The Effects of School Grade Span Configuration on Student Achievement in Middle School-Aged Children

Grade Span Configurations.Grade span configuration refers to the grade range within a school building. Numerous grade span configurations exist among schools and grades are grouped together for various reasons. Typical variations in grade spans include either kindergarten (K) or prekindergarten (PK) through fifth or sixth grade, with a transition into middle school through eighth or ninth grade with the final transition into high school through the 12th grade. Other grade span configurations that are prevalent include K-8 with one transition to 9-12, K-6 with one transition to 7-12, and one setting with no transition throughout the entire educational span (PK/K-12). There are even more variations of these configurations, some including more or fewer transitions throughout a student’s school career (Tennessee State Department of Education, 2013a). Many factors must be considered when deciding how to configure grade spans. Some of these considerations include: number of students and the cost of transportation, socioeconomic status of the student population, school system goals for student achievement, number of transitions for affected students, parental involvement and budget constraints (Herman, 2004). The number of students varies greatly in rural and suburban areas. Although rural school districts usually cover a greater geographic area, they may still only have enough students to populate one school plant. Therefore, it may make more sense financially to have a larger grade span in a single school whereas suburban areas are not only more densely populated, but may actually have more than one school within a relatively small geographic area. Therefore, it is not uncommon to find multiple schools within those districts. The cost of transportation would vary greatly between these two areas as the denser the population, the closer the school location and the less transportation costs for the schools. Although the number of transitions have been shown to have an effect on student achievement, many times, transition is not a heavily weighted factor in grade configuration as some school systems have multiple transitions throughout a student’s academic career. Although there may be many reasons for selecting grade configuration within a given school, it would seem, in an ideal scenario, that the single-most important factor should be student achievement and which configuration best fits this goal.

K-8 Configurations. Research by Offenberg (2001) examined the efficacy of K-8 schools as compared to middle schools in Philadelphia. At the outset of Offenberg’s (2001) study, it had already been identified that K-8 schools in Philadelphia outperformed middle schools academically. The explanation at that time was K-8 schools did better because more middle class students enrolled in K-8 schools. Offenberg wanted to identify whether the higher academic achievement of K-8 schools was due to social class or school environment; therefore, every achievement model tested within the study controlled for socioeconomic status. Offenberg found that while social class is an important predictor of achievement, once statistically controlled, K-8 achievement scores were still significantly higher than scores from comparable students in middle grade only arrangements, suggesting social class is not the only reason why these two environments differ. The researcher found further that the percent of minorities attending the schools did not affect the findings. Additionally, when Philadelphia’s “Children Achieving” reform was implemented, the gains in K-8 schools were higher than those in middle schools. Another interesting finding was a strong negative relationship between achievement and school size, particularly in eighth-grade enrollment. The K-8 schools were typically smaller and there were fewer students per grade. Therefore, although the argument for K-8 schools is strong, it was clear that school size, or possibly class size, also appears to play a significant role in facilitating achievement gains. These findings were supported in earlier research by Lee and Loeb (2000) who found student achievement was higher and teachers’ attitudes were more positive about their responsibility for students’ learning in smaller schools (less than 400). Consequently, Offen-berg’s study concluded school size impacted student achievement both directly and indirectly.

In a disquisition in 2002, Pardini reported that Cleveland middle schools which had been reconfigured into K-8 schools demonstrated better attendance and achievement scores from sixth graders than those still attending middle school. Many in this discussion argued that the middle schools were not successful because they had not completely implemented the middle-level programs, but had merely changed the name of the school and reconfigured the grades. Supporters of the K-8 schools contended that the middle-level program can be implemented in the K-8 schools and that K-8 schools do a better job at implementing the middle-level program due to the support of the constant environment a K-8 school offers.

Middle School Configurations: Concept Versus Environment. The middle school concept is a set of principles and practices developed after numerous studies of middle schools were conducted following a myriad of criticism by both policymakers and public. Although the middle grade concept has been in existence for more than a half century, current definitions of the concept are typically derived from two primary sources: Carnegie Council on Adolescent Development’s Turning Points: Preparing Youth for the 21st Century (1989) and the National Middle School Association policy: This We Believe: Successful Schools for Young Adolescents (2003). Based on these sources:

In the 1960s and 1970s when overcrowding became an issue, school districts found it to be more cost-efficient to move fifth and sixth graders to the middle school than to build more elementary schools. One of the problems with this is that middle schools were often housed in former junior high school buildings, and in some cases old high school buildings. Junior high school buildings were designed to organize teachers by subjects. Whereas this was not an issue for the middle-school concept for the subjects of language arts, social studies, math, and reading, the implementation of the middle-school concept did pose a major problem where science was involved. Science labs were often together, in one area of the building, which made it more difficult for science teachers to participate in small teaching teams with collaboration through proximity nearly impossible (George, 1988).

During the early implementation of middle schools, there was also much attention on school desegregation and racial equality. By rearranging the grade configuration into middle school systems, ninth grades were moved to the high schools and fifth and sixth grades were moved out of neighborhood K-8 schools, which increased the amount of desegregation possible across multiple school grades. So, the application of the middle school concept answered a societal issue as well. Unfortunately, in many locales, desegregation only occurred at the middle school level, leaving segregation in both elementary and high schools largely intact. Despite attempts to desegregate middle grade schools, inadequate integration of the middle-level programs persisted for various reasons, which eventually led school officials back to the concept of K-8 schools (George, 1988).

Many of those in favor of the middle school option feel separating middle school-aged children from elementary school-aged children allows the middle school faculty, staff, and administration to better focus on the needs of young adolescents both academically and emotionally. Advocates feel the middle school concept is more developmentally appropriate for students (Weiss & Kipnes, 2006) and fosters social ties among students and teachers while utilizing the strengths of teachers with different specialties (Alexander & Williams, 1965). One goal of the middle school was to help students make the transition from elementary school to high school. As it was later seen, many schools were called middle schools, but did not implement the middle school concept (Mac Iver & Epstein, 1992) making this goal difficult to achieve.

Middle schools typically involve interdisciplinary teaching teams who can work collabo-ratively to focus the developmental needs of the children served by the team (Jenkins & McEwin, 1992). Davis (1995) defined team teaching as “two or more faculty in some collaboration in the planning and delivery of a course” (p. 7). This idea is frequently attributed to W. M. Alexander and was intended to transition children from a single teacher in elementary school to multiple teachers in high school by having two to four teachers work together with a group of students. Alexander’s idea was that this transition with a few teachers would allow the students to still feel connected to an extent, have a sense of community, and reduce the chances of students feeling lost when they reached high school (Alexander, 1995). In 1993, research by Lee and Smith examined the link between the implementation of middle school components and student achievement. These researchers concluded that in order for the middle school concept to be implemented, certain elements must be present. The middle schools must reduce or eliminate departmental structure, there must be heterogeneously grouped instruction, and there must be team teaching. According to Davis (1995), not only does team teaching involve two or more teachers collaborating to plan, conduct, and evaluate lessons to the same group of students, but the teachers must also be involved in almost every aspect of the lessons throughout the course. This lends support to Alexander’s original definition as opposed to the “trade off” variation some schools use. Team teaching is one of the most important aspects of the middle school model; however, the interpretation of the collaboration concept seems to vary. Some researchers contend this concept can be implemented effectively whether the environment is middle school or K-8 (Anfara & Lipka, 2003; Mac Iver & Epstein, 1991).

In 2008, Cook et al. concluded that it was better to separate sixth graders from older students in an effort to decrease discipline problems in that age group, as exposure to their older counterparts appeared to contribute to some of the discipline issues. This opens the question as to where the cut-off for middle school should be, which supports research by Jenkins & McEwin (1992) suggesting that young adolescents should be separated from both younger and older peers to accommodate their developmental needs. Even though research shows achievement declines upon entering middle school (Combs et al., 2011; Pardini, 2002), proponents of the separate middle grade school contend that declines may be due to higher expectations of the teachers (Whitely et al., 2007).

Percent Minority. Studies have long demonstrated an achievement gap between African American students and their non-Hispanic White counterparts (Borg, Borg, & Stra-nahan, 2012; U.S. Department of Education, 2009). The achievement gap between these two groups has given rise to focusing on how to improve the learning experience for African American populations. Some of the factors that appear to have an effect on African Americans’ achievement include socioeconomic status, parents, community, and environment (Kuykundall, 1989) as well as safety issues (Lacoe, 2012). Research by Combs et al. (2011) controlled for parental involvement and socioeconomic status. The researchers still found a difference between achievement levels of middle-school students versus their K-8 counterparts. Research by Offenberg (2001) also controlled for percent minority and found that it did not change the conclusions that middle grade students in K-8 schools outperform their grade level peers in middle schools. Kuykundall (1989) emphasized the significance of improving the school environment by putting programs in place to improve Black students’ achievement by enhancing students’ self-image. Further, Lacoe (2012) found “that school safety may be a contributing factor to racial … achievement gaps” (p. 2). Although children are a product of the community in which they live, research by Simmons and Blythe (as cited in Offenberg, 2001) suggested that having numerous years in the same school environment improves students’ sense of belonging as well as self-esteem and may counteract negative influences in the community outside the school. Other efforts to improve student achievement, such as older students taking pride in their community and feeling protective of the younger students which were also discussed in this study, could potentially address safety.

Socioeconomic Status. “Researchers examining student performance consistently find that one of the most important influences on student achievement is socioeconomic status (SES) of students” (Tajalli & Opheim,

2004, p. 44). When examining high- and low-poverty schools, many times high-poverty schools are found to be predominately minority in racial makeup and low-poverty schools are found to be predominately White, in their racial makeup. Although some contend this combination (i.e., high minority and high poverty) has a consistent negative effect on performance, several studies have examined both percent minority and SES and the findings still demonstrated higher achievement in the K-8 environment (Bromberg & Theokas, 2013; Offenberg, 2001).

As previously mentioned, SES was found to be an important factor in predicting achievement in research conducted by Offenberg (2001). Offenberg wanted to see if controlling for SES would mediate the outcome of higher performing K-8 schools vs their lower performing middle school counterparts. The results demonstrated that each 1% increase in proportion of low-income students reduced the school’s average scores between 0.27 and 0.41 points on the NCE Stanford Achievement Tests, suggesting that SES is not a significant factor in student achievement between the two environments. Research by Bromberg and Theokas (2013) not only examined the minority and SES achievement gap, but also focuses on how to close said gap by pushing students of all color and SES to higher levels of achievement. Borg et al. (2012) also examined the achievement gap between high-poverty and low-poverty schools and found that while school resources (i.e., quality of teachers and smaller class sizes) affected student achievement, these factors were not as important predictors as student background variables, including household income, race, and gender.

School Size/Class Size. Since the 1940s, school size has been a concern in studies of student achievement (Cotton, 1996). As the need for schools grew, so did class and school size. While there is no definite dividing line between small and large schools, on average, the most effective size for an elementary school is 300-400 students and 400-800 for middle school (Williams, 1990). Two studies using Texas schools found statistically significant relationships between class size and student achievement. The first study by Ferguson (1991) found a statistically significant relationship between teacher quality, defined as “good teachers [having] distinguishable impacts on students’ exam scores—effects that are separate from those, for example, of well-educated parents” (Ferguson, 1991) and class size in first through seventh grade. This study demonstrated for every student above an 18:1 student-teacher ratio, student achievement fell. The second study by Wenglinsky (1997) examined data in over 200 Texas districts and found that smaller class size had a positive effect on fourth grade math as well as improved social environment for eighth graders, which also improved achievement. These effects were greatest in low SES students. These findings were corroborated by Molnar et al. (1999) who examined the Student Achievement Guarantee in Education (SAGE) program in Wisconsin. This program was similar to the Student/Teacher Assessment Ratio (STAR) in Tennessee that examined class size. SAGE used even smaller class sizes than STAR, in that the SAGE ratio was 12-15 students per teacher versus the STAR program of 13-18 students per teacher as the reduced size. These programs both used 21-25 students per teacher in the “larger” comparison groups. SAGE targeted low-income schools and, like the better-known STAR program, the results demonstrated students in small classes performed better. Also, like the STAR program, students in smaller classes were better behaved, and classroom time spent on discipline decreased. Adding support to the smaller class size theory, Borg et al. (2012) found that smaller class size contributed to higher student achievement. This was especially true in low-performing schools.

Research by Nye, Hedges, and Konstan-topoulos (1999) further supports small class size. These researchers conducted a follow up study to the Tennessee STAR program to see if the effects of small classroom size were still evident 5 years later. They concluded small classroom size encourages rapport among students and enhances student-teacher relationships. Effects were observed to last at least five years and were sustained adequately enough to suggest implications for educational policy. The research findings by Nye and colleagues (1999) lend further support to the supremacy of the K-8 environment considering that most K-8 schools typically have smaller class sizes than do their middle-school counterparts.

Student Mobility. Student mobility refers to “the phenomenon of students changing schools for reasons other than promotion” (Editorial Projects in Education Research Center, 2004). Reasons for mobility include parent’s job transfer or loss, homelessness, and other factors. Mobility appears to affect achievement and behavior as well as increase the risk of dropping out (Hartman & Leff, 2002). Although research shows a pattern of lower achievement for the high-mobility students, the impact mobility has on nonmobile students has not been studied in great proportions. However, when a new student enters a classroom, specifically after the beginning of the school year, the teacher frequently spends time reviewing in an effort to get that student caught up with the rest of the class, which inturn, slows/reduces instructional time for other students (Knox, 2011). Therefore, mobility could affect not only the mobile student, but also the nonmobile student where learning and achievement are concerned.

Research by Wright (2001) examined various risk factors on student achievement, including student mobility. Wright found that student mobility was not only a confounding factor associated with both family income and ethnicity, but also had less effect on achievement than either of these two factors. Hence, although student mobility impacts student achievement, other factors connected to mobility have a larger impact.

Environment. Almost every study on student achievement addresses school environment as playing a key role in student achievement as well as student behavior. School environment is much more than the building and includes quality of relationships among the students, faculty, and staff (Pardini, 2002). However, if the building is not designed for the middle school concept, particularly junior high school buildings, the success of the middle school concept, primarily where team teaching is concerned, is nearly impossible to fully achieve. On the other hand, if the middle school concept is applied, regardless of grade span configuration, then the building is minimized as a factor of the environment. Because middle school is a concept and not just a grade configuration, many educators and researchers feel that it can be implemented in any environment. Some educators and researchers argue that children attending K-8 schools have a stronger sense of community and belonging (Pardini, 2002). They also contend that in K-8 environments, the older children develop stronger leadership roles and are protective of their younger counterparts (Herman, 2004). K-8 proponents often express concerns with the middle schools due to the additional transition required. In their estimation, students who transition to a new environment for middle school lose the sense of community and belonging developed in the elementary environment as they are starting over with new administrators, staff, and faculty. Further, depending on what elementary schools feed into the middle school, many students are starting over with friends as well. All this can be a lot to bear in addition to the onset of puberty (Herman, 2004).

Transitions. Transitions occur when students move from one school environment to another. This includes transitions from elementary to middle school, middle school to high school, and the variations of each. The number of transitions may vary from district to district and within districts (Tennessee State Department of Education, 2013a). With each transition comes the usual new faces in class and a more challenging curriculum. Along with that also comes the unfamiliarity of a new school building, new faculty and administrators, and new expectations of achievement. These stressors can be overwhelming each time a transition occurs. To make matters worse, along with the transition to middle school usually comes puberty and awareness of the opposite sex. Evidence suggests that the effects of these transitions are mostly negative, resulting in a decline not only in performance, but also in motivation and self-esteem due to transition.

Not all outcomes of transition are negative. Research by Whitely et al. (2007) examined the effect of transition on academic achievement in Canadian schools. The researchers found no differences in academic achievement between those who transitioned into middle school (seventh grade) and those with no transition. What is interesting to note here is that Canadian educators work on programs to actively support the transition for students. These programs may have resulted in the successful adjustment after the transition. Paglin and Fager (1997) suggested that school systems with multiple grade spans should have “articulation and transition activities” in place to help students successfully make the transition (p. 9). Based on the findings of Whitely et al. (2007) and corroborated by Paglin and Fager (1997), it is possible that transitions may be more successful if students are better prepared for the expectations of each transition itself.

Standardized Tests. The No Child Left Behind Act required that states measure student achievement through standardized tests (U.S. Department of Education, 2002). Standardized tests are used to measure students’ knowledge in content areas and can be formative or summative in nature. Many saw the mandate of these tests as a benefit while some saw the tests as a liability (Phelps, 2005). One type of summative measure is an achievement test. According to the Tennessee Department of Education (2013b, p. 1), “the achievement test is a timed multiple choice assessment that measures skills in reading, language arts, math, science and social studies.” Achievement tests measure performance in specific objectives and can provide a description of student performance on academic skills. Further, according to research by Whitely et al. (2007), although results of high stakes testing may align with differences in students’ grades, standardized achievement scores should not necessarily be affected by immediate changes in students’ grades.

Valued-Added Assessments. There is a growing effort to hold teachers and administrators accountable for the quality of education they provide as measured by the achievement scores of students. Previously, growth scores were calculated simply as a difference between a student’s scores for the current year and the previous. Current value-added scores are more complexly calculated using sophisticated equations, and scores may take more things into consideration. Simply value-added scores are difficult to justify when there are factors beyond the teachers’ and administrators’ control such as family demographics and socioeconomic status. Not taking factors such as these into consideration is unfair and can be counterproductive. The newer alternative approach is more defensible as the resultant value-added scores are claimed to remove any noneducational factors from the measure of the student’s educational growth. These scores measure schools and teachers based on students’ progress rather than the ability of students to reach an absolute score (Ballou, Sanders, & Wright, 2004). Value-added scores typically include a control for initial level of achievement and may use growth curve analysis based on multiple measures of achievement.

One widely used version of value-added assessment is the Education Value-Added Assessment System (EVAAS) developed by William Sanders and Associates (Ballou et al., 2004). This system does factor out demographics and socioeconomic status and gives a score with only educational factors considered. This approach however, is not without criticism as some researchers believe it does not do enough to control for these factors because these factors not only affect the starting point but also the rate at which a student learns (Kupermintz, Shepard, & Linn, 2001). Scores are also criticized in that it is difficult to determine whether and to what degree the achievement tests used in the value-added calculations measure learning that takes place outside of school.

Grade Span Configurations and Student Achievement. Many view the EVAAS variations as effective measures of student growth in achievement, but what effect do the transitions from one school to another have on student achievement? There are numerous studies that show that transitioning between schools has a negative effect on achievement (Coladarci & Hancock, 2002; Combs et al., 2011; Herman, 2004).

Combs et al. (2011) examined differences between fifth grade reading and mathematics in elementary schools as compared to middle schools. After matching for school size, socioeconomic status, mobility, and limited English proficiency, it was found that students in elementary settings had higher reading and mathematics achievement than those who had transitioned to intermediate school. These authors used the Texas Assessment of Knowledge and Skills standardized test to determine these results.

Coladarci and Hancock (2002) reviewed several studies that looked at grade span configuration and student achievement and found similar results between schools. Specifically, those students with fewer transitions outperformed their transitioning counterparts in various academic categories. One study in this review determined that the differences in academic achievement between those students who transitioned and those who did not was dependent on socioeconomic status (Becker, 1987).

An interesting finding in research by Offenberg (2001) was that student achievement was higher in those students with fewer transitions; further, when comparing achievement scores of ninth grade students who had attended K-8 schools to those that attended middle school, Offenberg found those students with no transitions had higher achievement scores. So, it would appear that while grade span configuration does effect student achievement, and transitions appear to have a negative effect, delaying the transition until ninth grade actually appears to be better for students academically.

One study does not support the idea that transitioning to middle school negatively impacts student achievement. Research by Whitely et al. (2007) examined schools in Canada to see if there was a decrease in achievement in those students who transitioned into middle school compared to those who did not transition. The researchers did not find any difference in achievement between the two groups although they did find a significant decrease in the magnitude of gain/growth in mathematics achievement between fifth- and seventh-grade students, regardless of transition status. These researchers contributed this overall decline to developmental factors as opposed to environmental. The similar achievement between the two groups was attributed to the many programs that have been put in place in both elementary and middle schools to help prepare the children for the challenges they may face when changing school environments.

The purpose of the present study was to examine if there is a difference in achievement scores between students who attended a 6-8 middle school and ones that attended a K-8 school based upon race and socioeconomic status. It also examined what (if any) effect school size would have on student achievement. This study was guided by the hypothesis that students in K-8 environments would have better achievement with fewer transitions. The research questions for this study were as follows: (a) Is there a statistically significant difference in mean achievement scores between middle school students and those students who attend K-8 schools when controlling for race and socioeconomic status? (b) What is the magnitude of the relationship of school size with student achievement?

Instrument. The data source for the present study was the School Report Card at the Tennessee Department of Education (2013b). The current study used public schools with only one systematic transition to sixth grade and public schools with no transition through at least sixth grade. Four hundred forty-two schools met the criteria for inclusion in this study totaling 241,844 students. Of these, 203 schools transitioned into the sixth grade and 239 schools had no transition.

Measures. The following variables were measured.

• DependentVariable. The dependent variable was the EVAAS estimates of gain from the Tennessee Comprehensive Assessment Program Achievement Test that is administered to Grades 3-8 in public schools in the state of Tennessee. The dependent variable was further divided into subjects of math, reading, science, and social studies.

• IndependentVariables. The independent variables for this study include transition status (either transition or not), percent of students with low SES, percent minority, school size. These variables were chosen based on previous research in this area.

• Socioeconomic Status. SES was determined using Economically Disadvantaged Student Percent from the Tennessee Department of Education (2013b). “Students [who qualify for this program] are from families who meet certain income criteria making them eligible to receive free or reduced meals at school” (Tennessee Department of Education, 2013a).

• Based on this 40% eligibility requirement, individual schools whose economically disadvantaged student percent was > 40% were considered for the high-poverty status. Those schools whose economically disadvantaged student percent was < 40% were considered to have low-poverty status.

• PercentMinority. Due to the fact that SES and percent minority appear go hand in hand, schools whose percent minority was > 40% were considered high minority status. Those schools whose percent minority was < 40% were considered low minority. For the purpose of the study at hand, minority was considered all non-White students.

• SchoolSize. School size has been determined to have an effect on student achievement. Small elementary schools are those schools that have fewer than 400 students (Abbott, Joireman, & Stroh, 2002, as cited in Jones & Ezeife, 2011; Lee & Loeb, 2000). Based on research by Lee and Loeb, the study at hand divided schools into two sizes; for elementary schools, small < 400 students and large schools > 400 and middle school and schools with no transition (K-8), small < 700 students, and large schools > 700 students.

Procedures. For the purpose of the present study, middle schools with a transition included sixth grade through eighth grade. Schools with no transition included PK/K through at least the sixth grade. Schools that offered PK/K through eighth, ninth and/or 12th grade were also included as there was no transition during the time of interest.

Descriptive statistics were used to show the characteristics of the school data on the variables of interest (Table 1). Means for mathematics ranged from -1.523 to 6.686, and standard deviations ranged from 1.305 to 7.506. Means for reading ranged from -4.855 to 2.460, and standards deviations ranged from 1.525 to 6.372. Means for science ranged from 0.173 to 7.780, and standard deviations ranged from 1.981 to 10.156. Means for social studies ranged from -1.849 to 5.433, and standard deviations ranged from 0.656 to 8.826.

A four-way factorial MANOVA was used to analyze the data and determine whether transition had an effect on academic achievement of middle school-aged children, specifically sixth graders (see Table 2). School size, SES, and percent minority were used as input variables since they have been reported to have an effect on student achievement in previous research (Combs et al., 2011, Offenberg, 2001). The hypothesis tested was consistent with previous research findings indicating that students attending a school that transitioned after sixth grade would have higher scores on the EVAAS from the standardized Tennessee Comprehensive Assessment Program Achievement Test than their counterparts who transitioned into middle school at sixth grade.

A statistically significant main effect was seen only for Transition. The only multivariate statistically significant interaction effect was between transition and percent minority (p = 0.002; partial η² = 0.073), indicating a small but notable effect. Therefore, SES and school size were removed from the model allowing the associated sums of squares and df to be included in the error term. This second model was used for all other results. Results from the second model indicated statistically significant main effects for transition in math (p < 0.001) and reading (p < 0.001). A statistically significant interaction was found between transition and percent minority (p = 0.007), all of which supports the original model. Therefore, these findings suggest that transition had an effect on student achievement depending on whether or not the school had a high percentage of minority students.

As a post hoc procedure, separate two-factor ANOVAs were conducted for each dependent variable. No statistically significant effects were found for science and social studies. The main effects for math and reading were statistically significant (p’s = 0.03 and 0.05, respectively). There were also statistically significant interactions in the areas of mathematics and reading. For mathematics, there was a statistically significant difference between low and high percent minority where transition was concerned, _F(1,436) = 7.75, p = 0.006, partial η² = 0.017. Those schools with a high percentage of minority that did not transition (3.692) scored higher than all other combinations, but especially high percentage of minority with transition (-1.324). For reading there was also a statistically significant difference between low percent minority and high percent minority where transition was concerned, F(1,436) = 9.16, p = 0.003, partial η² = 0.021. Those schools with a high percentage of minority that did not transition (-0.375) scoring higher than all other combinations, but especially high percentage of minority with transition (-4.708). These results suggest that students in schools with high percentages of minorities perform better when there is no transition than their transitioning counterparts in both math and reading. However, effect sizes are very small, so these results should be interpreted cautiously.

The current study accounted for school size and SES based on previous research, but school size and SES were not shown to have an impact on student achievement. Many studies have controlled for these same variables, and in some studies, these variables have been shown to have a statistically significant influence on student achievement in the presence of variables measuring transition (Borg et al., 2012; Bromberg & Theokas, 2013; Offenberg,

2001). In the present study, this was not the case. SES was not effective in predicting student achievement The fact that school size, not class size, was measured, may have masked the effect of student to teacher ratios on achievement gains.

According to the U.S. Department of Education Statistical Report (2009), the achievement gap between Whites and Blacks in mathematics and reading has existed for many years. Over time there have been gains in educational achievement by both Blacks and Whites; nevertheless, a significant achievement gap still exists.

Upon examining the means for math and reading, with high and low percent minorities, with and without transition (see Table 3), it would appear that the schools with a high percentage of minorities that did not transition had better EVAAS estimates of gain from Tennessee Comprehensive Assessment Program Achievement test scores in both math and reading, than their transitioning counterparts. The gap between high and low minority schools was smaller without the transition, suggesting that keeping students in one environment through at least the sixth grade, may contribute to reducing the achievement gap between Whites and Blacks. These findings are supported in research by Combs et al. (2011) and Simmons and Blyth (1987, as cited in Offenberg, 2001).

The present study examined factors that affect student achievement where transition is concerned. The results suggested that percent minority has a small effect beyond the effect of transition on both math and reading achievement. Findings lend support to the growing body of research that suggests K-8 school environments, or at least a school environment using transition after sixth grade, is superior to traditional middle grade environments where student achievement is concerned. The middle school concept has the potential to address the academic and emotional concerns of middle school aged children when it is implemented in its full form (Weiss & Kipnes, 2006). Based on the findings of the present study and other related studies, practitioners might be wise to implement the middle school concept within the K-8 school environment, giving students more of a sense of community, self-esteem, and ownership of their school and grades. This could also potentially lessen the White to Black gap in math and reading achievement.