Education of Parents and Intelligence of Children
- Author:
Claude L. Nemzek and F. H. Finch
University of Minnesota
That there exists some degree of relationship between the amount of schooling of parents and the measured intelligence of their children has been revealed by investigations dealing with a variety of problems. Jones (6), in a study of 120 superior children, found that a large proportion of the parents were college graduates. Goodenough (5), using the Kuhlmann Revision, tested 213 children ranging in age between 18 and 54 months. Each child was retested with the same test after an interval of six weeks. The r between I.Q.’s on the first test and education of mother was .319 ? .042; between I.Q.’s on the first test and education of father, .264 ? .043. The r between I.Q.’s on the second test and education of mother was .353 zt .040; between I.Q. ‘s on the second test and education of father, .349 ? .040. The differences in the figures for father and mother are obviously no greater than might occur by chance.
Van Alstyne (11), using the Kuhlmann Revision, tested a group of children between ages 33 and 39 months. She reports an r of .51 ? .06 between father’s education and child’s mental age, and an r of .60 ? .05 between mother’s education and child’s mental age. A high degree of variability in intelligence characterizes her sample, the standard deviation of I.Q.’s being 21.67.
Pater son and his colleagues (10), in their work upon mechanical ability, observed an r of .14 between Otis I.Q.’s of 100 junior high school boys and education of father. For the same group, education of mother gave a corresponding figure of .08. Carter (3), in his study of family resemblances, reported the following relationships:
i The findings presented in this paper are from an analysis of the records of the University of Minnesota High School Personnel Office, made in co operation with Professor John E. Anderson, Director of the’ University n ‘f Minnesota Institute of Child Welfare.
Ni; Ni Min. P.E. Max. P.E.
r between education of father and: Son’s Arithmetic Score Daughter’s Arithmetic Score Son’s Vocabulary Score Daughter’s Vocabulary Score… r between education of mother and: Son’s Arithmetic Score Daughter’s Arithmetic Score Son’s Vocabulary Score Daughter’s Vocabulary Score. ..
102 107 101 107 102 107 101 107 74 79 74 79 74 79 74 79 .13 .02 .05 .15 .00 .11 .02 .06 .065 .065 .07 .06 .07 .06 .07 .065 .08 .08 .08 .07 .08 .08 .08 .08
Ni refers to the number of offspring included in the sample; N% to the number
of families represented.
Carter used the Courtis Standard Research Tests in Arithmetic, Series B, and Vocabulary Tests A and B of the University of Minnesota College Aptitude Test.
Miller (8), in a study of 1,021 children considered dull by teachers and 337 children considered bright by teachers, noted that larger proportions of the bright children were the offspring of parents with the greatest number of years of education; whereas, larger proportions of the dull children were the offspring of parents with the smallest number of years of education. Moore (9), in her study of high school graduates, selected by means of intelligence and objective achievement test records on 27,219 cases a group of 341 whom she designates as gifted. A comparison of the gifted and total groups on the basis of amount of parental education is given below:
Gifted Group Total Group per cent per cent Parental Education One or both parents attended college 33.4 10.9 Both parents attended high school 17.9 16.1 One parent attended high school 17.0 17.0 Neither parent attended high school 31.7 56.0 Bayley (1), who studied the mental growth of 61 children during the first three years of their lives by means of repeated infant tests, calculated the relationships between mid-parent education and infant test scores for the various administrations of the tests. Her results are included in Table I.
Table I Relationships Between Mid-Parent Education and Infant Test Scores Tests for Months 1 2 3 4 5 6 7 8 9 10 11 N 52 58 61 58 58 57 52 53 56 56 52 -.01 ? .10 -.02 ? .10 -.27 ? .09 -.22 db .09 -.29 ? .08 -.23 ? .09 -.19 ? .09 -.05 db .10 .06 ? .10 .08 ? .10 .00 ? .10 Tests for Months 12 13 14 15 18 21 24 27 30 36 N 53 53 46 52 51 53 48 51 47 49 .03 ? .10 -.05 ? .10 -.02 db .10 -.00 ? .10 .16 db .09 .29 ? .08 .50 ? .07 .41 ? .08 .44 ? .08 .47 ? .08
There is a peculiar trend in the correlations for successive ages, which may possibly be due to the nature of the test employed. There is considerable evidence that such tests used at early ages do not measure the same traits as do tests designed for older children.2 Three studies of foster children included an examination of the relation between amount of schooling of foster parent and intelligence of the child. Burks (2) reports on this relationship for a group of foster children and also for a control group of own children : Foster Group
P.E. N Control Group P.E. r’s between I.Q.’s of children and: Grade reached by father Grade reached by mother .01 .17 .05 .05 173 194 .27 .27 .06 .06 102 103
The foster children on which Burks’s results are based were placed in the foster homes before they had attained an age of 12 months, and ranged in age from 5 to 14 years at the time of test. Control cases were matched with foster cases in age, sex, locality of residence, type of neighborhood, and occupational level of father. Freeman (4) reports a correlation of .42 ? .03 between educa2 See Shirley, Mary M. The First Two Years, Vol. I, Postural and Locomotor Development. Minneapolis, University of Minnesota Press, 1931 pp 128-134.
Furfey, Paul Hanly and Muehlenbein, Josephine. The Validity of Infant Intelligence Tests, J. Genet. Psychol., 1932, 40, 219-223. tion of foster parents and intelligence of foster children. These subjects ranged up to 17 years 6 months in age at placement, and had been in the foster homes for an average of 6 years 11 months before being tested.
More recently Leahy (7) reported the following relationships between education of parents and intelligence of foster children:
Parent N C Uncorrected C Corrected P.E. Mother. Father. 93 98 .463 .498 .338 .367 .12 .13
Despite the fact that these cases were placed in foster homes at an age much below that typical of Freeman’s cases, Leahy finds evidence that selection is to a large extent responsible for the observed relationship. The subjects furnishing the data for which the relationship between amount of parental education and intelligence of offspring is reported below were tested at the time of their admission to the University High School. A series of five group tests was given to each individual, and the results of the several tests were converted into the equivalents of Terman Group Test I.Q.’s, according to the method devised by Miller.3 The middle value from these five has been taken as the measure of intelligence. Means and standard deviations shown in the tables indicate that a marked degree of selection with respect to both variables is present. Approximately 42 per cent of the fathers were classified as professional, and all except 2.6 per cent of the remainder were in occupational classes II and III.4
Eesults are reported for six groups of subjects, as follows: 1. High school entrants?719 cases representing as many families, not classified as to sex or to grade in which tests were given. 2. High school graduates?boys who were tested at the beginning of the ninth grade (or later in a few cases) and who graduated from University High School. 3. High school graduates?girls who were tested at the beginning of the ninth grade (or later in a few cases) and who graduated from University High School.
3 Miller, W. S. The Variation and Significance of I.Q. ‘s Obtained from Group Tests, J. Educ. Psychol., 1924, 15, 359-367. 4 Occupational classification from Goodenough, Florence L., and Anderson, John E. Experimental Child Study. New York, Century Co., 1931. 546 p.
Seventh grade?cases tested at entrance to the seventh grade.
Eighth grade?cases tested at entrance to the eighth grade.
6. Ninth grade?cases tested at entrance to the ninth grade. The last five groups overlap group one to a large extent, but in no case do they overlap each other. However, a given family may contribute to more than one of these groups, where records on more than one child from a family were available. The number of families represented in these five groups is somewhat smaller than the respective numbers of cases reported, as more than one child from a family occasionally fell into the same category. Where this is true, the probable errors of r as reported are slightly below the true values.
Table II Correlation Between Mother’s Education and I.Q. op Offspring 1. Mother’s Education in Years. 2. I.Q. of Offspring. Groups of Subjects Mi M2 S.D.i S.D.2 N 1. H. S. entrants. . 2. H. S. graduates (boys) 3. H. S. graduates (girls) 4. Seventh grade. . 5. Eighth grade… 6. Ninth grade .172 ? .024 .216 ? .04 .210 ? .04 .084 dt .07 .245 ? .06 .074 ? .06 12.48 11.77 11.89 12.67 12.89 12.39 118.65 117.9 117.25 118.95 119.60 119.15 2.53 2.77 2.75 2.52 2.58 2.55 12.65 12.35 12.25 14.05 13.15 11.65 719 269 312 93 108 136 Table III Correlation Between Father’s Education and I.Q. op Offspring 1. Father’s Education in Years. 2. I.Q. of Offspring. Groups of Subjects Mi Mj S.D.i S.D.2 N 1. H. S. entrants. . 2. H. S. graduates (boys) 3. H. S. graduates (girls) 4. Seventh grade. . 5. Eighth grade. . 6. Ninth grade .225 ?.024 .268 ? .04 .223 ? .04 .187 ? .07 .313 ? .06 .143 ? .06 13.86 13.58 13.41 14.25 13.51 13.68 118.65 117.50 117.15 118.95 119.60 119.15 3.68 3.91 3.72 3.67 3.55 3.78 12.65 12.25 12.25 14.05 13.15 11.65 719 263 313 93 108 136
From the tables it is obvious that in these data there is a positive relationship between amount of parental education and I.Q. of offspring. While the obtained values seem to center about .2 there is no means of determining how well this figure will hold for the population in general, since the cases here available are largely from the upper part of the scale, both in education and in intelligence. The homogeneity of the data is such as to suggest that the observed value may be below that typical of the population of this country.
It will be noted that the raw correlations are consistently higher for the fathers than for the mothers, but here again there is a difference in variability in education that must be taken into account. The corrected values of r,5 given in Table IV, indicate that the apparent difference in the two groups is due to this factor, and that the same degree of relationship exists for the fathers and the mothers.
Table IV Corrected Correlations of Parental Education with I.Q. of Offspring Groups of Subjects Father’s education Mother’s education (corrected). .225 .242 .268 .298 .223 .279 .187 .122 .313 .329 .143 .109
Bibliography
1. Bayley, Nancy. Mental growth during the first three years: A developmental study of sixty-one children by repeated tests, Genet. Psychol. Monog., 1933, 14, 1-92. 2. Burks, Barbara Stoddard. The relative influence of nature and nurture upon mental development: A comparative study of foster parent-foster child resemblance and true parent-true child resemblance, 27th Yrbh. Nat. Soc. Stud. Educ., Pt. I, 1928, 219-316. 3. Carter, Harold Dean. Family resemblances in verbal and numerical abilities, Genet. Psychol. Monog., 1931, 10, 1-104. 4. Freeman, Frank N., Holzinger, Karl J., and Mitchell, Blythe Clayton. The influence of environment on the intelligence, school achievement, and conduct of foster children, 27th Yrbt. Nat. Soc. Stud. Educ., Pt. I, 1928, 102-217. 5. Goodenough, Florence L. The relation of the intelligence of pre-school children to the education of their parents, School and Soc., 1927, 26, 54-56. 6. Jones, Alice M. An analytical study of one hundred twenty superior children, Psychol. Clin., 1925, 16, 19-76. s Formula (186). Kelley, T. L. Statistical Method. New York, Macmillan, 1924. 390 p. EDUCATION OF PARENTS 269 7. Leahy, Alice M. A study of certain selective factors influencing prediction of the mental status of adopted children in nature-nurture research, J. Genet. Psycliol., 1932, 41, 294?329. 8. Miller, Joseph. Causes of failure and success in school?II, Educ. Method, 1933, 12, 364-366. 9. Moore, Margaret Whiteside. A study of young high school graduates, Teach. Coll. Contrib. Educ., 1933, No. 583. 10. Paterson, Donald G., Elliott, Kichard M., Anderson, L. Dewey, Toops, Herbert A., and Heidbreder, Edna. Minnesota Mechanical Ability Tests. Minneapolis: University of Minnesota Press, 1930. 11. Van Alstyne, Dorothy. The environment of three-year-old children, Teach. Coll. Contrib. Educ., 1929, No. 366.
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.