Bright Children
(Mental and Physical Correlations) :Author: S. C. Garrison and G. M. Pullias. Professors of Educational Psychology, George Peabody College for Teachers, Nashville, Tenn.
The purpose of the present paper is to present certain data to show the relation of mental ability to physical development. Only bright children, those making a high score on the Binet-Simon intelligence test, are used in this study. Our problem is very well expressed in the question: Does the child of high intelligence possess physical development commensurate with his mental ability?
The subjects used in this study were pupils from the Demonstration School of the George Peabody College for Teachers. Since the study was restricted to children of superior intelligence, the number was necessarily limited to forty-five. School grades four to eight, inclusive, were included in the study as follows: Thirteen from the eighth grade, eleven from the seventh, five from the sixth, ten from the fifth, and six from the fourth. The data resulting from the tests were handled separately for each grade. For convenience the children were given numbers from one to forty-five and ranked according to I. Q. within the several grades. Seventeen boys and twenty-eight girls were used in the study.
The Stanford Revision of the Binet-Simon test was used as a measure of mental ability. Only children with an I. Q. of 116 or above were selected for use in this study. The highest I. Q. found was 170. The percentile method was used in the study and according to Terman1 all I. Q.’s of 116 or above are in the best 10 per cent or above the 90-percentile. All the physical measurements were taken by the writers in April, 1921.
The following physical measurements were taken: Standing height, weight, strength of grip, and vital capacity. All apparatus used is fully described by Whipple.2 The stadiometer was used in finding the height of the children. The thickness of the shoe heels was deducted from the total height. The weight was obtained by accurate scales, the type especially devised for anthropometric work, and given in pounds or kilograms. No reduction was made from 1 Terman, L. M. The Measurement of Intelligence.
s Whipple, G. M. Manual of Mental and Physical Tests.
weight for clothing. Strength of grip was determined by the improved form of Smedley’s dynanometer and expressed in kilograms. Only the right hand grip was taken. The highest score of two trials was the one recorded for strength of grip. The vital capacity was tested by the wet spirometer in terms of cubic inches. Each subject was given two trials and the highest score recorded in each case. There is at present a growing tendency to use mental, educational, and physical measurements in educational diagnosis. The present belief is that the bright child does the best school work and is also the best physical specimen. The original data from which this study is derived are treated in this paper so as to show the relation physical measurements bear to intelligence in the case of bright children. The degree of rank correlation between these measurements is shown. This should not be confused with the coefficient of correlation. In showing the degree of rank correlation the percentile method is used. Since the number of children in the study is limited to forty-five, the correlative positions of the various scores are shown in terms of 10-percentiles.
Table 1 shows how the rough data were arranged for the purpose of studying the correlative positions of the several scores. Five such tables were made, one for each grade represented in the study. The method of arrangement needs but little explanation. The study is one of rank correlation and the procedure was as follows: The I. Q.’s secured by the mental test were placed in a vertical range from the highest to the lowest for each grade. The scores made by each child were then arranged and the corresponding percentile in which any particular score fell was placed in a column after that score. In finding the percentile standing of any pupil, a percentile table for the sex and age to which that pupil belonged was used. The table finally shows not only the scores made in the physical and mental tests, but the particular percentile into which the measurement for each of the pupils in each trait fails. This method of tabulation was followed for each of the grades. An examination of Table I shows that pupils 4, 5, 6, 11, and 12 have records in physical measurements which fall above the 90-percentile in only one case each. Pupils 9 and 10 have two records each in the physical measurements which fall above the 90-percentile, while pupils 2, 8, and 13 each have three records and pupils 1, 3, and 7, four each which fall there.
Table I.?Arrangement of Data for Purpose of Studying MentalPhysical Correlations. Eighth. Pupil. Sex. Actual Age. Yr. Mo. Binet. Weight. Lbs. Height. In. Strength of Grip. Kg. Vital Capacity. Cu. In. 12 12 13 13 14 13 13 13 14 13 14 15 13 158 139 127 126 125 123 123 122 121 120 120 119 116 90 90 90 90 90 90 90 90 90 90 90 90 90 116 91.3 106.5 89.3 86.3 82.3 139.3 115.5 131 101 104.5 86 106 90 80 90 40 20 40 90 90 90 70 60 10 SO 62.8 62 63 59 59 58 64.5 61 61.3 61 61 61 61 90 90 90 50 40 60 90 70 60 70 60 30 80 55 69.3 73.9 55 46.2 46.2 36.3 68.2 51.2 56.1 56.1 51.9 85.8 90 90 90 90 90 90 90 90 90 90 90 90 90 170 197 200 132 156 150 173 168 152 160 188 150 172 90 90 90 70 60 70 90 90 80 90 80 70
The frequencies of correlation for the 90-percentile of intelligence with the four physical measurements are given in Table II. Correlations for the eighth grade are shown in the second vertical column; for the seventh, in the third vertical column, and so on for the other grades. Thus it is seen from the table that for the eighth grade five pupils correlated by rank with one physical measurement; two with two physical measurements; three with three physical measurements; and three with four physical measurements. Table II.?Frequencies of Correlation of Physical Measurements with Stanford-Binet Intelligence Test for the Ninetieth Percentile. Number of Physical Measurements Correlating.
Number of Pupils Correlating.
Gr. VIII Gr. VII Gr. VI Gr. V Gr. IV Totals. Total. 45
The frequencies of correlation of the separate physical measurements with mental ability are indicated in Table III. The first line of the table reads: Of the thirteen pupils in the eighth grade, six correlated by rank with weight; of the eleven pupils in the seventh grade, two correlated; of the five in the sixth grade, one correlated; of the ten in the fifth grade, three correlated; of the six in the fourth grade, two correlated, and of the forty-five children in the five grades there was a total correlation of fourteen. The percentage of correlation for the eighth grade was 46; for the seventh grade, 18; for the sixth grade, 20; for the fifth grade, 30; for the fourth grade, 33; and for all the grades a correlation of 31. The table contains a similar showing for each of the physical measurements. The highest correlation is found in strength of grip; the lowest in weight. The correlation of strength of grip with mental ability as shown in this table is rather significant.
Table III.?Frequencies of Correlation of Separate Physical Tests with Intelligence Quotient. Tests. Number of Correlations. Gr. VIII Gr. VII Gr. VI Gr. V Gr. IV Total. Per Cent of Correlations. Gr. VIII Gr. VII Gr. VI Gr. V Gr. IV Total. Weight Height Strength of grip Vital capacity Number of pupils 46 30 100 54 18 27 100 36 20 60 100 40 30 60 100 40 33 66 100 16 31 44 100 40 45
The range of deviation for the scores failing to correlate is given in Table IV. The second vertical column of the table reads: Of the forty-five possible correlations with weight, fourteen actually correlated. Of the twenty-nine scores failing to correlate, eight fall in the 80-percentile, five in the 70-percentile, and so on for the rest. Table IV.?Showing Range of Deviation in Terms of Percentiles for Scores Failing to Correlate with Reference to the Various Physical Tests.
Weight, Height, ins. Strength of Grip, Vital Capacjty, cu. ins. Possible number of correlations Actual number of correlations Scores deviating one percentile Scores deviating two percentiles… Scores deviating three percentiles.. Scores deviating four percentiles… Scores deviating five percentiles… Scores deviating six percentiles…. Scores deviating seven percentiles. Scores deviating eight percentiles..
The writers are aware of the fact that the percentile method is often misleading, because of the inequalities in the distances between the percentile steps. However, when interpreted in terms of P. E., the differences are found to be not so large, too small in fact to outweigh the practical advantages of the precentile method. Table IV could very readily be converted into terms of P. E. deviations from the 90-percentile instead of percentile deviations from that point by referring to Thomdike, Mental and Social Measurements, Table 44, page 198.
Conclusions. 1. There is considerable correlation between physical development and mental ability indicated by the results of our measurements. Fifty-four per cent of the physical measurements fall above the 90-percentile, the point above which all our I. Q.’s fall. Seventysix per cent of the physical measurements fall above the 70-percentile. That is, in three-fourths of our cases the results for our physical measurements fall not more than two steps, on our percentile tables, from the results of our mental measurements.
2. Some physical measurements show a higher degree of correlation with mental measurements than other physical measurements. A very high correlation is shown between strength of grip and intelligence. 3. These results show definitely that superior mentality and superior physical development go along together.
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.