Birth Rate and Native Intelligence
- Author:
Willis W. Clark, M.A.
Sociologist, California Bureau of Juvenile Research, Whittier, Calif.
Introduction. It is commonly considered that families of low mentality have more children than those of normal intelligence; i. e., are especially prolific in reproduction. “Feeble-minded beget feebleminded and they propagate fast” is the essence of usual statements in regard to this phase of the social problem. Elderton (2, p. 21) states: “It seems only too true at the present time that the physically and mentally weaker stocks are reproducing themselves at a greater rate than those of sounder physique and intelligence.” Data which may be interpreted in relation to this problem, and which may tend to modify radical statements concerning the relatively higher birth rate of inferior stock, are available in the laboratory of the Department of Research, Whittier State School. Qualitative data. Family histories of boys committed to Whittier State School from the various counties of California have been prepared by trained field-workers during the past five years. The process of obtaining the history includes in practically all cases a visit to the home and relatives of each boy, and analysis and synthesis of mental, physical, and social factors affecting each member of the family (see 7, pp. 71-82), and finally the preparation of a family history manuscript, elaborating on the plan outlined by the Eugenics Record Office (3, pp. 35). The histories vary in size from 10 to 50 typewritten pages, averaging 20 pages, contain data concerning from 2 to 6 generations, and average 20 to 30 individuals for each history. The essential accuracy of the information concerning the number of siblings (brothers and sisters) used in this study is assured. Each boy has been given an intelligence test, using the Stanford Revision of the Binet-Simon Intelligence Scale in all cases. This is generally accepted as the most accurate measure of native intelligence available (see 6 and 8); the intelligence quotient (I. Q.), which is the ratio of mental age to actual or chronological age, may be used in preparing a statistical table indicating the frequency distribution of native intellectual endowment.
That there is a high correlation between the intelligence of children and parents is usually assumed as being governed by the laws of heredity, and has been verified by mathematical formulae. For example, Pearson (4, p. 24) found a coefficient of correlation of 0.58 between the intelligence of father and son. Unpublished correlations prepared by Popenoe (5), using data of Department of Research concerning the same cases included in this study, found further evidence of this positive relationship. Where the intelligence of one parent and one child was known, the correlation was found to be 0.43 (P. E. 0.023), frequency, 1301 cases. Again, using a more highly selected group in which the intelligence of both parents was known and was in the same general group; e. g., both normal, both feebleminded, etc.?a positive correlation of 0.70 (P. E. 0.029) was found, indicating a high relationship between the two factors. Thus it appears that we may assume, for the purposes of this study, that the I. Q. of the boys in the School would show the relative degree of intelligence of the parental stock.
Quantitative data. A distribution of the I. Q.’s of 323 boys in Whittier State School concerning whom we have both an intelligence examination and an accurate family history, is given in Table I. The I. Q. classification is given in groups of ten points. The range of number of siblings is from 1 to 14 children, giving a total of 1204 children. In considering the number of brothers and sisters in the family only those of full relationship are included, and still-births, as well as half-, step-, and foster-fraternity, are eliminated from the distribution. This table was used in the calculus of correlation. Table II presents a reclassification of the data into general social intelligence groups, as superior, average-normal, dull-normal and borderline, and feeble-minded according to the usual estimates of 1. Q. values. It is seen that 3.5 per cent are superior, 25 per cent average-normal, 50 per cent dull-normal or borderline, and 21.5 per cent feeble-minded.
Statistical analysis. Statistical measures of averages and relationship between the native intelligence of the boys (therefore, as previously indicated, the intelligence of the parents) as measured by intelligence examinations and the number of fully related brothers and sisters are given in Table III. The coefficient of correlation derived from Table I by use of Pearson’s formula is ?0.079 (P. E. 0.037). Strictly interpreted this indicates a very slight tendency for the more intelligent families to be smaller and for the less intelligent to be larger, but it is of such small variation to be practically negligible. The principal feature of the distribution is the wide range or scattering for all of the groups which itself indicates a low correlation. Considering further the averages presented in Table III, we find verification only of the same slight tendency. In the 323 families there were 1204 children of full relationship, an arithmetic average of 3.7 per family. The arithmetic average for the superior groups is 3.0 children; average-normal, 3.4; dull-normal and borderline, 3.9; feeble-minded, 3.8 children. The mode, not marked in any group, increases from 1 child for the superior and average-normal, to 3 children in the two lower intelligence groups. Finally, the medians for each of the groups show the same tendency as the other measures of averages,?there is a slight tendency for those of inferior intelligence to be members of the larger families.
Qualifying factors. Two factors concerning the validity of data for statistical purposes should be considered. They are (1) the families used as a basis for this study are a selected group, and (2) the average intelligence is low.
Each family considered have one factor in common; namely, a boy has been committed to a state industrial school. Otherwise they are from various racial groups, from city, town, and country, well-to-do and poverty stricken, of various grades of moral character, and from a great variety of home conditions.
The median I. Q. for the group of boys in this study is 81.5, while Williams (8, p. 25) found the median I. Q. for the usual population of the School to be 82, a negligible difference. This I. Q. indicates an average mental age of 13 years for adults in this group. While this is considerably below the expected average adult mental age for school children as found by Terman (6), it is practically identical with the average mental age of adults, which Doll (1) finds that the U. S. Army tests indicate.
Thus it appears that the data may be used as a relatively unselected group and with reasonable expectation of trustworthy results. Summary and conclusion. The data presented in this study show only a very slight tendency for boys of higher intelligence (by implication, parents of higher mentality) to be members of smaller families indicated by a correlation of ?0.079 (P. E. 0.037). It would seem to indicate that the usual assumption that persons of low mentality have much larger families is invalid, and that the tendency is practically negligible. However, there is nothing in these data which should militate against the desire to reduce the frequency of disgenic characters; all efforts to promote a higher birth rate among the more eugenic groups should be encouraged.
References.
1. Doll, Edgar A. The Average Mental Age of Adults. Journal of Applied Psychology, IV-4, December, 1919. 2. Elderton, Ethel M. The Relative Strength of Nurture and Nature. Galton Laboratory Lecture Series, III. London. 1915. 3. Laughlin, H. H. How to Make an Eugenical Family History. Eugenics Record Office, Bulletin No. 13. 1915. 4. Pearson, Karl. Nature and Nurture: The Problem of the Future. Galton Laboratory Lecture Series, VI. London. 1910. 114 THE PSYCHOLOGICAL CLINIC. 5. Popenoe, Herbert. Unpublished correlations prepared in Department of Research Laboratory, Whittier State School, Calif. 1919. 6. Terman, Lewis M. The Measurement of Intelligence. Houghton, Mifflin Co. Boston. 1916. 7. Williams, J. Harold. Individual Case History Outline. Journal of Delinquency, V-3, May, 1920. 8. Williams, J. Harold. The Intelligence of the Delinquent Boy. Whittier State School, Department of Research. Monograph No. 1. 1919. Table I.?Intelligence Quotient of One Boy and Number of Children in Family. i.Q. Number of Children in Family.
9 10 11 12 13 14 Total 121-130.. 111-120.. 101-110.. 91-100.. 81-90… 71-80… 61-70… 51-60… 41-50… 2 9 29 52 82 79 44 21 5 Total. 58 51 69 37 35 36 16 11 323 Table II.?Intelligence Classification of One Boy and Number of Children in Family. Classification. Number of Children in Family. 1 2 4 5 6 7 8 9 10 11 12 13 14 Total Families. Superior (I.Q.’s 111130) Average-normal (I.Q.’s 91110) Dull-normal and borderline (I.Q.’s 71-90) Feeble-minded (I.Q.’s 41-70) Total 11 81 161 70 323 BIRTH RATE AND NATIVE INTELLIGENCE. 115 Table III.?Statistical Averages and Intelligence Groups. Classification. Total Families. Total Children. Arithmetic Average. Mode. Median. Superior (I.Q.’s 111-130) Average-normal (I.Q.’s 91-110) Dull-normal and borderline (I.Q.’s 71-90) Feeble-minded (I.Q.’s 41-70) Total. 11 81 161 70 323 33 276 627 268 1204 3.0 3.4 3.9 3.8 3.7 3.25 3.43 3.93 3.79 3.76
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.