Otis Self-Administering Test Research Articles

Attributive and Contrast Projection: A Cognitive Approach

Summary It was hypothesized that more intelligent and less psychologically differentiated Ss would be less likely to use attributive and contrast projection. The Otis Self-Administering Tests of Mental Ability were used as the measure of intelligence, and the Group Embedded Figures Test was used to measure psychological differentiation. Nineteen female and 11 male college students described Favorable and Unfavorable target photographs and themselves on a list of trait-rating scales. Possible accuracy of ratings was controlled. Attributive projection and contrast projection were demonstrated and found to be related to each other (r = .44, p<.01). There was no relationship between cognitive ability and the measures of projection. The correlation between attributive projection and psychological differentiation was r = .345 (p<.05); the correlation between contrast projection and psychological differentiation was r = -.51 (p<.01), indicating that more differentiated Ss projected more. The results are consistent with Witkin's position that more differentiated Ss are more likely to project, since projection is a relatively sophisticated mechanism. They also support Witkin's premise that intelligence and psychological differentiation are separable.

Otis Self-Administering Tests of Mental Ability, Intermediate Form, in Iran

A Persian form of the Otis Self-administering Tests of Mental Ability, Intermediate Form, was administered to a representative sample of Iranian secondary school pupils (944 boys and 554 girls). Iranian Ss as a group scored strikingly lower than their American counterparts. Boys scored significantly higher than girls both in different grades and across grades. Scores correlated significantly with a past academic performance criterion. Test-retest reliability coefficients over a 2-wk. interval were low relative to those given in the test manual.

A Comparison of Listening Tests with Standardized Tests

A NUMBER OF investigations have been conduct ed to determine relationships between listening tests and standardized tests (1, 2, 4, 5, 6, 7). How ever, these studies have sought to determine these relationships as a secondary por pose. Their pri mary purpose was the construction ofalistening test. Therefore, the purpose of this investigation was to study the listening ability of college fresh men by comparing the scores made on certain lis tening tests with the scores made on certain stand ard tests. This was done by comparing the results of listening tests with reading tests, by comparing the results of listening tests with academic aptitude and intelligence tests, by comparing the results of listening tests with grade point averages, by com paring group performances on existing tests of lis tening ability, and by comparing 1 i s t e n i ng ability with results of a measure of personality adjustment. Three listening tests were administered. These were the Brown-Carlson Listening Comprehension Test, Form Am; the Michigan State College Listen ing Test, Form HI; and the Stephens College Test of Listening Comprehension, Form A. Total scores on each of the three listening tests were compared with the sub-test scores of each of the following tests: the ACE Psychological Examina tion, College Level, 1954 Edition; the Cooperative English Test, C-2: Reading Comprehension, (High er Level), Form Y; the Otis Self-administering Tests of Mental Ability, Form A, 20 Minute Norms the Iowa Silent Reading Tests, New Edition, Form Am, (Revised); and the Thurstone Temperament Schedule. One hundred and ten students from the freshman class of Hope College, Holland, M i c higan, served as subjects for this investigation which was carried on during the school year 1955-1956. There were 62 men and 48 women. The students were divided into two groups. Group Aconsisted of students enrolled in the regular freshmen English classes. Group B consisted of students enrolled in the reading improvement course, l? a student happened to be a member of an English class chosen for test ing and was also enrolled in the reading improve ment course, his test data was included in Group B. Each of the groups were further divided accor ding to sex. Raw scores were used as t he basis for all sta tistical computations. Coefficients of corre 1 a t i o n were calculated by means of the Spearman rank dif ferences formula. The table includes coefficients of correlation be tween the three listening tests and five standardized tests. According to these data it appears that the Brown-Carlson test is the most satisfactory of the tests of listening for use with men in Group A. The test seemed to differentiate the members of the

An Experiment in the Use of Goal Sheets in Ninth Grade Mathematics

The purpose of this experiment was to determine the effect of the use of goal sheets upon achievement in the 2nd semester of ninth grade mathematics. Two sections in ninth grade general mathematics, each containing twenty-four pupils, were used in the study. The classes met during the third and fourth periods in the morning, both groups meeting in the same room, with the same teacher. The sections were carefully equated on the bases of scores obtained on the Otis Self-Administering Test of Mental Ability and achievement in ninth grade mathematics during the first semester. The course material for the second semester was arranged in five units and presented to both groups in exactly the same manner. As work was begun on each new unit, each member of the experimental group was furnished with a goal sheet. The teacher explained the purpose of the goal sheets and pointed out that the full cooperation of all pupils in the use of the sheets was essential if any benefit was to be derived. These sheets were not supplied to the control group. The goal sheets were mimeographed forms, at the top of which the pupils entered their names, date each sheet was issued, subject, and unit. The teacher then read a list of goals or things to be learned during the unit. These goals were copied on the sheets by the pupils. A separate column was provided for checking off each goal as it was accomplished. For example, one of the goals was to learn how to bisect an The pupils wrote on their goal sheets, Learn how to bisect an angle. As each pupil became certain he had mastered this skill, he entered a check-mark next to this goal in the column provided for checking off accomplished goals. Each textbook had a paper pocket fastened to the inside of the back cover. The goal sheets were kept in these pockets where they could be referred to at any time. At the end of each unit, an objective test was administered to both groups. These tests were constructed by the teacher and covered the course material thoroughly. Since the list of goals for each unit was compiled di 209

Achievement in Silent Reading in an Elementary School

The results of studies inquiring into achievements in reading at every school level confirm the general observation that our schools have not de veloped satisfactory reading skills in a large number of elementary and high school pupils. We still find teachers who utterly fail to understand the individual differences of children and to meet their separate needs. The limited capacities of some children may account for their failure of retarda tion in reading. In most schools, however, the lack of adjusting the teaching program to the abilities and needs of each child from the time he enters school produces many cases of reading deficiencies which, with proper teaching technique, could have been prevented. The study reported here was undertaken to determine the achievement in reading of the pupils of a parochial elementary school, to demonstrate to the teachers the necessity of better individualized instruction, and to enlist the cooperation of the Parent-Teacher Association of the community in pro viding funds for supplying each grade with reading material of varying difficulty and of a wide variety of interests. The data of the study are derived from measuring achievements in reading of the pupils of an elementary school in a University city of the Midwest. Being parochial in character, this school draws the majority of its pupils from the Catholic people of the city who represent all classes? professional, industrial, merchant, artisan, and laborer. The community has a preponderance of well-to-do families. Accumulated records of the pupils furnished data concerning the socio-economic status and the mental ability of the subjects of our study. Regarding the former, scores obtained on the Sims Score Card for Measurement of Socio-Economic Status indicate that the great majority of the subjects of investigation live in desirable environ mental circumstances. When treated statistically, scores derived from the Otis Self-Administering Test of Mental Ability, Forms A and B, admin istered to the pupils of Grade IV through VIII, the Otis Primary Test, ad 594

Item Validities of the Otis Self-Administering Tests of Mental Ability for a College Population

slightly more difficult from Form A. (The records for Form A were randomly selected from iooo test records, but records for Form B represent all those available at the time of study.) It also appears that the test is either slightly more difficult for women or else that women are somewhat inferior to men in intel ligence. Since so much evidence has been presented in refutation of the second alterna tive, we are left with only the first, if the differences are interpreted as significant. In Table II are presented data showing the relative difficulties (per cent of correct re sponses) and relative validities (biserial cor relations) of all items in Form A of the Otis test. Presented for comparison with our own figures are those nreviouslv renorted hv Chapanis. Examination of the data in this table reveals the following facts: i. The median difference for men between the per cent of adults and the per cent of students who got each item correct is .08 (Range .00 to .25). 2. The corresponding median difference for women is .09 (Range .01 to .35).

THE EFFECT OF 10 MG. OF BENZEDRINE SULFATE ON THE OTIS TEST SCORES OF COLLEGE STUDENTS

Forty-two male college students were divided into two groups. One group was given 10 mg. of benzedrine sulfate, and the other a placebo at the first session. They were then given a form of the Otis self-administering tests of mental ability (higher form). One week later, the groups were rotated and tested with an alternate form. Under the conditions of the experiment, we found that: 1. The 10 mg. of benzedrine were without a favorable effect on the Otis test score. 2. A small but statistically significant practice effect was noted upon retesting.

A New Method of Finding the Intelligence Quotient on Otis Self-Administering Tests, Higher Examination

A New Method of Finding the Intelligence Quotient on Otis Self-Administering Tests, Higher Examination

Some Psychometric Observations on the Tuberculous Patients in a National Institution: I. The Otis Self-Administering Test of Mental Ability

Some Psychometric Observations on the Tuberculous Patients in a National Institution: I. The Otis Self-Administering Test of Mental Ability

What Makes Ability in English?

During the past generation success in the various high-school subjects has been measured with increasing objectivity. At the same time interest in the relation between general intelligence and success in a given field and between success in one field and in another has had a steady and generally wholesome growth. While much of the interest in these relations owes its greatest impetus to the desire to predict success in the high school on the basis of elementary-school work or to predict success in college on the basis of high-school achievement, a large share of it has been focused, particularly in recent years, on special subjects and subject combinations in the high school. Believing that in the field of his major interest, English, the race is not always to the brightest nor the stigma of failure to the dullest, the writer set about obtaining data for a study of his own. Intelligence quotients derived from the Terman Group Test of Mental Ability, the Haggerty Intelligence Examination, and the Otis Self-administering Tests of Mental Ability were available from the individual pupil records in the office. The Terman and the Haggerty tests had been administered when the members of the group studied, the Senior class in high school, were in Grade VII; the Otis test had been given to them when they were in the first year of high school. To get a fourth measure, the writer used Army Alpha to test the surviving members of the Senior class. Four measures were available for 92 of the 102 members of the class. These 92 pupils were then given the Inglis Tests of English Vocabulary. Finally their English marks for the six semesters that they had been in high school were averaged. When the averages of the four intelligence quotients for every pupil are correlated with the average English marks for the six semesters in high school by the method of the scattergram and the

A Note on the Validity and Difficulty of Items in form a of the Otis Self-Administering Tests of Mental Ability

Although the Otis Self-Administering Test of Mental Ability (i) is widely used, there seems to be no published data concerning spe cific item validities of the test. purpose of this study was to obtain measures of the validity and difficulty of test items in Form A of the higher examination of this test based on a group different from that on which the orig inal item validities were established. As a matter of interest, these were computed sep arately for the sexes in order to note whether there might be any significant sex differences in the validity or difficulty of test items. Otis (2; Pg. 3) states that the items were originally validated on about 1000 high school students and 1000 grammar school pupils . subjects in the present study were 100 men and 100 women who had taken the test in connection with an other investigation. group of 100 women had a mean Otis I. Q. of 108 with the ex tremes at 72 and 140. Ages varied from 18 to 33. group of men had a mean Otis L Q. of in with a range of 72 to 135. Ages varied from 18 to 32 with three exceptions? one 36, one 37, and one 52. mean age was 25. Concerning the original validation, Otis states (2; Pg. 3): students were divided in each case into two groups, a 'good group' and a 'poor group'. same number were taken from each grade from both groups. good group constituted the young students, and the poor group the old students. These groups had reached the same average edu cational status, therefore, but at different rates. Now it is the rate at which a stu dent can progress through school that the mental-ability test is chiefly used to pre dict. Therefore this is believed to be the best criterion by which to judge the valid ity of each item that goes into the test. number of times each item was passed bv each group was then found and only those items chosen which showed a distinct gain in number of passes by the good group over the number of passes by the poor group in spite of the fact that the median age of the good group was over two years less than that of the poor group. Each item justified its inclusion, therefore, be cause it distinguished between students who progressed slowly and those who progressed rapidly. The items in each form of each exam ination have been arranged in the order of difficulty, according to the number of passes of each item by the students taking the preliminary editions.

The Relationship of Certain Factors to Failure Among Superior Pupils in a Junior High School for Negroes

This investigation was undertaken for the purpose of throwing some light on the causes or predictive associates of failure among superior pupils in junior high school. In the Northeast Junior High School of Kansas City, Kansas, a school for Negroes, there were located twenty-seven cases, 18 boys and 9 girls, with I.Q.'s, computed from Otis Self-Administering Tests of Mental Ability, of 110 or above, whose work on the average was unsatisfactory-failing or no higher than the lowest passing mark. For purposes of comparison, two other groups of superior pupils were employed: (1) a group of 27 pupils eminently successful in their school work, and (2) a group of 92 pupils whose work was average or better. Of the total group of 146 pupils of I.Q. of more than 110, 90 or 62 per cent were girls and 56 or 38 per cent were boys. They were distributed by grade as follows, 7th grade, 54; 8th grade, 38; 9th grade, 54.

Types of Errors and Remedial Work in the Fundamental Processes of Algebra

This investigation was conducted to supplement in an explicit and tangible manner the analyses of the specific errors which are most fre quently made in the fundamental processes of algebra. The study was confined to the difficulties of the ninth grade. The data were collected from 104 ninth-grade pupils in the East High School, Superior, Wis consin. With the exception of one boy, whose I. Q. was only 65, the intelligence quotients of these pupils ranged from 72 to 134. There were from 79 to 104 papers for each test, depending upon the daily attend ance. Since this investigation was conducted in regular classes, all the usual difficulties and inadequacies of classroom investigations were en countered. The results are, therefore, only indicative of what could be discovered or produced by controlled experimentation. This report de scribes such activities as any classroom teacher can conduct. These pupils were from five city ward schools, two parochial schools, and from outlying country schools, and ranged in age from twelve years to sixteen years. The results of the Otis Self-Administering Tests, Form A, were used as the basis of grouping. Due to conflicts in in dividual programs some shifting had to be done, and, as shown in Table I, the resulting groups were fairly heterogeneous. Later in the year, after the algebraic ability of individuals had been discovered, pupils were shifted, as far as their schedules would permit, into homogeneous groups. TABLE I Classification of Groups Number of pupils Average Range of Group in group I. Q. I. Q. A . 26 . 115.2 . 100 to 134 B . 28 .'. 108.3 . 85 to 126 C . 22 . 93.9 . 91 to 114 D . 28 . 87.7 . 65 to 102

How Constant is the IQ Yielded by the otis Self-Administering Test of Mental Ability

By the use of partial correlations one factor was held constant so that the correlation of the other two might be determined. These partial correlation coefficients are shown in Table III. This table shows that in every instance there is a much higher correlation between activity scores and test scores with intelligence held constant than there is between intelligence and test scores with activity held constant. This means that the test grades were influenced more by the activity of the child than by his intelligence. Objective tests are, perhaps, our best instruments for measuring the effectiveness of instruction. Therefore, numerous visitors mean more pupil activity, which in turn will give them higher test scores, and which indicate that children actually recite better in the presence of visitors.

A Conversion Scale for Comparing Scores on Three Secondary School Intelligence Tests

In 1926, the Secondary School Examination Board, composed of a number of private college preparatory schools, appointed a committee to report on the advisability of including an intelligence test as a part of the examinations for entrance to the member schools. An additional function of this committee was to recommend a test to be used. It was found that certain schools were using the results of such tests quite extensively, and that others had experimented or were experimenting with them. Naturally different tests had been used by different schools, so it was necessary as one phase of the work to resolve the various scores to some common denominator in order to enable school executives to translate the scores made on one test into approximately equal scores on another test. This paper presents the conversion or absolute scale worked out by means of which scores made on the Terman Group Tests of Mental Abil ity, Otis Group Intelligence Scale, Advanced Examination, and Otis Self-Administering Test could be stated in somewhat comparable terms. It is recognized, of course, that scores on this absolute scale are not exactly comparable. The aggregation of items on the various tests meas ure different functions, so a score or an IQ obtained on the Terman test is not quite the same thing as one obtained on the Otis Advanced Ex amination, though correlations between the two are high.2 The writer knows of no published correlations between the two varieties of the Otis test, but it should be high since the S elf-Administering Test is made up of items selected from the Advanced Group Intelligence Scale. A cor relation of .72 =!z .03 was obtained between the Terman and the Otis Self-Administering Test in the case of 111 boys used in this present study, who took the two examinations during the same month. It is to be distinctly understood that the absolute scales which follow are in tended as a convenient help to administrators who may have pupils enter

An Investigation of the Intelligence of Foster Children

THE purpose of this study' was twofold. Its practical purpose was to determine whether residence in a foster home appears to be of benefit to the children so far as their intellectual advancement is concerned. A more theoretical purpose was to determine whether the environment is a factor in intelligence or whether intelligence is determined entirely by the individual's inborn capacity. The bearing of the results upon the practical purpose will be obvious, so that they will be presented chiefly from the theoretical point of view. The investigation of the relative effect of inborn capacity and of the environment upon mental traits meets with serious difficulties. The major difficulty is that variations in environment and variations in capacity are commonly associated. That is, a child in a good environment is likely to have parents with superior ability. This makes it difficult to know whether to attribute the individual's achievement to the one or the other factor. In the case of foster children the environment has been altered. It may be possible, then, by statistical manipulation to determine whether this alteration has had an effect upon the children's intellectual ability. Accordingly a study was made of 401 foster children. A series of personal interviews was had with each child by the field worker of the study, Mrs. Blythe C. Mitchell. At these interviews the child was given the Stanford Revision of the Binet Scale and the International Test devised by Stuart C. Dodd. Interviews were also had with one or both of the foster parents. In many cases the foster parents consented to take the Otis self-administering test and a vocabulary test. On a specially prepared home rating blank,

24. Lange, Irene Dunn, The Value of the Otis Self-Administering Test of Mental Ability as a Means of Predicting the Success of Pupils in Junior High School. September, 1927. Pp. 63

24. Lange, Irene Dunn, The Value of the Otis Self-Administering Test of Mental Ability as a Means of Predicting the Success of Pupils in Junior High School. September, 1927. Pp. 63

An Experimental Study of a Visual Method in Latin Instruction

The emphasis on visual methods in Latin instruction is perhaps best shown by a comparison of Latin textbooks of twenty-five years ago and those of today. One significant difference is the greater number and better quality of illustrations in the later textbooks. There is a very definite effort in recent textbooks to illustrate, so far as possible, various aspects of Roman life in order to make it real to the pupil and to present pictures related to the words which compose the vocabulary. This article is concerned with an attempt to measure experimentally a new type of visual approach to the teaching of beginning Latin and to compare it with more common methods of teaching. Two groups of college students were selected; there were originally twelve students in each group, but only nine in each group completed the experiment. As shown in Table I, the control group was slightly superior in intelligence as measured by the Otis Self-Administering Test of Mental Ability, Higher Examination, Form A. The control group was also somewhat superior as measured by the Ohio State University Intelligence Test. None of the subjects had had any previous instruction in Latin. The groups were unequal in previous foreign-language training. The experimental group contained five individuals who had had previous training in a foreign language;

The Intelligence Quotient as a Prognosis of Success in Physics

The data presented in this article were obtained during the school year 1924-25 from a study made of three classes in highschool physics in the North High School, Minneapolis. About ninety pupils were enrolled, but data are presented for only fiftyeight pupils because the records for the others are incomplete. Mental ability was measured by means of the Miller Mental Ability Test, Form A, and the Otis Self-Administering Test of Mental Ability, Form A. The I.Q.'s obtained from the Otis test were computed according to the instructions given in the manual accompanying the test; those obtained from the Miller test were corrected so that they would more nearly approximate the StanfordBinet I.Q.'s. Both sets of I.Q.'s, then, are in approximate agreement with the Stanford-Binet I.Q.'s. It should be understood that the pupils were regularly classified pupils, fifty-six being in the eleventh grade and two in the twelfth grade. Forty were more than sixteen years of age. Only four were younger than fifteen. The group was a fair sample of what might be expected of an eleventh-grade class so far as maturity is concerned. Success in physics was measured by a group of tests covering the conventional topics studied in high-school physics, namely, metric and English units, simple machines, mechanics of liquids, mechanics of gases, forces and strains, motion and acceleration, sound, heat, light, and magnetism and electricity. There are twelve comprehensive tests represented, each consisting of at least two sections, namely, subject-matter information and problems for solution. The test on heat is made up of three information tests and one problem test. The highest possible score on all the tests is 415. Table I shows the total scores on the physics tests given during