will be included inspecial longitudinal analyses. In addition, indi-vidual twins who were deleted from the CycleIII sample were also scheduled for examination,as they were in Cycle II, to provide data on pairsof twins for future analysis. These data are notincluded in the report as part of the nationalprobability sample of youths.
The sample was selected in Cycle III, as it hadbeen for the children in Cycle II, to containproportional representation of youths fromfamilies having only one eligible youth, twoeligible youths, and so on, so as to be representa-tive of the total target population. However,since households were one of the elements in thesample frame, the number of related youths inthe resulting sample is greater than would resultfrom a design which sampled youths 12-17 yearswithout regard to household. The resultingestimated mean measurements or rates should beunbiased but their sampling variabilities will besomewhat greater than those from a morecostly, time-consuming, systematic sample de-sibm in which every kth youth would be selected.
The total probability sample for Cycle IIIincluded 7,514 youths representative of theapproximately 22.7 million noninstitutionalizedUnited States youths of 12-17 years. The samplecontained youths from 25 different States, withapproximately 1,000 in each single year of age.
The response rate in Cycle III was 90 petcent,
with 6,768 youths examined out of the totalsample. These examinees were closely represen-tative of those in the population from which thesample was drawn with respect to age, sex, race,region, population density, and populationgrowth in area of residence. Hence it appearsunlikely that nonresponse could bias the find-ings appreciably.
Reliability
While measurement processes in the surveyswere carefully standardized and closely con-trolled, the correspondence between true popu-lation figures and HES results cannot be ex-pected to be exact. Survey data are imperfectfor three major reasons: (1) results are subject tosampling error, (2) the actual conduct of asurvey never agrees perfectly with the design,and (3) the measurement processes themselvesare inexact, even though standardized and con-trolled.
Data recorded for each sample youth areinflated in the estimation process to characterizethe larger universe of which the sample youthsare representative. The weights used in thisinflation process are a product of the reciprocalof the probability of selecting the youth, anadjustment for nonresponse cases, and a post-stratified ratio adjustment that increases preci-sion by bringing survey results into closeralinement with known U.S. population figuresby color and sex within single years of age forages 12-17.
In the third cycle of the Health ExaminationSurvey, as for the children in Cycle II, thesample was the result of three principal stages ofselection: the single PSU from each stratum, the20 segments from each sample PSU, and thesample youth from the eligible youths. Theprobability of selecting an individual youth isthe product of the probability of selection ateach stage.
Because the strata are roughly equal in popu-lation size and a nearly equal number of sampleyouths were examined in each of the samplePSU’S, the sample design is essentially self-weighting with respect to the target population;that is, each youth 12 through 17 years of agehad about the same probability of being drawninto the sample.
35
The adjustment upward for nonresponse isintended to minimize the impact of nonresp onseon final estimates by imputing to nonre-spondents the characteristics of “similar”respondents. Here, “similar” respondents in asample PSU were defined as examined youths ofthe same age in years and sex as youths notexamined in that sample PSU.
The poststratified ratio adjustment used inthe third cycle achieved most of the gains inprecision that would have been attained if thesample had been drawn from a populationstratified by age, color, and sex and makes thefinal sample estimates of population agreeexactly with independent controls prepared bythe Bureau of the Census for the U.S. noninsti-tutional population as of March 9, 1968, ap~proximate midpoint of the survey for Cycle III,by color and sex for each single year of ages12-17. The weight of every responding sampleyouth in each of the 24 age, color, and sexclasses is adjusted upward or downward so thatthe weighted total within the class equals theindependent population control. Final samplefrequencies and estimated population frequen-cies as of the approximate midpoint of thesurvey are presented in table I by age and sex.
Extent of Missing Test Results andImputation Procedures Used
In addition to youths who were selected forthe sample but, for various reasons, not exam-
ined, there were some whose examination wasincomplete in one procedure or another. Theextent of missing data for the WISC is shown intable 11 according to sex and age. For 109youths, or 1.6 percent of all those examined,one or both of the WISC subtest results were notavailable. There were a number of reasons forthis, the primary one being operational andlogistical survey problems, such as lost recordsor lack of time to complete the examination.Other reasons included difficulties with theEnglish language or illness on the part of theyouth. Since the reason for missing test resultsin most cases was not directly related to thecharacteristic being measured, raw scores wereimputed for almost all of these examinees. Incertain, infrequent instances, imputation wasnot considered appropriate, as for example, theimputation of Block Design scores for a blindyouth, or the imputation of Vocabulary testresults for foreign-language-speaking youths whocould not understand English well enough totake any of the psychological tests.
Imputation was accomplished in the followingmanner. An intercorrelation matrix of allpsychological test data and selected socio-economic variables was derived to identify thosevariables that were most highly associated witheach raw test score. As a result, five variableswere chosen for the imputation of Vocabularyand Block Design raw scores: other available testscores, educational level of the head of thehousehold (four categories), age, and two con-
Table 1. Sample and estimated population frequency distributions of youths 12-17 years of age in the noninstitutional population of
the Unitad States: Health Examination Survey, 1966-70
I I
Number of youths in sampleEstimated number of youths
Agein population as of midsurvey
Total Boys Girls TotalII
Boys Girls
Number in thousands
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,768 3,545 3,223 22,692 11,489 11,203
12years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,190 643 547 4,002 2,032 1,970
13years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,208 626 582 3,952 2,006 1,846
14 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,2C4 618 588 3,852 1,951 1,90115 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,116 613 503 3,751 1,900 1,851
16 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,082 556 536 3,625 1,836 1,789
17 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 958 489 469 3,510 1,764 1,746
36
Table 11. Extent of missing WISC test results among examinees in the Health Examination Survey, 1966-70
Sex and age
Both sexes
12-17 years, ., . .,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12 yeers ... .,,.,, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 years, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14yeers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 years, : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
~
12-17 years, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13years a m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Girls
12-17 years, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13years . .. m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14years, .. o . . . .. o..,....,.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17years a m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
trol variables, race and sex. Imputation of amissing test result for an examinee was accom-plished by randomly selecting a match amongthe group of examiners ofthesame agein years,parental level of education (four categories),race, sex, and available raw score test resultsmost highly correlated with the scores to beimputed. The raw score of this “matched”examinee was then imputed to the examineewith the missing score. When data for any ofthese variables were not available, amatch wasselected using information on as many of thevariables as were available in the youth’s record.
Totel
108-
1616241518IB
60
9101787
9
49
9677
119
Bothsubtestsmissing
I
Vocabulary 810ck Designonly only
missing missing
Number of examinees
71
128
15111114
40
75
12835
31
533389
17
334331
7
112
2
1
10
22231
21
355143
13
143
2
3
8
21212
Sampling and Measurement Error
In the present report, reference has beenmade to efforts to minimize bias andvariabilityof measurement techniques. The probabilitydesign of thesurvey makes possible thecalcula-tion of sampling errors. The sampling error isused hereto determine howimprecise the surveytest results may be because they result from asample rather than tkom the measurements ofallGlements in the universe. The estimation ofsampling errors for a study of the type of theJ+IealthExamination Survey is difficult for at
37
least three reasons: (1) measurement error and${ ~reM sampling error are confounded in thed%a, and it is difficult to find a procedure thatwill either completely include both or treat oneor the other separately; (2) the survey designand estimation procedure are complex, andaccordingly, require computationa.lly involvedtechniques for the calculation of variances; and(3) thousands of statistics are derived from thesurvey, many for subclasses of the populationfor which the number of sample cases is small.Estimates of sampling error are obtained fromthe sample data and are themselves subject tosampling error, which may be large when thenumber of cases in a cell is small or, occasion-ally, even when the number of cases is substan-tial.
Estimates of approximate sampling variabilityfor selected statistics used in this report areincluded in the detailed tables and in tables IIIthrough VII. These estimates, called standarderrors, have been prepared by a replicationtechnique that yields overall variability throughobservation of variability among random sub-samples of the total sample. The method reflectsboth “pure” sampling variance and a part of themeasurement variance, and is described in pre-viously published reports.44~45
Hypothesis Testing
In accordance with usual practice, the intervalestimate for any statistic was considered to be
Table 111. Standard errors for mean raw scores on the Vocabulary subtest of the W ISC by grade, sex, and age for non institutionalized youths: United States,
1966-70
Present grade in school
Total Lower
than 5th 6th 7th 8th 9th 10th Ilth 12th
5th
Sex and age
Both sexes
12-17 years . . . . . .
12 years . . . . . . . . . . . .
13years . . . . . . . . . . . .
14years . . . . . . . . . . . .
16years . . . . . . . . . . . .
16 years . . . . . . . . . . . .17 years . . . . . . . . . . . .
BOYS—
12-17 years . . . . . .
12 years . . . . . . . . . . . .
13yaars . . . . . . . . . . . .
14yaars . . . . . . . . . . . .
15years . . . . . . . . . . . .
16 years . . . . . . . . . . . .
17 years . . . . . . . . . . . .
Girls—
12-17 years . . . .
12years . . . . . . . . . . . .
13years . . . . . . . . . . . .
14 years . . . . . . . . . . . .
15 years . . . . . . . . . . . .
16 years . . . . . . . . . . . .
17 years . . . . . . . . . . . .
Sample frequencies . . . . .
Standard error
0.59 2.05.
2.00
4.07*
*
●
. . .
1.16—
1.17
3.73●
. . .
. . .
.-.
4.09—
3.51*
. . .●
●
. .
74
—
0.54 0.64-
-..●
1.09
0.64
1.3B1.18
0.65
0.53 1.21 0.s43.82-
5.06*
●
●
●
. . .
5.11
7.64●
●
*
. . .
. . .
4.65
1.01.
0.84
1.82
2.86●
.-.
. . .
1.01
1.07
1.71
2.37*
. . .
. . .
1.19
0.66 0.63.
●
1.00
0.81
; .05
1.22
2.36
0.78
●
0.B7
0.74
1.25
1.38
1.83
0.68
0.43.
. . .●
●
1.34
0.331.08
0.55
1.91_
. . .
. . .
. . .*
●
1.98
2.76
1.27_
. . .
. . .
. . .
. . .●
1.44
1,81
_ . —0.50
0.62
0.81
0.67
0.570.69
0.64
0.66
0.72
0.93
0.74
0.75
0.79
0.55
0.36
1.58
1.32
2.77●
.
0.85
0.64
1.89
1.52
2.62●
●
0.52
0.92
0.45
1.32
2.53
2.16.
0.58
1,24
0.62
1.32
2.04
3.11.
0.66
1,76
2.77
2.76●
●
.
1.99
. . .
. . .●
●
1.600.65
1.18
.-.
. . .●
✎
1.180.51
0.67
2.11
4.60
3.74●
.
●
1.40
. . .●
1.46
0.72
1.47
1.89
0.79
. . .
. . .
. . .1.92
0.52
1.40
0.46
. . ..
●
1.21
0.50
1.10
876
. . .-..
●
●
1.61
0.58
0.75
. . .
. . .
. . .●
✎ ✎ ✎
2.96
2.62
. . .-... . ...-
.2.05
1.74
. . .
. . .●
●
2.2s
1.43
0.99
0.53
0.71
0.80
0.71
0.62
0.75
6,760
6,87●
●
. . .
. . .
. . .
25
1.07
2.19.
●
. . .
. . .
390
0.40
1.49
2.13●
●
●
1,056
1.13
0.52
1.54
4.26●
. . .
1,261
3.23
5.25●
●
●
.
88
. . .1.58
0.68
1.44
1.44●
1,109
. . .●
1.66
0.84
1.70
1.86
1,072
. . .
. . .-..
.1.83
0.68
577
. . .
. . .
. . .*.
2.77
55
. . .
. . .
. . .
. . ..
1.80
53
. . .
. . .●
●
2.24
1.54
145
38
Table IV. Standard errors for mean raw scores on the Block Design sub test of the W KC by grade, sex, and age for non institutionalized youths: United States,
1966-70—
Present grade in school
High Higher
school than
graduate 12th
Left
before
graduating
Lower
than 5th 6th 7th Bth 9th 10th llth 12th
5th
Special
placementSex and age Total
Both sexes
12.17 years . . . . . .
12yeare . . . . . . . . . . . .
13 years . . . . . . . . . . . .
14years . . . . . . . . . . . .
15years . . . . . . . . . . . .
16 years . . . . . . . . . . . .
17yl?ars, . . . . . . . . . . .
Boys—
12-17 years . . . . . .
12years . . . . . . . . . . . .
13years ., . .,.,.,...
14 years . . . . . . . . . . . .
16years . . . . . . . . . . . .
lfiyears . . . . . . . . . . . .
17years, , ..,.,..,..
Girls—
12.17 year5 . . . . . .
12years, . . . . . . . . . . .
13 years...,,,,,,,,,
14years . . . . . . . . . . . .
15years . . . . . . . . . . . .
16yeara . . . . . . . . . . . .
17 yeare . . . . . . . . . . . .
Sample frequencies . . . . .
Standard error
0.36 0.402.84=
3.33●
.
●
●
. . .
4.99
1.72 1.08-
1.04
2.28
3.76.
. . .
. . .
1.lB
0.63 0.42 0.4B 0.52 0.46 1.59—
. . .
. . .
. . ..
.
1.73
3.11
1.61_
. . .
. . .
. . .
. . ..
1.55
1.57
0.91 1.24—0.41
0.54
0.56
0.51
0.56
0.43
0.3B
1.5B
4.05●
●
●
. . .
1.51
0.59
1.31
1.74
2.80●
●
0.76
1.480.46
0.73
2.41
2.78●
0.51
.1.09
0.48
0.86
1.43
3.03
0.54
1.19
2.B9
4.00.
.
.
1.49
. . .●
1.31
0.62
1.14
1.73
0.6B
..-●
✎
1.25
0.67
1.09
0.44
. . .
. . ..●
1.37
0.52
0.78
. . .
. . .●
●
2.01
1.42
2.52
0.56
0.62
0.70
0.72
0.73
0.51
0.37
10.16●
●
●
. . .
. . .
1.88
1.40
3.26●
. . .
. . .
. . .
2.B5
1.12
2.43
5.07*
. . .
. . .
1.t8
0.71
1.61
2.31
2.s1●
.
0.66
1.13
0.63
1.02
2.08
3.17●
0.51
●
1.25
0.71
1.08
2.19
3.08
0.66
1.57
4.35.
●
.
●
1.49
. . ..
1.16
0.93
1.22
1.55
0.67
. . .
. . .
. . .2.01
0.65
1.50
0.77
. . .
. . ...
1.54
O.BO
0.65
. . .
. . .
. . .●
✎ ✎ ✎
3.17
3.01
. . .
. . .
. . .
. . ..
1.58
3.25
. . .
. . .
. . .
. . ..
3.34
53
. . .
. . ..●
2.B4
3.08
1.3B
0.59
0,65
0.70
0.53
0.76
0.56
6,764
2.81●
●
. . ..
. . .
25
2.9B.
. . .●
.
. . .
74
1.21
2.30*
●
. . .
. . .
390
0.81
1.29
2.35●
●
●
1,067
1.97
0.67
0.95
4.09.
. . .
1,261
2.09
2.83●
.
.
.
67
. . .1.77
0.71
1.26
2.03.
1,112
. . .●
2.14
O.BO
1.55
4.15
1,071
. . .●
●
1.66
1.04
1.25
675
. . .
. . .
. . .*
1.74
0.81
577
—
. . .
. . .
. . .●
✎
3.28
55
. . .
. . .●
●
3.07
2.22
147
the range within one standard error of the the difference as computed above. If the result-ing t value was *2.00 or more, the differencewas considered statistically significant at approx-imately the 5-percent confidence level. Forexample, the mean Block Design raw score for12-year-old boys was 25.6, while the mean for12-year-old girls was 22.4, a difference of 3.2.The approximate standard error of the differ-ence between means was 0.81. Since the differ-ence between the means was about four timesthe standard error, the difference was consideredsignificant beyond the 5-percent confidencelevel.
tabulated statistic with 68 percent confidence,and the range within two standard errors of thetabulated statistic with 95 percent confidence.The latter is used as the level of statisticalsignificance in this report.
An approximation of the standard error of adifference d = x - y of two statistics x and y isgiven by the formula Sd = (S: + S; )% where S.and SY are the sampling errors, respectively, of xand y. Of course, where the two groups ormeasures are positively or negatively correlated,this formula will give an overestimate or under-estimate of the actual standard error.
Thus, the procedure used in this report fortesting the significance of difference betweenmeans consisted in dividing the difference be-tween the two means by the standard error of
Small Categories
In some tables, statistics may be shown forcells for which the sample size is so small that
39
Table V. Standard errors for mean scaled scores on the Vocabulary subtest of the WISC by grade, sex, and age for noninstitutionalized youths: United
States, 1966-70
Present grade in school
Lower
than 5th 6th 7th 8th 9th 10th Ilth 12th
5th
I I I
High HigherSpecial
Left
school than before
graduate 12thplacement
graduating
Sex and age Total
Both sexes
12-17 years . . . . .
12year5 . . . . . . . . . . .
13 years . . . . . . . . . . .
14 years . . . . . . . . . . .
15year$ . . . . . . . . . . .
16 years . . . . . . . . . . .
17years . . . . . . . . . . .
B&l&
12-17 years . . . . .
12yeara . . . . . . . . . . .
13years . . . . . . . . . . .
14years . . . . . . . . . . .
15years . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . . . . . .
Girls—
12-17years . . . . .
12years . . . . . . . . . . .
13years . . . . . . . . . . .
14years . . . . . . . . . . .
15years . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . . . . . .
Standard error
0.17=
.
0.26
0.16
0.27
0.30
0.60
0.22—
●
0.24
0.20
0.34
0.34
0.50
0.18—
. . .
0.42
0.19
0.36
0.35●
—
0.16_
. . .●
0.31
0.16
0.38
0.27
0.18
. . .●
0.42
0.19
0,41
0.44
0.20
. . ..
0.47
0.22
0.48
0.50
0.130.16 0.30 0.19 0.14 0.14 0,53 0.33 0.37 0.211.18 0.62
0.14
0.16
0.22
0.17
0.16
0.18
0.17
1.70●
,
*
●
. . .
1.47
0.59
1.10●
●
●
. . .
0.35
0.26
0.52
0.73●
..-
-..
0.30
0.110.43
0.35
0.73●
●
0.24
0.28
0.11
0.36
0.68
0.49●
0.16
0.52
0.77
0.72●
●
*
0.68
. . .●
●
0.35
0,09
0.28
0.16
. . .
..-*●
0.31
0.14
0.18
. . .
. . .
. . .●
●
0.53
0.86
. . .
. . .
. . .
. . .●
0.38
0.50
. . .
. . .●
●
0.38
0.15
0.30
. . .
. . .●
●
0.59
0.32
0.23
0.20
0.19
0.25
0.20
0.21
0.20
0.15
2.35●
●
. . .
. . .
. . .
1.59
0.35
1.03,
..-
. . .
. . .
1.26
0.33
0.48
0.52●
. . .
. . .
0.35
0.19
0.52
0.41
0.68●
●
0.16
0.40
0.16
0.36
0.53
0.72+
0.18
0.62
1.37
0.95*
●
*
0.37
. . .
. . .-..
0.52
0.18
0.36
0.14
.-.
. . .●
●
0.440.16
0.21
. . .
. . .
. . .●
✎ ✎ ✎
0.92
0.71
. . .
. . .
. . .
. . .*
0.55
0.47
0.15
0.18
0.21
0.18
0.17
0.21
2.36●
●
. . .
. . .
.-.
1.02.
. . .
●
●
. . .
0.31
0.61.
●
. . .
. . .
0.12
0.41
0.56●
●
●
0.35
0.13
0.41
1.20*
. . .
0.91
1.34●
*
*
*
. . ..●
0.30
0.14
0.29
. . .
. . .
. . .●
0.48
0.21
. . .
. . .
. . .●
✌
0.75
. . .
. . .
. . .
. . .●
0.47
. . .
. . .**
0.52
0.35
the sampling error may be larger than the administration of the psychological test batterystatistic itself. Such statistics areincluded in this to insure the quality of the data. AU psycholo-
gists were initially trained in the survey proce-dures by the Psychological Advisor. In addition,the field psychologists exchanged all test formsdaily and checked for any apparent errors in testadministration, counting, scoring, or recording.Each field psychologist tape recorded one entiretesting session each week. The tapes were sent tothe supervisor who reviewed them and madenotes of errors and suggestions regarding testingprocedures. These notes were then sent to thefield psychologists. Periodically the supervisorvisited the mobile examination center for directobservation and supervision of the work. Thetest forms were also spot checked when theyarrived at headquarters. Approximately onceeach week, six WISC subtests, chosen randomly
report, along with their corresponding standarderrors, in the belief that the information, whilenot meeting strict standaxds of precision, maylend an overall impression of the survey find-ings and may be of interest to subject-matterspecialists.
Quality Control for the PsychologicalTest Battery
The maintenance of standardized administra-tion and scoring procedures and methods ofrecording results is essential to large data collec-tion operations such as the Health ExaminationSurvey, as it is to psychological testing itself.Several procedures were incorporated into the
40
Teblo V1. Standard errors for mean scaled scores on the Block Design subtest of the WISC by grade, sex, and age for noninstitutionalized youths: UnitedStates, 1966-70
Present grada in schoolHigh Higher Left
Totel LowarSpecial
school than beforethan 5th 6th 7th 8th 9th 10th Ilth 12th graduate 12th
placement
5thgraduating
Sex and age
Both sexes
12.17 years . . . . .
12y00rs . . . . . . . . . . .13yews . . . . . . . . . . .14yows . . . . . . . . . . .15y0ars . . . . . . . . . . .16 years . . . ...+....17ycws . . . . . . . . . . .
&
12-17 years . , , . .
12years . . . . . . . . . . .13ytxws.. . . . . . . . . .14years . . . . . . . . . . .15ytmrs, . . . . . . . . . .16yeisrs . . . . . . . . . . .17years . . . . . . . . . . .
Girls—
12-17 years . . . . .
12yums . . . . . . . . . . .13ycws . . . . . . . . . . .14years . . . . . . . . . . .15ycws . . . . . . . . . . .16ycnrs . . . . . . . . . . .17yetws . . . . . . . . . . .
Standard error
0,08.
0.090.110.120.110.110.10
0.08
0.120.130.150.150.160.11
0.08
1.08 0.25 0.14.
0.140.290.340.74
●
*
0.17.
0.170.330.480.66
*●
0.15—
0.170.300.42
*●
●
.
0.090.48 0.09 0.10 0.12-
. . .**
0.260.150.23
0.11.
. . .
. . .
. . .
0.430.150.32
0.17—
. . .●
●
0.330.230.26—
0.11 0.31—
. . .
. . .
. . .*●
0.34
0.75
. . .
. . .
. . .●
.-.
0.76
0.67
0.41_
. . .
. . .
. . .
. . ..
0.36
0.52
0.32—
0.350.961.08
●
.●
0.43
0.24. . .
1.63**,●
. . .
1.40
0.421.50
***
. . .
0.36
0.351.44
.
. . .
. . .
. . .
0.90
0.230.510,77
*
-... . .
0.25
0.240.531.02
*. . .. . .
0.30
0.330.100.150.520.52
●
0.11
0.310.130.210.420.62
●
0.11
●
0.240.100.180.270.59
0.11
.
0.250.160.210.410.59
0.15
. . .●
0.280.130.270.31
0.16
. . .●
0.270.220.310.30
0.14
. . .
. . .●
●
0.280.12
0.18
. . .
. . .●
●
0.330.18
0.14
. . .
. . .●
●
0.390.29
0.51
2.76●
●
*
..-
. . .
1.27
0.481.33
●
●
.●
0.52
. . .
. . .
. . .
. . ..
0.48
0.71
. . .
. . .**
0.510.65
0.31
0.120.140.140.100.140.13
1.95●
●
. . .*
. . .
0.83*
. . .●
*
. . .
0.30
0.51●
.
. . .
. . .
0.410.140.210.95
*
. . .
1.CX3
1.71●
●
●
..
. . .0.410.150.280.34
.
—
. . .●
0.420.160.350.75
. . .
. . .
. . .●
0.360.17
. . .
. . .
. . .●
●
0.73
. . .
. . .
. . .
. . ..
0.74
. . .
. . .*●
0.650.42
from among those given by each psychologist reached, were referred to the Psychologicalduring the week, were exchanged and restored. Advisor for decision. Finally, a log of-any
unusual occurrences that might affect-the valid-ity of scores was maintained.
Scoring disagreements were discussed by the twopsychologists, and if agreement could not be
41
Table V1l. Standard errors for mean scores on the Vocabulary and Block Design short form of the WISC for ncminstitutionalized youths, by grade, sex, and
age: United St;tes, 1966-70
Present grade in school
High Higher
Total
Left
LowerSpecial
school than before
than 5th 6th 7th 8th 9th 10th Ilth 12th graduate 12thplacement
5th
graduating
Sex and age
Standard errorBoth sexes
12.17 vears . . . . . 0.71
=●
1.19
0.68
1.25
1.49
2.91
0.80
●
0.95
0.68
1.45
1.97
2.49
0.91
0.66
. . .●
1.45
0.79
1.60
1.26
0.80
. . .●
1.81
1.07
1.41
1.71
0.87
0.630.66 6.00 3.00 1.47 0.91 2.170.55
—
1.59
0.48
1.32
3.21
2.51●
0.65
1.82
0.68
1.45
2.55
3.47●
0.74
1.85
0.62
1.53
5.79●
..-
0.62 1.88_
. . .
. . .
. . .
. . .
●
1.92
2.52
1.76 1.17
12 years . . . . . . . . . . .
13 years . . . . . . . . . . .
14 years........,..
15 years . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . . . . . .
0.62
0.76
0.89
0.77
0.66
0.75
0.70
9.26●
●
●
●
-..
6.67
2.53
7.27●
●
●
. . .
1.88
1.33
2.78
3.94●
-..
. . .
1.46
0.62
2.01
1.87
3.13●
●
1.12—
0.89
2.33
2.44
2.94●
●
0.81
1.89
4.51
3.80●
●
●
2.67
. . .●
●
1.58
0.60
1:36
0.67
. . .
.-.
. . .●
●
2.28
3.72
. . .. . .
●
●
1.52
0.64
0.87
●
●
2.12
0.95
2.1912-17years . . . . ..
12years . . . . . . . . . . .
13years . . . . . . . . . . .
14years . . . . . . . . . . .
15yaars . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . . . . . .
0.83
0.84
1.03
0.95
0.89
0.81
0.63
12.69●
●
.-.
. . .
.-.
8.19
1.57
6.78●
. . .
..-
..-
5.78
1.42
2.48
4.23●
. . .
. . .
1.75—
1.56
3.11●
●
.-.
. . .
—
. . .
.-.
. . .
. . .●
2.52
2,93
2.74
6.CE
16.65●
●
●
2.34
. . .
. . .
. . .2.33
0.74
1.88
0.75
. . .
. . .●
●
2.M
0.70
0.94
. . .
. . .
. . .●
✎ ✎ ✎
3.83
3.72
. . .
. . .●
✃
2.82
2.37
1.24
Girls.
12-17 years . . . . .—
12years . . . . . . . . . . .
13years . . . . . . . . . . .
14years . . . . . . . . . . .
15yaars . . . . . . . . . . .
16years . . . . . . . . . . .
17years . . . . . . . . . . .
0.68
0.91
0.96
0.75
0.76
0.91
12.32●
●
. . .
. . .
. . .
4.46●
. . .
●
●
. . .
—
0.71
1.95
2.49●
●
●
. . .
. . .
. . .
. . .●
2.92
3.82
16.55●
,*
●
●
. . .2.00
0.92
1.74
1.66●
. . .●
●
0.842.18
3.25
. . .
. . .
. . .,
2.09
0.99
. . .
. . .
. . .●
●
4.05
. . .
. . .●
●
3.11
1.99
●
●
1.57
0.90
1.47
42 *U. S.GOVERNMENT PRINT3NGOFFICE:1973 543.a77/g
Series 1.
Series 2.
Sem”es 3.
Series 4.
Series 10.
Series 11.
Sm”es 12.
Series 13.
Series 14.
Series 20.
Sm”es 21.
Series 22.
r
VITAL AND HEALTH STATISTICS PUBLICATION SERIES
Formerty Public Health Service Publication No. 1000
Programs and collection procedures. —Reports which describe the general programs of the NationalCenter for Health Statistics and its offices and divisions, data collection methods used, defLnLtions,and other material necessary for understanding the data.
Data evaluation and methods research. —Studfes of new statistical methodology includfng: experi-mental tests of new survey methods, studies of vital statistics collection methods, new analyticaltechniques, objective evaluations of reliability of collected data, contributions to statistical theory.
Analytical studies--Reports presenting analytical or interpretive studies basedon vital and healthstatisacs, carrying the analysis further than the expository types of reports in the other series.
Documents and committee re#orts.— Final reports of major committees concerned with vital andhealth statistics, and documents such as recommended model vital registration laws and revisedbirth and death certificates.
Lkata from the Health Interview Survev. —Statistics on illness, accidental injuries, disability, useof hospital, medical, dental, and other services, and other health-related topics, based on datacollected in a continuing national household interview survey.
Data porn the Health Emmination Survey. —Data from dfrect exatrdnatiotl, testing, and measure-ment of national samples of the civilian, noninstitutional population provide the basis for two typesof reports: (1) estimates of the medically defined prevalence of specific diseasee in the UrdtedStates and the distributions of the population with respect to physical, physiological, and psycho-logical characteristics; and (2) analysis of relationships among the various measurements withoutreference to an explicit finite universe of persons.
Data from the Institutional Population Surveys. —Statistics relating to the health characteristics ofpersons in institutions, and their medfcal, nursing, and personal care received, based on nationalsamples of establishments providing these services and samples of the residents or patienta.
Data from the Hospital Discharge Survey. -Statistics relating to discharged patients in short-stayhospitals, based on a sample of patient records in a national sample of hospitals.
Data on health resources: manpower and facilities. —Statistics on the numbers, geographic distri-bution, and characteristics of health resources includLng physicians, dentists, nurses, ot%er healthoccupations, hospitals, nursing homes, and outpatient facilities.
Data on mortlzlity. —Various statistics on mortality other than as included in regular annual ormont~y reports-special analyses by cause of death, age, and other demographic variables, alsogeographic and time series analyses.
Daka on natlrlity, marriage, and divorce. —Various statistics on natality, marriage, and divorceother than as included in reguLar annual or monthly reports-special analyses by demographicvariables, also geographic and time series analyses, studies of fertility.
DZta from the National Natality and Mortality Survejs. — Statistics on characteristics of birthsand deaths not available from the vital records, based on sample surveys stemtnfng from theserecords, includfng euch topics as mortali~ by socioeconomic class, hospital experience in thelast year of life, medical care during pregnancy, health insurance coverage, etc.
/
For a list of titles of reports published in these series, write to: Office of InformationNational Center for Health StatisticsPublic Health Service, HRARockville, Md. 20852
