An Exploratory Study of U. S. Pre-Service Teachers’ Beliefs About Homeschooling Outcomes

Yvonne Malone

Tennessee Technological University, YMalone@tntech.edu

 

Misty J. Cecil

Tennessee Technological University

Abstract

The authors investigated the perceptions of undergraduate college students who had been admitted to upper division classes in the teacher education program about homeschool outcomes. The purpose was to compare any belief differences about homeschooling outcomes based on gender, on the subjects’ chosen teaching level (elementary vs. high school), on their own schooling experience, and whether they had a friend or acquaintance who had been homeschooled. The study showed that the main predictor of homeschool opinion was participant history of homeschool. Participants who had been homeschooled for some time during their school career had positive beliefs about homeschool outcomes, while those who had not experienced homeschooling registered more negative beliefs about the outcomes.

Keywords: nontraditional education; parents as teachers; homeschooling legislation; statistics on homeschooling, homeschooling outcomes, beliefs about homeschooling.

The number of U. S. students schooled in the home continues to increase as the practice becomes mainstreamed. Statistics gathered by the National Center for Education Statistics (NCES) show that 1.5 million students (2.9 percent of school-age population) were homeschooled in 2007, compared to the estimate in 2003 of 1.1 million (2.2 percent of school-age population), and 1999’s estimate of 850,000 home-schooled students (1.7 percent of school-age population). (Bielick, 2008). The Department of Education estimates that the number of children educated at home is growing about 15% per year (Winters, 2000). The percentage of homeschooled students of the total school population remains fairly constant at 1.7 percent from first grade through twelfth grade (Bielick, Chandler, & Broughman, 2001), although Ray (as cited in Clark, 1994) found in his 1990 study that the number of homeschooled students levels off after fifth grade, as students return to traditional schools in middle and high school grades.

Some individuals, including educators, school administrators, and parents have concerns as to whether children schooled at home can be adequately educated by parents who have not been formally trained as teachers. Will their academic experience be less rigorous than children taught by government-certified teachers? In addition to academic standards, critics question whether homeschooled children will suffer from a lack of socialization opportunities (Ray, 2010). Public school teachers with a negative bias toward homeschooled students could have an adverse effect on the students’ public school experience. For students who entered public school from homeschool, the biggest difficulty in the transition was attributed to the reception by school personnel. Some parents felt that there was discrimination on the part of school officials when enrolling their children. The parents reported that principals and counselors expressed “skepticism” and were not helpful and supportive in the process (Koonce, 2007).

The current study sought to examine the effect of several factors (e.g., participant history of homeschool, friend history of homeschool, participants’ gender and their planned teaching level) on pre-service teachers’ attitudes toward homeschooling outcomes. A secondary purpose of the study was to investigate if similar relationships exist with attitudes toward public school, using public school as the dependent variable, and if findings differ with separate analyses filtering by gender, and by teaching level.

The importance of the study is justified by the growing popularity of the homeschool movement in the United States. The profile of homeschool families provided by surveys reveals that the typical family is white, conservative, middle SES, with parents who are better educated than the general population (Lines, 2000). However, Hammons (2001) suggests that the motivations of parents who homeschool their children range from “. . . conservative concerns about the values taught in public schools to more liberal worries that public schools stress conformity over creativity” (para. 2). He believes that the added emphasis on standardized testing may, in fact, prove to be the largest supplier of new converts to the practice. In fact, those who choose to homeschool are “. . . an eclectic mix of Christian fundamentalists, aging hippies, and inner-city minorities chastened by highly dysfunctional public schools” (Hammons, 2001, para. 2).

Homeschooling has only been legal in all 50 states since 1993. Each state supervises homeschool practices and this leads to varying degrees of regulation. Low regulation states (14) require notification of intent to homeschool. Moderate regulation states (20) require parents to send notification, test scores, and/or professional evaluation of student progress. High regulation states (6) require all of the above, plus other requirements such as curriculum approval by the state, teacher qualification of parents, or home visits by state officials. There are ten states with no notice requirements (Homeschooling, 2007).

Although minorities are currently underrepresented among those who homeschool, there may be a trend of increasing participation on their part in the homeschooling movement. A small (n=254) survey of college students at a two-year college and at a four-year private university revealed that while only 25 percent of white students said they would consider schooling their children at home, almost half of African-American students answered “yes” or “maybe” to the question, and two-thirds of other non-white minorities indicated the same (Lines, 2000). There are national support groups for a variety of minority groups, including Jews, Mormons, Catholics, Muslims, the handicapped, and persons of color (Lyman, 1998). Muslim Americans are the most rapidly growing sub-group to homeschool and their rate of participation is expected to double every year until 2009 (Bielick, et al., 2001). There are newsletters for Muslim and Jewish families who homeschool, and the practice is not confined to one type of neighborhood, but exists in urban, suburban and rural areas of the country (Clark, 1994).

Interestingly, homeschooling began as a liberal movement in the mid twentieth century. In the late 1950’s and early 1960’s a group of families believed that schools were too conservative, and, following the philosophy of educator John Holt (who published a critical review of the public school system in 1964 entitled How Children Fail), set out to provide an unstructured education that allowed the child flexibility in pursuing his own interests while being supported and encouraged by his parents and other adults (Lines, 2000).

It is a challenge to obtain an exact count of how many students participate in home-based education (it is not listed as a census question), but most researchers and educators agree that the practice is growing, rather than declining. Lines (2000) estimates that in some states the population may be growing as much as 15-20% a year, and that it could include 3 to 4 percent of school-aged children nationwide, although most researchers acknowledge that the limited data used to gauge the extent of the movement is often imperfect for differing reasons, not the least of which may be philosophical reluctance on the part of many homeschool parents to cooperate with researchers. Other challenges to research include convenience sampling that may not represent the population and imperfect data collection methodology.

One small study (n=69) found that, while 75% of respondents indicated their desire to include religious teachings as a reason for homeschooling their children, a larger number (98%) said they did it because they believed their children could achieve a higher level of academic achievement at home than in the public school, and 98% reported that they disliked the social influence of the peer group in the public school (Abell, 2002). This is not the only study that points to a shift away from religious reasons for homeschooling and toward dissatisfaction with the school curriculum and concerns about school safety. Lines (2000) cites the results of a Florida Department of Education survey sent to families who homeschooled for over a decade. “Religion” was cited as the main reason for the practice until 1995, but, since then, a variety of reasons have prevailed, including issues in the public school environment and dissatisfaction with the instructional program. Almost half (49 percent) of parents who chose to educate their children at home stated that their main reason for doing so was because they believed they could do a better job of it compared to the school system (Bielick, et al., 2001). Data from the National Household Education Surveys Program (NHES) in 2003 showed that 31% of parents surveyed said the main reason they homeschooled was because they were concerned about the environment of other schools (Princiotta, Bielick, & Chapman, 2004).

Meighan (1995) cites several advantages to homeschooling: education that is multi-dimensional and flexible, learning that focuses on and is directed by the student, and learning that is often more cooperative and democratic than learning in the traditional school. The informal environment that homeschooling provides allows “differentiated instruction,” not a one-size-fits-all version that is typical in public schools where teachers must meet the varied needs of twenty or more students in the classroom. The personal approach of schooling at home provides a natural environment to customize the curriculum for learning disabled and academically gifted children alike (Ray, 2002).

Ray (2000) makes the point that homeschooling is, in fact, home-based education. Although the parents are the primary decision makers about their children’s academic and social activities, there is often much learning and socialization that takes place outside of the home itself. The growth of the Internet has contributed to homeschooled students’ sense of community, allowing them to connect with other students schooled at home, and providing them with educational resources only dreamed of twenty years ago. Indeed, approximately 94% of home scholars have access to a computer, compared with about half of all American households combined (Basham, 2001). In the NHES 2003 survey, 41% of homeschooled students reported that they had engaged in some type of distance learning (Princiotta, Bielick, & Chapman, 2006). Many homeschooled students belong to a learning cooperative where they meet with other homeschooled students for specialized training.

In addition, public schools sometimes offer support for homeschooled students by providing them with curricular materials, allowing parents or students to use their facilities for meetings and/or to get information, and permitting students to participate in extracurricular activities and to attend classes (Bielick, et al., 2001). Parents make use of tutors for specialized subjects like foreign languages, or they often trade off teaching subjects with other families. There are learning cooperatives where their students can join others in biology labs, quilting bees, and other subjects. Home educators also increase their knowledge by means of homeschool conferences, networking, and internet chat rooms (Lyman, 2002).

Review of the Research

Wichers (2001) reviewed several studies to determine whether homeschooled students could compete academically in higher education. She looked at data from Ohio (ranked 5^th in number of homeschooled students) and at previous studies that considered academic achievement. In the Ohio data, homeschooled students scored on average between the 65^th and 85^th percentiles on a standardized achievement test administered by the Ohio Department of Education. Results from the 1994 Iowa Test of Basic Skills further showed that 80% of students schooled at home scored above the national average. “Ultimately, the Ohio report found that 53% of homeschooled children achieved individual scores in the top quarter percentile rank of the students who took the test” (p. 146).

Homeschooled students’ scores on the Iowa Test of Basic Skills and on the Stanford Achievement Test (both are standardized tests with national norms) are repeatedly as good as or better than those of traditionally schooled students. Also, the number of homeschooled students chosen as semi-finalists in the National Merit Scholarship Awards shows a steady increase, from 70 in 1998, to 137 in 1999, to 150 in 2000 (Lyman, 2002). Homeschooled students are also well represented in academic contests like the National Geography Bee and the National Spelling Bee. In addition, in a nationwide study that included a random sample of 1,516 families, Ray (1990) found that the median score on all areas of standardized tests for students in this study was at or above the 80^th percentile.

One of the largest studies conducted (Rudner, 1999) looked at the achievement of 20,760 homeschooled students. Some of the major findings were that median scores for every subtest at every grade were well above average when compared to students in either public, Catholic, or private schools, and the homeschooled students’ scores were typically in the 70-80^th percentile (compared to students who attended public school). In grades 1-4, homeschooled students performed, on average, one grade level above students their age who were in public and private schools; and, by the time they reached 8^th grade, homeschooled students were performing four years ahead of their counterparts in public and private schools. (In comparison, one researcher in a large school district estimated that 5% of their students are enrolled above grade level.) Based on these findings, it is not surprising that students who were exclusively schooled at home achieved higher academic scores on achievement tests than did those who spent some of their scholastic career in other educational programs. This is an interesting finding because the majority of homeschooled students enter other types of schooling at some point during their academic career.

Ray (1988) is quick to point out some cautions when interpreting these findings as proof of the educational superiority of homeschooling. First of all, when parents of students schooled in traditional settings are highly motivated and involved in their students’ education, there may be little difference in the achievement levels of the two groups. Also, the benefit of individual tutoring by an adult is incalculable and must surely play a role in differences in achievement levels of the two groups. Finally, most studies make use of standardized achievement tests to compare learning success, and these tests are considered by some to be biased in favor of white, middle-class people, the group that is most often seen in the homeschooled population.

Critics of homeschooling argue that the practice places the good of the individual above the good of society by removing children and social capital from the public schools, which weakens the ability and resolve of the public school to improve (Lubienski, 2000). The National Education Association (NEA) is the leading public teachers’ union and is opposed to the homeschooling movement. It has issued an official statement that says, in part, “. . .home schooling programs cannot provide the student with a comprehensive education experience” (para. 2) and it supports a resolution that only “persons licensed by the appropriate state education licensure agency” (para. 2) be allowed to teach children in the home (Wagner, 1997). Kathleen Lyons, a spokeswoman for the NEA, further clarifies the union’s position when she states, “Our society is loose enough as it is. The thing that binds us together in this country is public education” (“Homeschooling:” Issues and Controversies on File, Support for Homeschooling, 2000). Other professional educators discredit homeschooling, as evidenced in a comment by Thomas Shannon, Executive Director of the National School Boards Association, who remarked that homeschooling is “a giant step backward into the 17^th century” (Ramsey, as cited in Clark, 1994, para. 17).

Higher education has been more accepting of these students in the last ten years, as evidenced by the opening of Patrick Henry College in 2000, located in Purcellville, Virginia and specifically designed for homeschooled students. Other colleges (more than 200 at last count, according to Clark (1994), are enthusiastically welcoming homeschooled students, including elite schools like Stanford University, Wheaton College, and Harvard University (who has assigned an admissions officer to review homeschooled students’ applications (Winters, 2000). An admissions officer at Stanford University remarked that, “Homeschoolers bring certain skills – motivation, curiosity, the capacity to be responsible for their education – that high schools don’t induce very well” (Hammons, 2001, para. 21). College admissions personnel are impressed by the standardized test scores of homeschooled applicants whose SAT scores are, on average, 58 points higher than students schooled in the traditional manner. The homeschooled class of 2000 scored “an average of 1,100 on the SAT – a full 81 points above the national average – and 22.8 on the ACT, compared with the national average of 21” (Winters, 2000, para. 5).

A study was conducted by Jones and Gloeckner (2004) to examine university admission officers’ attitudes toward homeschooled students who applied for college admission. Overall, 73% of admissions officers expected those schooled at home to fare as well or better than those educated in traditional settings. The authors compare these findings with a study conducted by Barnebey (as cited in Jones & Gloeckner, 2004) which revealed a more negative attitude among college admissions officers, with 50% of them expecting homeschooled students to be less successful than students from traditional schools. Another startling difference between the two studies is that only 16.4% of the admissions officers in Jones and Gloeckner’s study recommended that students schooled at home attend a community college or junior college before enrolling in a four-year institution, while in Barnebey’s study, 65.5% of the admissions officers recommended that route for homeschooled applicants. In the more recent study, although most schools stated that they had an official policy for accepting homeschooled students, the researchers found that higher institutions of learning with student enrollments from 10,000-19,999, those in rural settings, church-affiliated institutions, state schools, and doctoral programs were the most amenable to accepting individuals schooled at home.

Although Ray (as cited in Basham, 2001), found that homeschooled children performed significantly better on tests of achievement when their parents are university graduates, he found no advantage on achievement scores for students whose parents had ever held a teaching certificate. Remarkably, children educated at home scored in the 80^th to 90^th percentile on achievement tests whether their mother held a college degree or had not completed high school. When the researcher looked at 8^th grade students’ math achievement in the public schools, he found that having college educated parents correlated with scores at the 63^rd percentile, while those students whose parents had not completed high schools scored at the 28^th percentile. Those students taught at home whose mother had not completed high school scored 55 percentile points higher than students in the public schools whose parents had a comparable education. When looking at parent education levels, Rudner (1999, Parent Education Levels section, para. 1) concludes that, “. . .at every grade level, the mean performance of homeschool students whose parents do not have a college degree is much higher than the mean performance of students in public schools.”

While many critics have conceded that students schooled at home are academically prepared as well or better than students in public schools, there is still the matter of whether they are as socially developed and well-adjusted as traditionally schooled youngsters. Public educators perceive that social interaction is missing in the homeschool environment and they cite this as one of their major objections to the practice of schooling at home. A survey of 115 educators found that more than 80 percent believed that the social development of the child suffered when they were homeschooled, and 59 percent believed that lack of competition socially and academically was a disadvantage of homeschooling Gorder (as cited in Clark, 1994).

Durkin (1995, p. 614) defines socialization as “the process whereby people acquire the rules of behavior and systems of beliefs and attitudes that equip a person to function effectively as a member of a particular society.” He describes the dilemma of agreeing upon what initially appears to be a basic concept:

. . . it [socialization] proves remarkably difficult to determine exactly how it occurs since it can never be observed directly. It is possible to monitor how people attempt to socialize others (e.g., how parents teach, guide, or correct their children) but the ways in which the recipients of such attentions develop their understanding and learn to regulate their own behavior can only be inferred from their observable responses. What is observable may provide only a partial insight into what is occurring within the organism. As a result, explaining how socialization comes about has been a fundamental source of controversy among theories of human social development. (p. 614)

Providing socialization opportunities for students has become more of a function of the schools than it previously was, and society has come to believe that the traditional school system can do a better job of socialization than parents who homeschool (Medlin, 2000). Nyberg and Egan (as cited in Medlin, 2000, p. 108) argue that public schools have “. . . been made responsible for an expanding range of socializing activities that previously were considered the proper roles of other social institutions, such as the family. . .”. Detractors of the homeschool movement argue that schooling children at home limits the opportunity for them to form social skills and may even inhibit their ability to become socially competent adults. In essence, these homeschool detractors believe that only traditional schools can teach “key social skills such as cooperation, respect for others, and self-control” (Medlin, 2000, p. 108). Mayberry (as cited in Medlin, 2000) reported on a survey which found that 92% of public school superintendents believed that schooling students in the home would not provide them with the experiences they needed to be adequately socialized.

Homeschooling parents invoke a variety of sources to expand their academic resources, and they also rely on available social networks in the public schools and the community. Lawsuits have resulted in some school districts allowing homeschooled children to participate in extracurricular activities such as football and band (Kantrowitz & Wingert, 1998), and Oregon, Iowa and California allow these students to participate in a variety of opportunities offered by the public schools (Clark, 1994). Other public schools provide resources and instructional support for homeschooling parents and/or allow homeschooled students to take classes (Lines, 2001).

The community provides an abundance of socializing opportunities for homeschooled students, from organized sports and 4-H clubs to volunteer activities and scouts. Other activities include book clubs, Little League teams, charity projects, and field trips organized by groups of homeschooling parents (Clark, 1994). Delahooke (as cited in Medlin, 2000) found that homeschooled children participate in more activities than those who attend public schools, and the researcher concluded that this could be the result of flexible scheduling and the more efficient use of time by parents who homeschool. Medlin (2000) speculates that a finding by Rudner (1999) that public school children watch, on average, considerably more television than homeschooled children, may account for this greater participation by home-schooled children in extracurricular activities. In addition, Montgomery (as cited in Medlin, 2000) found that parents who homeschool consciously seek activities for their children that will develop their leadership abilities.

While students in the public schools are segregated by age, home-schooled students may have more opportunities to interact with children of varying ages and with adults. Medlin’s 1998 study (as cited in Medlin, 2000) sought to determine the amount of diversity in home-schooled students’ social contacts. Parents reported that their children had “close relationships” with adults (including the elderly), and with other children who attended conventional schooling, and that they had “moderately close” relationships with individuals from socioeconomic, ethnic and religious backgrounds that differed from their own.

Research has been mixed on whether homeschooled students feel they are at a disadvantage when it comes to forming social relationships. Shirkey (as cited in Medlin, 2000) asked children ages 6 to 13 to list disadvantages of homeschooling and concluded that they felt they had few friends. However, Montgomery (as cited in Medlin, 2000) reported that only 2 of 87 homeschooled students listed “having few friends” as a concern, and Mullins (as cited in Medlin, 2000) concluded that students were satisfied with the socialization they experienced as homeschooled students, especially if they were involved in the decision to homeschool.

Research that looks at social development differences between homeschooled students and those in traditional schools has consistently yielded more positive outcomes for the former group. Two separate studies, one by Stough and one by Smedley (both cited in Medlin, 2000) used results from the Vineland Adaptive Behavior Scales to compare social development in the two groups. Stough concluded there was no significant difference in the two groups, and further stated that “insofar as self concept is a reflector of socialization, it would appear that few home-schooled children are socially deprived, and that there may be sufficient evidence to indicate that some home-schooled children have a higher self-concept than conventionally schooled children” (as cited in Aiex, 1994, para. 11). Another researcher reported higher scores for homeschooled students on the test’s subscales of communication, daily living skills, socialization and social maturity, and a mean score for the homeschooled group at the 84^th percentile, while the matched sample of traditionally schooled children had a mean score at the 23^rd percentile. This led the researcher to deduce that “children kept home are more mature and better socialized than those who are sent to school.” Smedley (as cited in Medlin, 2000, p. 114). Smedley further postulated that:

In the public school system, children are socialized horizontally, and temporarily, into conformity with their immediate peers. Home educators seek to socialize their children vertically, toward responsibility, service, and adulthood, with an eye for eternity (as cited in Klicka, 2002, para. 12).

Finally, Lee’s 1994 study (as cited in Medlin, 2000, p. 114) using the Vineland Adaptive Behavior Scales yielded similar results as Smedley’s study, with higher scores for the homeschooled group on the subscales of family and community, leading Lee to conclude that, “socialization of children in homeschools is effective without exposure to large groups of children. …Homeschool parents are imparting positive family socialization, which is not inferior to the public school culture.”

There are a number of studies that have investigated different aspects of social and emotional adjustment of homeschooled individuals. Shyers’ 1992 study (as cited in Medlin, 2000) compared two groups of students (n=140) aged 8-10 years. Half of the students had been exclusively homeschooled, and half had attended only conventional schools. The two groups were matched by age, race, gender, socioeconomic status, and number and frequency of extracurricular activities. Shyers looked at self-concept, assertiveness and behavior. There were no significant differences between the two groups on the first two measures, but, on the third, Shyers trained observers to rate the subjects as they played and worked together, using the Direct Observation Form of the Child Behavior Checklist developed by Achenbach and Edelbrock in1983 (as cited in Medlin, 2000). (It was a blind study on the part of the observers because they were not told which students were homeschooled and which were traditionally schooled.) The outcome was dramatic: The children who were traditionally schooled had a mean score on problem behaviors that was eight times higher than the homeschooled children’s mean score.

Knowles (as cited in Klicka, 2002) looked at adults who had been homeschooled and found that they were not socially deficient as evidenced by several factors: two-thirds were married (the norm for adults their age), all were employed, and none were on welfare. Almost all of the subjects said they would choose to be homeschooled again. Some of the advantages they mentioned included a close relationship with their parents and the individualized instruction that is unique to homeschooling, and three fourths said that they credited homeschooling with equipping them to relate to individuals from all walks of life. The researcher was struck by the fact that almost two thirds of the subjects were self-employed and Knowles saw this as evidence for “. . .the contention that homeschooling tends to enhance a person’s self-reliance and independence” (para. 16)

Ray (1988) reviewed several studies that looked at the affective domain of homeschooled individuals. Among them was a study conducted by Taylor, who found that the self-concept of these students was significantly higher when compared to conventionally schooled youth; another by Wartes, who concluded that homeschooled students are not socially deprived; and a third study by Delahooke, whose research showed that homeschooled children scored in the “well adjusted” range of the Roberts Apperception Test for Children (similar to children taught in private schools.) In addition, Delahooke’s study (as cited in Ray, 1988) found that homeschooled children were not as dependent on peer influences as children educated in traditional schools. Finally, Taylor (as cited in Aiex, 1994; and in Medlin, 2000) had more than 220 children complete the Piers-Harris Children’s Self-Concept Scale. Their mean scores were significantly above the test norms.

The Current Study

While there has been research regarding the beliefs of college admissions officers (Jones & Gloeckner, 2004) and public educators (Gorder, as cited in Clark, 1994) about students who are homeschooled, there is a scarcity of research on attitudes of pre-service teachers. Some students are homeschooled for their entire school career, but others enter the traditional school system sometime during their school career (Ray, 1994). It would be advantageous to know if teachers hold erroneous beliefs about homeschooling that would be detrimental to the students’ public school experience.

In addition, because the National Education Association (NEA), the country’s largest teachers’ union, is undeniably critical of the practice of homeschooling (Homeschooling, 2000), it is safe to assume that young, impressionable college students preparing to become teachers may be swayed by the principles and policies of the organization that influences their trade. The NEA’s negative attitude toward homeschooling could adversely affect teachers’ behaviors toward, and expectations of, their students who had been formerly schooled at home.

 

Procedures

The study was conducted at a mid-sized public university in the southeastern United States. The subjects were upperclassmen who were enrolled in a required testing course in the teacher education program and who volunteered to participate in the study. The only identifiers of the subjects were gender, whether they would teach at the elementary or high school level, whether they were homeschooled, and, if so, for how many years, and if they had a friend or acquaintance who was homeschooled. The instrument contained both positive and negative statements about homeschooling and traditional schooling, and the students were asked to indicate their agreement or disagreement using a 5-point Likert scale rating that that ranged from (strongly disagree) to (strongly agree). Examples of the questions were “Homeschooled children have poorer social skills than students who attend public schools,” “Children who attend public school will be more likely to attend the college of his/her choice,” “Children who attend public schools watch more television at home than those who are homeschooled,” and “A child will benefit more academically from homeschooling than from public schooling.” (see Appendix).

Scoring

A homeschool composite scale (Homeschool Scale) was created using the sum of items one, three, six, seven, eight, twelve and fourteen of the survey (see Appendix). The items included in the homeschool scale were positive statements about homeschool, and a high score on this scale indicates a high opinion of homeschool. A public school composite scale (Public School Scale) was created using the sum of items two, four, five, nine, ten, eleven, and thirteen of the survey (see Appendix). The items included in the public school scale were positive statements about public school, and a high score on this scale indicates a high opinion of public school.

Analysis of Data

The design of this study was descriptive-correlational. The data were analyzed using multiple linear regression analyses. The backward method was used on each data set initially, then the enter method was used to determine the unique statistical contribution of each variable in its relationship with the dependent measure(s).

Results

Opinion about Homeschool

A multiple linear regression analysis was performed using participant history of homeschool, friend history of homeschool, gender, and future teaching level to predict opinion about homeschool (Homeschool Scale). The overall regression was significant, R =.17, F(1, 311) = 9.03, p<.05. Participant history of homeschool was found to be the only significant predictor of homeschool opinion. The equation explained 2.8% of the variability in opinion about homeschool. The best model for predicting opinion about homeschool is: Homeschool Scale = 5.64 (Participant History of Homeschool) + 21.86.

Filtered by Gender – Males

A multiple linear regression analysis was performed on the data from the male participants using participant history of homeschool, friend history of homeschool, and future teaching level to predict opinion about homeschool (Homeschool Scale). The overall regression was significant, R =.20, F(1, 76) = 3.31, p<.05. Participant history of homeschool was found to be the only significant predictor of homeschool opinion among the male participants. The equation explained 4.2% of the variability in opinion about homeschool. The best model for predicting opinion about homeschool for the male participants is: Homeschool Scale = 4.25 (Participant History of Homeschool) + 23.55.

Filtered by Gender – Females

A multiple linear regression analysis was performed on the data from the female participants using participant history of homeschool, friend history of homeschool, and future teaching level to predict opinion about homeschool (Homeschool Scale). The overall regression was significant, R =.17, F(1, 232) = 6.88, p<.05. Participant history of homeschool was found to be the only significant predictor of homeschool opinion among the female participants. The equation explained 2.9% of the variability in opinion about homeschool. The best model for predicting opinion about homeschool for the female participants is: Homeschool Scale = 5.64 (Participant History of Homeschool) + 21.86.

Filtered by Level – Elementary School

A multiple linear regression analysis was performed on the data from the participants planning to teach elementary school using participant history of homeschool, friend history of homeschool, and gender to predict opinion about homeschool (Homeschool Scale). The overall regression was significant, R =.21, F(1, 173) = 7.93, p<.05. Participant history of homeschool was found to be the only significant predictor of homeschool opinion among those students who plan to teach elementary school. The equation explained 4.4% of the variability in opinion about homeschool. The best model for predicting opinion about homeschool for the participants planning to teach elementary school is: Homeschool Scale = 10.59 (Participant History of Homeschool) + 21.41.

Filtered by Level – High School

A multiple linear regression analysis was performed on the data from the participants planning to teach high school using participant history of homeschool, friend history of homeschool, and gender to predict opinion about homeschool (Homeschool Scale. The overall regression was significant, R =.20, F(1, 134) = 5.68, p<.05. Gender was found to be the only significant predictor of homeschool opinion among those students who plan to teach high school. The equation explained 4.1% of the variability in opinion about homeschool. The best model for predicting opinion about homeschool for the participants planning to teach high school is: Homeschool Scale = -2.12 (Gender) + 23.91.

Opinion about Public School

A multiple linear regression analysis was performed using participant history of homeschool, friend history of homeschool, gender, and future teaching level to predict opinion about public school (Public School Scale). The overall regression was significant, R =.19, F(1, 310) = 11.29, p<.05. Participant history of homeschool was found to be significant, and friend history of homeschool approached level of significance in predicting public school opinion. The equation explained 3.5% of the variability in opinion about public school. The best model for predicting opinion about public school is: Public School Scale = -6.30 (Participant History of Homeschool) + 27.55.

Filtered by Gender – Males

A multiple linear regression analysis was performed on the data from the male participants using participant history of homeschool, friend history of homeschool, and future teaching level to predict opinion about public school (Public School Scale). Participant history of homeschool approached level of significance (p<.05). The enter method was performed by entering only participant history of homeschool to predict the dependent variable and the variable was found to be non-significant in predicting opinion about public school.

Filtered by Gender – Females

A multiple linear regression analysis was performed on the data from the female participants using participant history of homeschool, friend history of homeschool, and future teaching level to predict opinion about public school (Public School Scale). The overall regression was significant, R =.26, F(2, 230) = 8.61, p<.05. Participant history of homeschool and friend history of homeschool were both found to be significant predictors of public school opinion among female participants. The equation explained 7.0% of the variability in opinion about public school. The best model for predicting opinion about public school for female participants is: Public School Scale = -8.16 (Participant History of Homeschool) – 1.58 (Friend History of Homeschool) + 27.55.

Filtered by Level – Elementary School

A multiple linear regression analysis was performed on the data from the participants planning to teach elementary school using participant history of homeschool, friend history of homeschool, and gender to predict opinion about public school (Public School Scale). The overall regression was significant, R =.21, F(1,173) = 7.88, p<.05. Participant history of homeschool was found to be the only significant predictor of public school opinion among those students who plan to teach elementary school. The equation explained 4.4% of the variability in opinion about public school. The best model for predicting opinion about public school for the participants planning to teach elementary school is: Public School Scale = -10.14 (Participant History of Homeschool) + 27.64.

Filtered by Level – High School

A multiple linear regression analysis was performed on the data from the participants planning to teach high school using participant history of homeschool, friend history of homeschool, and gender to predict opinion about public school (Public School Scale. The overall regression was significant, R =.18, F(1,135) = 4.71, p<.05. Participant history of homeschool was found to be the only significant predictor of public school opinion among those students who plan to teach high school. The equation explained 3.4% of the variability in opinion about public school. The best model for predicting opinion about public school for the participants planning to teach high school is: Public School Scale = -4.93 (Participant History of Homeschool) + 27.43.

Discussion

Participant history of homeschool was found to be a significant predictor of the dependent variable, opinion about homeschool. Additionally, participant history of homeschool showed a positive relationship with opinion about homeschool. In other words, the participants who were homeschooled had high scores on the homeschool scale, indicating that those who were homeschooled had a high opinion of homeschool.

Separate analyses were performed for the male participants and the female participants. The results for each gender revealed the same statistical relationships as the overall results; those who were homeschooled had a high opinion of homeschool. The separate analysis for the participants planning to teach elementary school revealed the same relationship as the overall results, as well. The separate analysis for the participants planning to teach high school showed a significant negative relationship between gender and opinion about homeschool. This finding indicated that males who plan to teach high school had a higher opinion of homeschool than the females who plan to teach high school.

Participant history of homeschool was found to be a significant predictor of the dependent variable, opinion about public school. Additionally, participant history of homeschool showed a negative relationship with opinion about public school. In other words, the participants who were not homeschooled had high scores on the public school scale, indicating that those who were not homeschooled had a high opinion of public school.

The separate analysis for the female participants found that participant history of homeschool, and friend history of homeschool were both significant predictors of opinion about public school. Both factors had a negative relationship with the dependent variable, opinion about public school. This could be interpreted that the female participants who were not homeschooled and/or did not have friends who were homeschooled had a higher opinion of public school than those who were homeschooled and/or had a friend who was homeschooled.

Separate analyses were performed for the participants planning to teach elementary school and the participants planning to teach high school. The results for each teaching level revealed the same statistical relationships as the overall results; those who were not homeschooled showed a higher opinion of public school than those who were homeschooled.

The purpose of this study was to explore the relationships between the factors participant history of homeschool, friend history of homeschool, gender, and future teaching level to predict the dependent variable, opinion about homeschool (Homeschool Scale) among pre-service teachers. The secondary purpose of this study was to investigate if similar relationships exist with opinion about public school as the dependent variable. Another goal of this study was to determine whether the findings would differ with separate analyses filtering by gender, and by teaching level. The results revealed that the most significant factor in predicting either opinion about homeschool or opinion about public school is personal experience with homeschool. Similar relationships were, in fact, found when the data was filtered by gender and by level.

One implication of this study is that education programs for pre-service teachers should emphasize teaching the future teachers to understand and accept students who come from different educational backgrounds/settings. It is inevitable that personal experience will affect opinion, but teachers should strive to remain neutral in regards to students’ educational background. Teacher education programs should stress the importance of eliminating biases held against different educational backgrounds, such as homeschool education.

This study revealed that pre-service teachers’ own educational background is significantly related to their opinion about public school and homeschool. In other words, if the participants attended public school they had a higher opinion of public school than homeschool. Conversely, if they were educated in the home, they had a higher opinion of homeschool than public school. Differing backgrounds and opinions can often lead to misunderstandings, misconceptions, and biases. Further research can be done in this area to investigate whether different educational backgrounds and opinions are related to biases about homeschool and public school. Future research could use scenarios describing students from various educational backgrounds and ask the participants a series of questions to investigate bias. It would be advantageous for future research to use the experimental method with a somewhat equal number of participants having public school experience and homeschool experience.

Certain limitations should be noted. This research study was exploratory in nature, and investigated opinions about home education and public school education outcomes. It was not intended to explore causal relationships. For that reason the research was designed as a descriptive-correlational study rather than as an experiment. In order to find definite cause and effect relationships, a controlled experiment should be performed in future studies.

References

Abell, Deborah (2002). Homeschooling: An overview. Institute for Regional Analysis and Public Policy. Retrieved from http://irapp.moreheadstate.edu/cerl/pdf/cerl_5.pdf

Aiex, Nola Kortner (1994). Homeschooling and socialization of children. ERIC Digest. Retrieved from http://www.eric.ed.gov/PDFS/ED372460.pdf

Basham, Patrick (2001). Homeschooling: From the extreme to the mainstream? The Fraser Institute, 51. Retrieved from http://www.fraserinstitute.org/workarea/DownloadAsset.aspx?id=2952

Bielick, Stacey (2008). 1.5 million homeschooled students in the United States in 2007. National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2009030

Bielick, Stacey, Chandler, Kathryn, & Broughman, Stephen P. (2001). Homeschooling in the United States: 1999. National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubs2001/2001033.pdf

Clark, Charles S. (1994). Home schooling: Is it a healthy alternative to public education? CQ Researcher, 4(33). 769-792.

Durkin, Kevin (1995). Socialization. In The Blackwell encyclopedia of social psychology (pp. 614-618). Cambridge, MA: Basil Blackwell.

Hammons, Christopher W. (2001). School at home. Education Next. Retrieved from http://educationnext.org/files/ednext20014_48.pdf

Home Schooling. (2000). Issues and Controversies on File. Retrieved from http://2facts.es.vrc.scoolaid.net/icof_story.aspx?PIN=i0201120&bts=1

Homeschooling. (2007). Iowa Civic Analysis Network. Retrieved from http://www.uiowa.edu/~ican/Papers%202007/Homeschooling.pdf

Jones, Paul, & Gloeckner, Gene (2004). A study of admission officers’ perceptions of and attitudes toward homeschool students. The Journal of College Admission, 185, 12-21.

Kantrowitz, Barbara, & Wingert, Pat (1998). Homeschooling: More than a million kids and growing: Can it work for your family? Newsweek, 64-70.

Klicka, Chris (2007). Socialization: Homeschoolers are in the real world. Homeschool Legal Defense Association. Retrieved from http://www.hslda.org/docs/nche/000000/00000068.asp

Koonce, Jeffrey B. (2007). The transitional experience of home-schooled students entering public education: How can public schools better serve the home-schooled student’s transition to public education? (Doctoral dissertation). Retrieved from https://mospace.umsystem.edu/xmlui/bitstream/handle/10355/4777/short.pdf?sequence=2

Lines, Patricia M. (2000). Homeschooling comes of age. The Public Interest, 140, 74-85.

Lines, Patricia M. (2001). Homeschooling. ERIC Digest, (151). Retrieved from http://scholarsbank.uoregon.edu/jspui/bitstream/1794/3374/1/digest151.pdf

Lubienski, Chris (2000). Whither the common good? A critique of homeschooling. Peabody Journal of Education, 75(1&2), 207-232.

Lyman, Isabel (2002). Answers to homeschool questions: Knowledgeable answers to seven frequently asked questions about homeschooling explain the benefits of home education and dispel some typical misconceptions. The New American, 18(9), 31-34.

Lyman, Isabel (1998). Homeschooling: Back to the future? Cato Institute. Retrieved from http://www.cato.org/pubs/pas/pa-294.html

Meighan, Roland (1995). Home-based education effectiveness research and some of its implications. Educational Review, 47 (3), 275-286.

Medlin, Richard G. (2000). Homeschooling and the question of socialization. Peabody Journal of Education, 75(1&2), 107-123.

Nyberg, David, & Egan, Kieran (1981). Socialization and the schools. New York: Teachers College Press.

Princiotta, Daniel, Bielick, Stacey, & Chapman, Christopher (2004). 1.1 Million homeschooled students in the United States in 2003. National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubs2004/2004115.pdf

Princiotta, Daniel, Bielick, Stacey, & Chapman, Christopher (2006). Homeschooling in the United States: 2003. National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubs2006/2006042.pdf

Ray, Brian D. (1988). Homeschools: A synthesis of research on characteristics and learner outcomes. Education and Urban Society, 21(1), 16-31.

Ray, Brian D. (1990). A nationwide study of home education: Family characteristics, legal matters, and student achievement. Salem, OR: National Home Education Research Institute.

Ray, Brian D. (2000). Homeschooling for individual’s gain and society’s common good. Peabody Journal of Education, 75(1&2), 272-293.

Ray, Brian D. (2002). Customization through homeschooling. Educational Leadership, 50-54.

Ray, Brian D. (2010). Academic achievement and demographic traits of homeschool students: A nationwide study. Academic Leadership Live: An online journal. 8(1). Retrieved from http://www.academicleadership.org/cgi/article/print/Academic_Achievement_and_Demographic_Traits_of_Homeschool_Students_A_Nationwide_Study

Romanowski, Michael H. (2001). Common arguments about the strengths and limitations of home schooling. The Clearing House, 79-83. Retrieved from http://bhssnell.weebly.com/uploads/5/0/9/8/5098492/9_3_romanowski.pdf

Rudner, Lawrence M. (1999). Scholastic achievement and demographic characteristics of homeschool students in 1998. Education Policy Analysis Archives, 7(8). Retrieved from http://epaa.asu.edu/ojs/article/view/543

Wagner, David (1997). No place (to learn) like homeschool. Insight on the News, 13(33), 12-13.

Wichers, Michelle (2001). Homeschooling: Adventitious or detrimental for proficiency in higher education. Education, 122(1), 145-150.

Winters, Rebecca (2000). From home to Harvard: Homeschooled kids have earned a college of their own–and admission to elite, traditional campuses. Time, 156(11), 55. ¯

APPENDIX

I was homeschooled.                    ______yes          _______no         If yes, how long? _______

I have a close friend (or acquaintance) who was home-schooled._____yes           ______no

Male______                               Female______                (your gender)

I plan to teach                 Elementary_____ High School______

	Strongly Disagree	Disagree	Slightly Disagree	Slightly Agree	Agree	Strongly Agree
A child will benefit more academically from homeschooling than from public schooling.
The public school system provides a child with more opportunities for social interaction.
A home-schooled child will become a more independent adult.
Children who attend public school will be more likely to attend the college of his/her choice.
Home-schooled children have fewer friends, resulting in lower self-esteem.
Children who attend public school are more likely to be exposed to drugs.
Home-schooled children are less likely to become sexually active at a young age.
Children who attend public school are more violent than those who are homeschooled.
Home-schooled children are at a disadvantage when competing for jobs in the work force after high school.
A child who attends public school is more likely to be involved in extra-curricular activities in college.
Children who are homeschooled have poorer social skills than students who attend public schools.
Children who attend public school watch more television at home than those who are homeschooled.
Home-schooled children have fewer resources (e.g., technology) for learning than those in public schools.
If it is economically feasible, I would like to homeschool my child.