A GROWING NUMBER of parents are choosing home schooling over public or private school education (Holt, 1983, p. 392). The reason for such choices vary, but research indicates that the majority choose to do so for religious reasons, such as objections to public school curriculum, fear of humanism, and lack of moral teachings or philosophical differences with the school system, such as the belief in a child’s need for more flexibility and creative activities and pursuit of individual interests (e.g., Divoky, 1983; Van Galen, 1986, 1987). Almost all home school parents desire to exercise more control over their children’s environment, development, and learning.
Those who choose to home school their children, whether for religious reasons or educational reasons, are confronted with the recurring criticism related to lack of socialization opportunities for the child. Proponents of home schooling view socialization in schools as negative, and outline ways in which it is possible to provide positive socialization through home schools (e.g., Ballman, 1987; Moore, 1984; Pagnoni, 1984). Those who perceive socialization as being inadequate in home schools point to the lack of peer association as depriving children of natural social contact.
An accepted way of measuring social skills is through the use of a self-concept instrument, normally a self-report instrument, based on the belief that children with positive self-concepts are well-adjusted and inclined to be socially competent (Moore, 1979, p. 58). Self-concept is viewed as a “central construct for the understanding of people and their behavior” (Fitts, 1971, p. 3). It has been found to be relatively fixed and stable throughout life and to emerge from social interactions (Kinch, 1963, p. 481). Kink’s (1983) study revealed that home schooling parents in California listed positive self-concept as the greatest benefit of home schooling. Taylor (1986) studied self-concept in home school children as a measure of socialization, and Breckenridge (1989) compared self-concept in children attending Christian schools and public school as a measure of socialization skills. This study was an attempt to compare children from the three educational settings: home school, Christian schools, and public school.
In an attempt to address the concerns regarding socialization, this researcher chose to study children in Texas Baptist churches to compare the self-concept scores of children in grades four, five, and six. Preliminary work for this research found that three educational settings were represented and that comparison of children across the three settings would provide additional information. It was assumed that children who attend the same church come from families who are somewhat similar in values, particularly spiritual values and educational values. Since these values could be consistent across the three groups, true differences in other areas could become more evident. In order to obtain a larger sample, only churches with large congregations were used in the population. As a result, all of the churches were in large metropolitan areas, thus representing a similar socioeconomic status for the sample.
This study used the psychological construct of self-concept to measure the social skills of fourth, fifth, and sixth-grade children in Texas Baptist churches who were educated in three different settings: home schools, Christian schools, and public schools. Mean scores were compared according to grade, gender, and school type to determine if significant differences existed.


The Piers-Harris Children’s Self-Concept Scale (The Way I Feel About Myself) was developed in 1969 by Ellen V. Piers and Dale B. Harris. Demographic information was collected on the cover page of the instrument related to gender, grade, and school type.
The instrument contained eighty first-person, declarative statements which were read aloud and repeated to the children as they followed along and responded “yes” or “no” to each statement. Answers were keyed to high self-concept; thus a higher total score indicated a positive concept of self. An overall score was figured, as well as six cluster scores including: physical appearance and attributes, anxiety, intellectual and school status, behavior, happiness, satisfaction, and popularity. Reliability data, using the original standardization sample of 95 items, produced reliability coefficients ranging from .88 to .93, based on Kuder-Richardson and Spearman-Brown formula. Test-retest coefficients over two- and four-month periods were in the satisfactory range of .72 to .77 (Piers, 1984, p. 50).
Content validity was built into the scale by using children’s statements about themselves as the universe to be measured as self-concept. By writing items pertaining to that universe of statements, the authors defined self-concept for their scale. For construct validity, the PHCSCS scale was administered to eighty-eight adolescent institutionalized retarded females. As predicted by Piers and Harris, these girls scored significantly lower than normals of the same chronological or mental age. This was interpreted as meaning that the PHCSCS did measure self-concept and discriminated between high and low self-concept (Piers, 1984, p. 95).
Concurrent validity was established by comparing PHCSCS scores with other measure of self-concept. Mayer (cited in Shreve, 1973) found a correlation of .68 with the Lipsitt Children’s Self-Concept Scale. Other studies with the PHCSCS reported significant negative correlations with measures of anxiety and reported personal problems. When compared with the Thomas Self-Concept Values Test against criteria set forth by the American Psychological Association, the most nearly satisfactory test was found to be the PHCSCS (Shreve, 1973, p. 106).

Description of the Sample
Letters were mailed to 94 churches reporting an attendance of 630 or higher in Sunday School in the 1988 Texas Baptist Annual. Additional letters were mailed to 50 churches listed by the Baptist General Convention of Texas as having Ministers of Childhood Education. The number of letters mailed totaled 144 with 11 churches meeting the qualifications for the study. These qualifications included: a total of 25 or more children in grades four, five, and six enrolled in Sunday School; employment of a staff member for childhood education; and ministering to all three groups under studyChome school, Christian school, and public school. A total of 311 surveys were completed with 257 usable instruments used in data analysis. These were distributed into the research groups as indicated in Table 1.


–This table is missing in this web page presentation; you may find the PDF format elsewhere on this web site with the table included–

Table 1. Subjects used in the research.

The packets with instruments, instructions for guidance in completion of the scale, and a stamped, self-addressed return envelope, were prepared and mailed to all churches qualifying for the study. Staff members then administered the instrument to all children in the three grades under study. Administration by a staff member was chosen in order to eliminate potential parental influence in the testing process. Instruments were then collected and returned to the researcher. Using three-way analysis of variance and appropriate t-tests for significance at the 0.05 level, the data were analyzed. A brief descriptive summary of the findings was prepared and mailed to assisting church staff members.


No statistically significant differences were found across school type, grade, or gender in the overall self-concept scores of older children in Texas Baptist churches. An F(5, 239) = 1.601, p<.05 indicated no significant differences. This was interpreted to mean that older children who attended home schools, Christian schools, and public schools did not significantly differ in their self-concept scores as measured by the Piers-Harris Children’s Self-Concept Scale. In addition, no significant differences were found among any of the three grades, F(2, 254) = .448, p<.05; nor were any significant differences found in overall scores between males and females, t(239) = .796, p<.05.
An insignificant F indicated that the differences in the means of the groups under study could not be attributed to anything other than sampling error. Table 2 shows means and standard deviations.


–This table is missing in this web page presentation; you may find the PDF format elsewhere on this web site with the table included–

Table 2. Mean scores and standard deviations for “total self-concept.”

A significant factor may have been the low return rate from home school children, with only 13% of the sample being home schoolers. In collecting the data, it was found that many of the home school children in Texas Baptist churches were in the lower grades–kindergarten through third–rather than in the older grades. However, since the instrument was designed for use with children grades four through twelve, an older group was necessary for valid results. The low number of respondents in this one group, however, increases the chance for a Type II error, retaining a false hypothesis, which in this case states that no differences exist when in reality, differences may exist.
When compared to the normative population mean of 51.84 (out of a possible 80) as reported in the Piers-Harris Children’s Self-Concept Scale Manual, the mean score of self-concept for the sample under study (M = 61.264) was significantly higher than the population mean, t(239)= 11.40, p>.05. This can be interpreted to mean that children who attended Texas Baptist churches had significantly higher self-concepts than the normative population of public school children.

Discussions and Implications

The results of this study relate specifically to the claims of both proponents and opponents of home schools. Using self-concept as a measure of socialization skills, it was evident that in this sample there were no significant differences in children’s self-concept among the three educational settings under study. This may indicate that other factors, rather than educational setting, more significantly enter into self-concept development. These factors many include such areas as parental self-concept, relationships with parents, individual academic achievement, or other factors as revealed in previous studies (e.g., Beane, Lipka, & Ludewig, 1980; Brookover, Thomas, and Paterson, 1964; Calsyn & Kenny, 1977; Kagel, 1981; McDonald, 1980; Purkey, 1970; Sears, 1970). However, it also indicates that claims regarding superiority of either setting over another in terms of self-concept development or socialization in general may be unfounded. If indeed there are no differences across the groups, socialization may be more directly tied to factors other than influences within the school setting.

Grade Level
There were no significant differences in measured self-concept among fourth, fifth, and sixth-grade children in Texas Baptist churches. Previous findings indicating that a child’s self-concept deteriorates as he progresses through grade levels were not supported in this study (Felker, 1974, p. 50).

There were no significant differences in measured self-concept between males and females in Texas Baptist churches. This was also supportive of previous studies regarding self-concept differences in gender (Rosenberg, 1979, p. xiv).
Several cluster scales which were additionally analyzed revealed several significant findings. There were significant differences in males and females in the study regarding anxiety in relation to self-concept. Males (M = 11.019) appeared to be less anxious than females (M = 10.086), t(239) = 2.447, p<.05.
This sample also showed that fifth-grade students scored higher than either fourth-grade or sixth-grade students in behavior. This indicated that fourth and sixth-grade students acknowledged more behavioral difficulties than fifth-grade students. This may be due to the absence of behavioral problems in fifth-grade children or to a tendency for fifth grade students to deny their problems.
Differences were found to be significant across school type in regard to intellectual and school status and in relation to anxiety. Christian school students scored higher in both intellectual and school status and in anxiety clusters, than either home school or public school students. Christian school students also assessed their abilities to perform academic tasks as being better than did home school or public school students. They also appeared to experience fewer emotions associated with high levels of anxiety, such as worry, nervousness, and fear.
As parents consider schooling alternatives for their children, it will be necessary for them to take under advisement research regarding various aspects of child development and how each aspect can be strengthened in the potential school setting. One of the aspects considered should be the development of self-concept and the need for its enhancement in the childhood years. It appears from the research that for Baptist children, educational setting does not significantly influence self-concept in either a negative or a positive way. Therefore, the choice to home school, in general, cannot be construed as a choice to necessarily enhance or damage a child’s self-concept; nor can the public school be blamed entirely for its negative contributions to the child’s view of himself. These observations, however, are made in light of the fact that the children involved in the study were a part of a local church ministry, specifically Baptist churches, and this fact opens the door for the possibility that factors inherent in religious upbringing may contribute to self-concept in such a way that educational environment may not be as predominant an influence.
As a result, it may be assumed that older children who attend church, as a whole, do not suffer from choices to attend either type of school, at any grade level, regardless of gender. However, parents and educators are encouraged to be sensitive to other factors that may contribute to low self-concept, such as academic achievement, physical abilities, and perceived opinions of teachers and parents. Educators in churches and in schools could enhance self-concept development by projecting positive attitudes toward children and giving them opportunity to succeed at various tasks.
Additional research comparing the three educational settings which takes into account factors such as parenting styles, differences in school philosophies, and various socioeconomic factors would give deeper understanding  of self-concept and its effects on the development of the child.


Ballman, Ray E. (1987). The how and why of home schooling. Westchester, IL.: Crossway.
Beane, James A., Lipka, Richard P., & Ludewig, Joan W. (1980). Synthesis of research on self-concept. Educational Leadership, 38, 84-88.
Breckenridge, Lillian (1989). A comparison of self-concept in Christian school and public school students in grades four, five, and six from selected Church of God churches. Doctoral dissertation, Southwestern Baptist Theological Seminary, Fort Worth, TX.
Brookover, Wilbur B., Thomas, Shailor, & Paterson, Ann. (1964). Self-concept of ability and school achievement. Sociology of Education, 37, 271-278.
Calsyn, Robert J., & Kenney, David A. (1977). Self-concept of ability and perceived evaluation of others: Cause or effect of academic achievement? Journal of Educational Psychology, 69(2), 136-145.
Divoky, Diane (1983). The new pioneers of the home-schooling movement. Phi Delta Kappan, 64, 397.
Felker, D. (1974). Building positive self-concepts. Minneapolis, MN: Burgers Publishing Co.
Fitts, William H. (1971). The self concept and self-actualization. Nashville, TN: Dede Wallace Center.
Holt, John (1983). Schools and home schoolers: A fruitful partnership. Phi Delta Kappan, 64, 392.
Kagel, Carol M. (1981). The association between children’s self-concepts and parental attributes. Doctoral dissertation, University of Cincinnati, OH.
Kinch, John W. (1963). A formalized theory of the self-concept. The American Journal of Sociology, 68, 481-486.
Kink, Linda Lee (1983). An analysis of sacramento area parents practicing home schooling. Masters thesis, California State University.
McDonald, Dan A. (1980). Enhancing a child’s positive self-concept.” In T. D. Yawkey (Ed.), The Self-Concept of the Young Child, p. 51-62. Provo, UT: Brigham Young University Press.
Moore, Raymond (1979). School can wait. Provo, UT: Brigham Young University Press.
Moore, Raymond (1984). The cradle and the gap, In A Blueprint for Education Reform, pp. 231-250. Washington D.C.: Free Congress Research and Education Foundation.
Pagnoni, Mario (1984). The complete home educator. Burdett, NY: Larson Publications.
Piers, Ellen V. (1984). Piers-Harris Children’s Self-Concept revised manual 1984. Los Angeles, CA: Western Psychological Services.
Purkey, William Watson (1970). Self-concept and school achievement. Englewood Cliffs, NJ: Prentice-Hall.
Rosenberg, Morris (1979). Conceiving the self. New York, CA: Basic Books.
Sears, Robert R. (1970). Relation of early socialization experiences to self-concepts and gender role in middle childhood. Child Development, 41, 267-289.
Shreve, E. E. (1973). A Critical analysis and evaluation of evidence regarding the reliability and validity of four selected measures of self-concept. Doctoral dissertation, University of Southern California.
Taylor, John Wesley (1986). Self-concept in home-schooling children. Doctoral dissertation, Andrews University, Berrien Springs, MI.
Van Galen, Jane A. (1986). Becoming home schoolers. Paper presented to the American Educational Studies Association, Pittsburgh, PA. (ERIC Reproduction Service NO. ED 284-325)
Van Galen, Jane A. (1987). Home schooling in context. Paper presented at the Annual Meeting of the American Educational Research Association, Washington, D.C. (ERIC Reproduction Service No. ED 284-324)

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply