• No results found

The Impact of Mentoring on Standardized Test Results of African American Males in the Elementary and Middle Grades.

N/A
N/A
Protected

Academic year: 2020

Share "The Impact of Mentoring on Standardized Test Results of African American Males in the Elementary and Middle Grades."

Copied!
110
0
0

Loading.... (view fulltext now)

Full text

(1)

ABSTRACT

ANDERSON, KENNETH ALONZO. The Impact of Mentoring on Standardized Test Results of African American Males in the Elementary and Middle Grades. (Under the direction of Terrance P. O’Brien).

This study examined the effects of mentoring on standardized achievement for African American males in the elementary and middle grades. Mentoring has been deemed effective regarding social development and various academic indicators, but scant empirical data is available regarding the effects of mentoring on standardized testing. Thus, standardized test data of African American males participating in a district-wide mentoring program were examined over a three-year period. Results show that mentoring has smaller effects on standardized testing, while other factors continue to have substantial impact on student results. Recommendations for improving the academic success of African American males through mentoring and other school-based

(2)

THE IMPACT OF MENTORING ON STANDARDIZED TEST RESULTS OF AFRICAN AMERICAN MALES IN THE ELEMENTARY AND MIDDLE

GRADES. by

KENNETH ALONZO ANDERSON

A dissertation submitted to the Graduate Faculty of North Carolina State University

in partial fulfillment of the requirements for the Degree of

Doctor of Philosophy

CURRICULUM AND INSTRUCTION

Raleigh 2005

APPROVED BY: ________________________

Terrance P. O’Brien Chair of Advisory Committee

Committee Members: ________________________

Jessica T. Decuir-Gunby _________________________

Alan J. Reiman

(3)

PERSONAL BIOGRAPHY

(4)

ACKNOWLEDGEMENTS

(5)

TABLE OF CONTENTS

Page

LIST OF TABLES ...v

LIST OF FIGURES ... vi

CHAPTER I - INTRODUCTION ...1

Background Information...1

African Americans ...2

African American Males...3

Relationships...5

Democratic-Capitalism ...6

Standardized Tests ...8

Mentoring...11

Equity...14

Purpose...14

CHAPTER II – LITERATURE REVIEW...17

High Stakes Testing ...17

Socioeconomic Status ...20

Exceptionality ...23

Youth Development Programs...25

Mentoring and African American Males ...27

Program Evaluations...32

CHAPTER III - METHODOLOGY...35

Participants...36

Procedures...39

Data Analysis ...41

CHAPTER IV - RESULTS ...44

Year 1 Results ...44

Year 2 Results ...49

Year 3 Results ...52

Discussion...60

CHAPTER V – CONCLUSIONS AND RECOMMENDATIONS ...65

African American Male Achievement ...65

Strategies and Interventions...67

Socioeconomic Status ...69

Exceptionality ...70

School Curriculum ...70

Expectations...71

Democratic-Capitalism ...73

Limitations ...76

Delimitations...77

Significance...78

Recommendations for Future Research...78

References...81

Notes ...90

(6)

LIST OF TABLES

Page

CHAPTER I (Introduction) ...1

Table 1 (NAEP Test Results)...3

CHAPTER IV (Results) ...44

Table 2 (Year 1 Frequencies)...45

Table 3 (Year 1 Descriptive Statistics) ...46

Table 4 (Year 1 Multivariate Results) ...47

Table 5 (Year 1 Univariate Results) ...47

Table 6 (Year 1 Multivariate Growth Results) ...48

Table 7 (Year 1 Univariate Growth Results) ...48

Table 8 (Year 2 Frequencies)...49

Table 9 (Year 2 Descriptive Statistics) ...50

Table 10 (Year 2 Multivariate Results) ...51

Table 11 (Year 2 Univariate Results) ...51

Table 12 (Year 2 Univariate Growth Results) ...52

Table 13 (Year 3 Frequencies)...53

Table 14 (Year 3 Descriptive Statistics) ...55

Table 15 (Year 3 Multivariate Results) ...56

Table 16 (Year 3 Univariate Results) ...57

Table 17 (Year 3 Multivariate Growth Results) ...59

(7)

LIST OF FIGURES Page

(8)

CHAPTER I INTRODUCTION Background Information

The issue of meeting the needs of African American males and preparing them to survive the often uninvited, but ever-important standardized test is quite a challenge for school districts throughout the United States. Due to exceedingly high stakes, educational leaders often feel compelled to modify curricular programs to meet stringent test

requirements (Amrein & Berliner, 2003). Unfortunately, these modifications are often extreme and usually occur at the expense of sound pedagogical and developmental practice. Despite these pressures, some districts have remained firm in their commitment to development of the “whole child”, by meeting specific needs of their students. Many programs, such as the ADEPT Project (see Roth, Murray, Brooks-Gunn, & Foster, 1999) have been developed to meet student needs; but as with most programs, financial and other resource allocations are usually associated with these commitments. In similar regard, accountability is typically associated with these types of programs.

(9)

African Americans

As shown in the next few sections, attaining access to more opportunities that promote human flourishing are particularly important for African American males. Many individuals in this group have been and continue to be marginalized in this society

(Johnson, 2000). Thus school districts must create environments that prepare these students to thrive as citizens in this country and abroad. One approach to providing these opportunities is to prepare African Americans to meet and exceed requirements on standardized tests.

(10)

among African Americans. In similar regard, the next few sections will address issues specific to African American males.

Table 1

National Mathematics Scale in Public Urban Districts 2003 Scale

Score Americans African Americans European

4th Grade 216 243

8th Grade 252 287

National Reading Scale in Public Urban Districts 2003 Scale

Score

African Americans

European Americans

4th Grade 197 227

8th Grade 244 270

Note: Test scores were obtained through the National Assessment of Education Progress in Reading and Math conducted by the U.S Department of Education,

National Education Center for Education Statistics, 2003.

African American Males

(11)

Another challenge for African American males is the overwhelming negative perception of African Americans as a whole. This negative perception can be traced back well into the Colonial Period of the United States (Roark et al., 2000). In addition, other historical occurrences have contributed to this negative perception of African Americans. Throughout history, many African Americans have experienced many negative

encounters, primarily racism, dating back to early civilization in the United States. Racism can be defined as “the assignment of people to an inferior category” to determine “their social, economic, civic, and human standing on that basis” (Fields, 2001, p. 48). This inferior assignment occurred to preserve privilege, or access to valuable resources of various types based solely on one’s group membership (Johnson, 2000). As a result of racism, many people of color are consciously and unconsciously assigned to these inferior categories. Regrettably, this racism has matriculated into the schools (DeCuir & Dixson, 2004) and student performance has suffered because of it.

(12)

American males. Further, standardized testing can be beneficial if used to truly assess student progress and to inform teaching, rather than create labels and provide rewards or sanctions (Darling-Hammond, 2004; Nieto, 2000). Also, I suggest that an unwelcome truth in education is that as long as we exist in a capitalistic society of finite resources, structures that assign access to privilege are necessary. My only hope is that access to privilege is gained more equitably throughout the population.

Reverting back to this notion of inferior assignment, in the name of preservation of privilege, brings up questions of why? Why is there racism? Why is there inequity among African Americans and other ethnic groups? Why are some of our students achieving more so than others? Why are our schools contributing to these inequities? I briefly alluded to the fact resource restraints and fragmented relationships contribute to this but will now provide a more in-depth view of my beliefs. While this study may not provide a singular answer to all these questions, it is my hope that my philosophical view, in addition to the results of this study will bring light to some serious issues affecting public education and therefore our society.

Relationships

Bitting (2003) asserts that the central purpose of pedagogy is to make relations for the intent of stretching the mind “beyond its original dimensions” (p. 49). James Comer (2004) also realized the importance of relationships and developed the School

(13)

fruitful relationships. This barricade of sorts rest with our political and economic system: democratic-capitalism. To explicate my claim, I will use components of capitalism and democracy to support my ideas. In particular, I will highlight how the purpose of maximizing relationships is often caught in this democratic-capitalistic vacuum.

Democratic-Capitalism

A dominant theory of democracy rests with social capital (Englestad & Osterud, 2003). Although difficult to define in a few sentences, Englestad and Osterud (2003) suggest that with social capital theory “a working democracy depends on a population that is engaged in different social networks and voluntary associations in which they learn how to trust each other and thereby get training in overcoming problems of

collective action” (p. 5). These authors go further to note that this social capital theory is flawed conceptually and in practice. Instead they suggest that with democracy, although it reflects partisanship or collective action, implementation is quite different and highly partial. This partiality is rooted in power and powerlessness is usually linked to economic poverty (Englestad & Osterud, 2003). While this study is about schools, I think some interesting correlations can be drawn from this conceptualization of democracy and the following assertions on capitalism. Chiefly, as represented through efforts such as James Banks (2002), citizenship education is a function of our schools are therefore one of the mechanisms in which this “training” for democracy is provided. Thus citizenship education suggests that schools are products of our democracy.

During the Reconstruction, 1865-1887, our former leaders of this country realized the need for citizens of this country to have particular types of knowledge for this

(14)

institutions were designed to train citizens for living in a democracy. In a sense, public schools were created to some degree indoctrinate for initiation into our society. While the school curriculum has changed a bit, as shown in Pinar, Reynolds, Slattery, and

Taubman’s (2004) description of curriculum from 12 perspectives, I would argue that defacto purpose of schools has not changed very much. Schools now exist to shape our young ones for navigation and transformation in this society, but to some degree conform to what exists.

To clarify, consider Fulcher’s (2004) views on capitalism. Fulcher (2004) suggests that a central feature of capitalism is to make a profit. He further states that:

“In a capitalist society, both capital [value-laden entity] and labour [sic] have an abstract and disembedded quality, since both are separated from specific economic activities and are therefore able in principle to move into any activity that suitably rewards them. In real life this mobility is

constrained by the existing skills and experience of both the owners of capital and worker, and by the relationships and attachments that they have formed” (p. 15).

(15)

attachment” development. Ideally, schools would be free from these limiting contrivances, but in reality, our limited societal norms will not allow such.

Weaver (1986) alluded to this by asserting that the core institutions of a democratic-capitalism cannot be respected. He argues that institutions, such as

government and corporations cannot be trusted. Moreover, he suggests practices from these institutions in fact subvert the democratic process. Not only is this claim suggested, but Englestad & Osterud, (2003) and Weaver (1986) further support this claim of distrust for our institutions by citizens with empirical evidence. Weaver (1986) further argues that substantial change in a democratic-capitalistic society usually occurs when problems are exaggerated. I simply understand this as perception. This leads me to my central

argument.

Since policies and practices in our major institutions are influenced by perception and schools are products of these institutions, this necessarily means that policies and practices in our schools are somewhat perceptive. Likewise, a major perceptive device in our schools is that of the standardized tests. If a student can successfully master this test, they are perceived as smart, whether they really are or not. Thus, no matter how

counterproductive this may sound to a true educative process, I offer this as an unwelcome truth.

Standardized Tests

Despite the numerous shortcomings of standardized testing, there are some

(16)

results can provide valuable information to education leaders. Although stakes are becoming more and more disproportionate, it may be unrealistic to advocate for the discontinued use of standardized testing in public schools. After all, standardized testing reflects the capitalistic principles of our society.

Notwithstanding, schools have an economic issue associated with it. Schools are regularly inundated with requests for demonstrations from computer vendors, test developers, and textbook publishers, indicating that schools are a major part of the economy and thus subject to economic rules. As you may recall from Fulchur’s (2004) claim, some of these rules are that one may be constrained by his individual “skill” and/or his “relationship and attachment development” (p.15).

Famous African American leaders W.E.B Dubois and Booker T. Washington both seemed to understand this concept quite well. They both understood the importance of education and training, regarding the advancement of African Americans, but shared different approaches. As a metaphor, Dubois represents our democracy, in that he

advocated for intellectual and relationship development among African American males. Although Washington recognized the importance of intellectual and relationship

development, he promoted skill development. Maximizing relationships, mastering intellect and skills all share one common thread; they allow one to produce capital and gain economic power, and thus power in our democracy.

(17)

to “draw the line in the sand”. Standardized testing does just that. My thoughts intersect with the concepts debated by Dubois and Washington, in that we must train, while educating. Standardized testing is a litmus test in determining what relationships have been formed. Even if these assessments do not “really” capture these relationships, it is a tool used to capture perceived relationships. This is where Fulchur’s use of the term “abstract” comes to life. Plainly stated, ethnic minorities, particularly African American males need to learn how to handle this abstract concept of the standardized test to create more opportunities for themselves.

Fulchur’s (2004) presupposition of “abstract-ness” is further illustrated by our use of standardized tests. For example, many debates have occurred regarding the limitations and biases of standardized tests as a predictor of skills mastered and individual potential; yet these tests are still widespread, serving as a large gateway to personal and

professional development for students and citizens in the United States. Some of these gateways include access to the next grade level, access to high school, college, graduate schools and many professions.

Despite the unfortunate corollaries of standardized testing, programs and

interventions have been developed and shown promise in increasing the performance of African American males on standardized tests. Some of these interventions include direct instruction for standardized tests, while others include a more indirect approach such as focusing on child development, not just content knowledge and skills. Examples of direct instruction approaches that were deemed successful regarding standardized are the 6-8 week summer programs presented by Eccles and Gootman (2002). Contrarily, a

(18)

has shown much success with student development and simultaneous standardized testing performance is the School Development Program presented by James Comer (2004). In general, the stakes associated with standardized tests are irrational, but as long as these stakes continue to exist, we, as educational leaders must continue work to find ways to help African American males and other unsuccessful groups achieve better results. In the end, some of their life options may depend on it.

Mentoring

It is clear that there are host of variables to consider regarding the achievement of African American males. Some of these issues include poor performance on standardized tests, historical events, dealing with negative perceptions, peer pressure, and conditions within the African American community, many of which lead to a reduced sense of self-worth (White, 2004). Further, beyond conditions of the students, many educational leaders are faced increasingly rigid mandates with disproportionate ambiguity in expectations and lack of funding (O’Brien & Reed, 1999). Consequently, many educational leaders are forced to make tough decisions on how to engage African

American males [and others] while meeting legislative requirements with little empirical data or resources to support their decisions.

(19)

foster positive transformations in dispositions and performance. Although this study was conducted with adults, I feel that changing dispositions may be generalizable to school-aged children. In addition, my posture seems to be reflected in the work of Tarek Grantham (2004) who used stages of identity development to develop a model for mentoring. As will be explained later, this model presents deliberate efforts to help African Americans pursue and perform better in gifted education. My hope is that these interventions also have mediating effects for standardized testing.

Although mentoring has the potential to influence standardized test results, empirical data on the effects of mentoring and standardized testing is somewhat scant. Based on the potential for mentoring to improve certain academic indicators, such as self-esteem, school attendance, and grade point average, it is plausible that mentoring may have some effect on standardized achievement for African American males. Albeit plausible, evidence is not very conclusive. The next few paragraphs highlight the complexities of generating this conclusive evidence and propose how mentoring can possibly influence standardized test performance.

(20)

their communities (Grantham, 2004). These experiences can debilitate and limit one’s success in school and lead to a reduced sense of self (White, 2004).

Although the experiences mentioned above can encumber student development, positive experiential encounters can help increase one’s view of self (White, 2004). One’s view of self may be related to one’s academic self-esteem, or one’s “global feelings and beliefs” (p. 2) about school. Academic self-esteem has been found to significantly contribute to the achievement of African Americans in school (Justice et al., 1999; Lockett & Harrell, 2003). Further, Brown & Jones (2004) found that one’s beliefs about how school can help them in the future also plays a significant role in how students perform in school. As shown by the authors presented above, it seems plausible that mentoring experiences can increase one’s beliefs in themselves and one’s academic self-esteem. Thus, it may not be beyond the realm of possibility that this increased esteem can enhance one’s performance on standardized tests. Performance may increase not only through increased confidence, but may also be improved through participation in more intellectual activities, such as studying more, being more attentive, and completing assignments more thoroughly as a result of having a mentor.

Although mentoring has the potential to improve academic indicators, there are a number of other variables that influence performance on standardized tests.

(21)

Equity

What I hoped to demonstrate in this introduction is that schools are products of our democracy-capitalistic society. After having shared my sentiments, my sincere hope is that schools can continue to inform and partially shape this society. My belief is that informing and to some degree shaping is what we can expect from schools at best. Thus, instead of fighting to rid our schools of standardized testing, I embrace this sorting mechanism as a tool for offering more equity. Equity, in this case, reflects using

standardization as an opportunity and a challenge to making our institutions [government, businesses, prisons, etc.] look more like our society. This equity can become more of a reality in two ways: 1) if standardized test results are used to inform and alter our pedagogic strategies to better meet the needs of students and 2) if more individuals from diverse ethnic and socioeconomic backgrounds can learn to master this art of perception we call standardized tests. My posture may sound somewhat cynical, but as long as we live in this democratic-capitalist society, I have accepted that the bell curve will remain in tact. My only passion is to continue to strive to reach more equity within this distribution. Mastering the standardized test is merely one way of contributing to this equitable

reshuffling of sorts. Now that my philosophical frame has been established, I will present the purpose of this study.

Purpose

(22)

educational officials are forced to make decisions regarding intervention strategies with little or weak evidence and this lack of empirical support may contribute to these improbable expectations (U.S. Department of Education, 2003).

Much discourse has surfaced in areas such as youth development and mentoring regarding their effects on student achievement. In fact, many of these studies offer conclusions that suggest youth development programs enhance academic achievement (Brown & Jones, 2004; Davis, 2003; Valencia & Villarreal, 2003). However, the methods used in some of those studies are often highly subjective or lack conclusive evidence. Likewise, there is a push from the federal government to evaluate programs using more conclusive empirical designs (U.S. Department of Education, 2003). As a result, the intention of this study was threefold: 1) mentoring was assessed to evaluate its effects on standardized achievement of African American males and 2) techniques using archival data were employed to provide the research literature with an empirically rich body of work regarding the effects of mentoring on standardized achievement of African American males and 3) the results of this study can be used to inform program and curriculum developers on how to proceed when planning interventions.

Before presenting the fundaments of this study, the next chapter provides an in depth presentation of some of the key issues regarding African American males and standardized testing. First, the implications of standardized assessment and the achievement gap will be discussed as it relates to African American males. Next,

(23)
(24)

CHAPTER II LITERATURE REVIEW High Stakes Testing

Even before the No Child Left Behind Act, a general shift towards school accountability, which places demands on the public school to show achievement on a variety of measures, primarily through standardized assessment had become more widespread (Abrams & Madaus, 2003; Doran, 2003). However, the idea of testing or assessment can be traced back to China as early as 210 B.C. Many assessments in this era were performance-based and were used to determine qualifications for civil and military employment (Madaus & O'Dwyer, 1999). The idea of assessment has evolved since the Chinese era, but still can only be divided into four distinct categories: 1) oral or written response to questions 2) product-based 3) performance-based or 4) selection from predetermined answer choices. The first three types of assessment focus on qualitative evaluation, while the fourth focuses on quantitative comparison. As society moved

through the three paradigmatic periods of premodernism, modernism and postmodernism, there has been a shift in the demand for quantitative as opposed to qualitative assessment. Factors such as increases in population, the introduction of the factory model, fiscal administration reductions and reduced subjectivity can be credited for a shift in demand for quantitative assessment (Madaus & O'Dwyer, 1999).

(25)

decision-making. In fact, some of the results of standardized assessments have been used to restrict individuals. For instance, the National Teachers Examination, in the United States largely grew out of efforts to legally combat desegregation litigation and restrict access to a higher quality of life for African-American citizens (Baker, 2001b). Before World War I, many African Americans and other immigrants were subject to

psychological testing using biased tests. Not surprisingly, biased results emerged and these results were chiefly used to restrict and label these citizens regarding placement in the military (Spencer, 1975).

Although standardization in today’s terms is purported to close achievement gaps, it has been suggested that this approach contributes to facilitating deficits (Baker, 2001b). For example, many ethnic minorities perform poorer on achievement tests, which are often based on a European American reference frame (Gardner III & Miranda, 2001; Williams, Williams, & Mitchell, 2004). In addition standardized tests are sometimes viewed as threats to perpetuating stereotypes for ethnic minorities. These threats create anxiety, thereby causing reduced performance on standardized tests (Steele, 1997), all of which can lead to restricted access to college enrollment, graduate and professional schools.

(26)

1999). Other outcomes of high-stakes testing involve public scrutiny of school personnel for poor results or extrinsic rewards for high marks attained.

While standardized testing has succeeded in identifying low performing sub-groups, more guidance to closing these achievement gaps is necessary. The achievement gap can be defined as a general deficit in the average achievement of a particular group, relative to another group (Brown & Jones, 2004). This achievement can include, but is not limited to factors like grade point average and standardized test scores. Currently, many educational leaders, such as building-level and district-level administrators are charged with meeting stringent growth goals regarding particular groups from year to year. In fact, severe consequences, such as school reorganization or valuable incentives, such as financial rewards are usually attached with these growth requirements. Moreover, school administrators and teachers are faced with tremendous pressures of meeting these requirements with little guidance (Jerald, 2003).

As mentioned, standardized testing has done well in identifying achievement gaps, but guidance to closing these achievement gaps is less clear. Efforts to improve achievement can occur through mentoring, but evaluation of programs, such as mentoring, is usually minimized, due to the need to place funds in direct services (Flaxman, 1992). However, evaluative efforts, both summative and formative could prove beneficial to the body of knowledge on mentoring. Therefore, this paper will explore the effects of an established mentoring program on the achievement of African American males.

(27)

individuals of European decent (Brown & Jones, 2004). Two highly contentious topics in education that are thought to contribute to this achievement are socioeconomic status and exceptional status. These topics are discussed next.

Socioeconomic status

Socioeconomic status (SES) is one of those buzz words that has plagued

education for years. SES includes a number of variables such as parental income, parent education levels and types of employment and is commonly thought to influence student achievement (Ma, 2000; Okpala, Okpala, & Smith, 2001). These findings may suggest that having a middle class background is a prerequisite for academic success. Since there are accounts of individuals achieving in spite of low SES, the previous statement cannot be fully endorsed. However, there are conditions related to SES that may prove

detrimental to the success of students in schools. Some of these conditions include parental involvement, parental education levels, and resource allocations among schools. Although these conditions have been stated, I will unpack these conditions a bit more to provide clearer insight on their significance regarding African American male

achievement. Using historical periods in the United States, some important foundations of SES will be presented in the next few paragraphs. Likewise, some of the implications of SES on the achievement of African American males will be discussed.

(28)

period is that of the Reconstruction. The Reconstruction is a period after the Civil War that occurred from roughly 1865-1877. During the Reconstruction, three themes emerged: the rise of public schools, civil rights activism, and economic development (Roark et al., 2000). During this time, the 13th and 14th Amendments to the Constitution were passed, abolishing slavery and granting Blacks citizenship. Although some rights were granted, many African Americans still experienced a life very similar to that of the enslaved. For example, many “Black codes” were established that disenfranchised African Americans from flourishing, such as charging African Americans taxes to work in professions outside of domestic labor and agriculture (Roark et al., 2000).

(29)

emotional distress and lowered esteem for students. Consequently, this emotional distress or reduced self-esteem permeates with the children of these families and can have

negative effects on school achievement (Lockett & Harrell, 2003).

Several studies on identity development have shown that excessive negative stimuli has caused African American males to think negatively of themselves and their potential, and these beliefs affect classroom performance (Brookins, 2004; Day-Vines, Patton, & Baytops, 2003, Lockett & Harrell, 2003). Reflecting on this research, a major outcome of SES, as it relates to African American males appears to be associated with beliefs.

Another outcome of SES is resource allocation. In 1896, the landmark ruling of Plessy versus Ferguson was passed legalizing segregation (Zirkel, 2002). Although the Brown v. Board decision reversed this law in 1954, re-segregation of schools is emerging yet again. This resegregation has resulted in inequitable dispersion of resources, yielding lower achievement among many students of color (Borman et al., 2004). Many African American males reside in poorer districts and often attend underfunded schools that lack necessary instructional resources (Gardner III & Miranda, 2001).This inequitable

(30)

disparate inequities (Gardner III & Miranda, 2001; Rice, 2004). These inequities have lead to low teacher retention and recruitment, limited instructional resources, and reduced achievement for students in poorer districts (Baker, 2001a).

Exceptionality

Exceptionality, or special education, usually refers to students with disabilities. However, special education often includes gifted education. For this study, academically gifted students are treated as a group within special education. Special education was designed to provide appropriate services for students with distinct needs, but this concept has created some very contentious issues in education (Yell, Rogers, & Rogers, 1998). The Individual with Disabilities Act (IDEA) was amended in 1997 to improve the

education of students with disabilities. Several changes were made to this policy, but two will be discussed. First, IDEA was amended with changes to Individual Education Plans (IEPs), a personalized education plan for students with disabilities. The changes consisted of including benchmarks to foster student growth and development and provide

measurable progress. Also, amendments to IDEA included assessing students with disabilities using state and local assessments to provide more measurable results (Yell, Rogers, & Rogers, 1998).

(31)

many students with disabilities (Ysseldyke, Nelson, & Christenson, 2004). Moreover, although a purpose of standardized testing is to inform pedagogy, many of the results are not being used by teachers in the development of IEPs (Ysseldyke, Nelson, &

Christenson, 2004).

Mass testing of students with disabilities has created a negative social stigma for some of these students, resulting in reduced self-esteem and reduced expectations from school leaders (Gardner III & Miranda, 2001). Since the stakes associated with testing are increasingly high, students with disabilities or their parents are often blamed for their lack of success. Instead of modifying pedagogy to meet the needs of students with disabilities, these students are often subjected to unfruitful pedagogy, increased behavior

management practices, and derogatory treatment. The research literature shows that African American males are over-identified in special education as having a disability (Gardner III & Miranda, 2001), which means that many African American males are often recipients of this negative treatment. Additionally, students with disabilities are usually relegated to low-level coursework, which also contributes to reduced success of African American males (Ford, 1996).

Another disparity in special education is that of gifted education. African American males are less likely to be placed and retained in gifted tracks. When African American males are placed in high level tracks, such as gifted programs, they often underachieve (Ford, 1996). In gifted education, this underachievement and

(32)

Further, cultural congruence in instruction, or the lack thereof, also affects motivation and achievement in the classroom (Ladson-Billings, 1995). Some of these cultural differences that influence the achievement of African American males are the lack of African American teachers and administrators (Cooper & Jordan, 2003). Further, the curriculum presented to students often lack cultural connections with ethnic minority students. Thus, the academic identity of African American males is not reinforced in schools. Although, cultural relevance has improved, many changes have been additive. Rather than employing additive techniques, multiculturalism should pervade the curriculum (Zeichner et al., 1998). This shift to multiculturalism should reinforce diversity, rather than treating difference as deficits (Nieto, 2000). In addition to multiculturalism, youth development programs are also beneficial for school-aged children.

Youth Development Programs

Adolescents in general, are highly impressionable, either negatively or positively (McLearn & LaFrance, 1999). Factors such as single-parenthood, poor living conditions and low income levels contribute to increased chance of adolescents participating in risky behaviors and having reduced interest in education. African American males in urban areas are likely to experience one or some these negative conditions; thus their likelihood of dropping out of high school or participating in risky behavior is high. For example, of African American men born in the late 1960s, 20% had served prison time before

(33)

risky behavior. Although many African American males encounter negative experiences, many succeed through support and development programs (McLearn & LaFrance, 1999).

Youth Development Programs (YDPs) are programs designed to promote healthy development among adolescents using a framework that is designed to reduce the amount of risky behaviors that an adolescent may engage in (Roth et al., 1999). When developing YDPs, Roth et al. (1999) suggest that adolescents should be challenged and provided opportunities to interact with caring individuals in a safe environment on a regular basis. Consequently, Roth et al. (1999) suggests five goals for healthy development of

adolescents. First, adolescents should develop the “ability to be intellectually reflective” (p. 270). This goal can be met by providing adolescents with a broad knowledge base through access to individuals and information that promote high academic achievement through formal and informal means. The second goal “preparation for a lifetime of meaningful work” (p. 270) can be accomplished by helping adolescents become aware of and understand the life options they have. Further, adolescents should be taught values, standards, and behavior that typically increase opportunities for obtaining these options (Roth et al., 1999).

(34)

maintaining good health” (p. 271). Hence, adolescents should be taught healthy practices and made aware of the negative effects of unhealthy practices on the body and mind (Roth et al., 1999). Mentoring is a youth development option that helps promote healthy behavior; therefore mentoring as it relates to African American males is discussed next. Accordingly, the next section presents background information on mentoring and how it can promote positive development among African American males.

Mentoring and African American Males

In an increasingly global society, how an individual handles and copes with difference is critical. Wadsworth (1979), citing the famous developmentalist Jean Piaget, suggests that most learning and dispositional growth occurs when one’s expectations do not match his or her experience. This mismatched state is known as disequilibrium. In order to achieve balance or equilibrium, interventions are often times necessary. Considering these assertions, African Americans are often placed in schooling environments that are very different from their personal experiences and how these students negotiate within these environments is central to the student’s success. One major youth development strategy for handling ambiguity, managing difference,

improving academic achievement and quality of life for African Americans has occurred through mentoring.

(35)

mentoring as one-to-one relationships, Struchen and Porta (1997) place mentoring on a continuum, with endpoints ranging from one-to-one relationships to one mentor and a large group of mentees. Although there is no universally accepted definition, Grantham (2004) suggests that “love, commitment, and responsibility” (p. 242) are key attributes for successful mentoring.

Mentoring is quite beneficial when there is one mentor for one child, but there also advantages of group mentoring (Parker, 1995). Some of these advantages include opportunities for mentees to share similar problems and learn how others cope. Group mentoring also fosters a sense of belonging and can relieve pressure from the mentor. Establishing bonds can often be difficult and are sometimes unattained in one-to-one situations, but the dynamics of the group can alleviate these pressures. Further, group mentoring can encourage the importance of belonging to a group (Grantham, 2004), while fostering self-regulatory practices between the mentees. Self-regulatory practices often include behavior correction between mentees, with or without the presence of the mentor (Parker, 1995).

Utsey et al. (2003) also suggest that group mentoring is more culturally congruent with the life experiences of African American males and reduce burnout for mentors. Despite these differing definitions and approaches, mentoring programs can be used as interventions and those with strong conceptual goals have shown to be effective regarding student development (Burrell, et al., 2001).

Mentoring programs have proven valuable to African American males at various educational levels (Burrell et al., 2001; Lee, 1999). Mentors have been effective in

(36)

the number of African American males pursuing doctorates and other higher education degrees (Morgan, 1996; Struchen & Porta, 1997; Townsel, 1997).

Although mentoring has been successful in some cases, results have been mixed. For example, the Office of Minority Advancement at Auburn University implemented a mentoring program targeting adolescents in junior high school in order to retain African American male students who may be at risk for dropping out later in high school

(37)

Another mentoring program that was deemed successful in reducing destructive behaviors in African American males is the therapeutic group mentoring approach. This therapeutic group mentoring approach was employed with a group of adolescents in foster care (Utsey et al., 2003). Mentors included members of a social fellowship of African Americans who met with mentees for about 2 hours each week. Mentors were trained to be sensitive to adolescent developmental needs and a licensed mental health provider also participated in the weekly session. Case study results from this mentoring project, citing anecdotal data, suggest that efforts were effective in changing attitudes and behaviors of the participants. However, the authors suggest that more empirical research designs are necessary (Utsey et al., 2003).

Mentoring is also believed to be a critical link in attracting and retaining African American students to gifted education (Grantham, 2004). In doing so, mentors are often encouraged to focus on multiculturalism (Banks, 1994). Within this multicultural framework, mentors should use a collection of approaches, rather than a universal approach to provide effective guidance (Banks, 1994). Based on these concepts, Grantham (2004) developed the Participation Motivation-Racial Identity Model to

enhance the quality of mentoring programs, while also fostering success of these students in gifted education. Overall, the Participation Motivation-Racial Identity Model focuses on expectations and values by delineating three areas: Participation Competence

Expectation, Outcome Attainment Expectancy, and Value of Outcomes. Participation Competence Expectation focuses on helping African American males believe in

(38)

demonstrate the intrinsic and extrinsic benefits of participating in gifted programs (Grantham, 2004).

Although mentoring can be beneficial, mentoring programs often fail because they lack focus and are expected to produce drastic results in short periods of time (Struchen & Porta, 1997). Further, some African American male mentors have reduced levels of positive identity development themselves, which may inhibit the mentoring process (Grantham, 2004). When dealing with populations facing great adversity, such as African American males, time and focused efforts are necessary to achieve success. Similar to Reiman’s (2004) thoughts on dispositions, or the act of interpreting, judging and acting, dispositions within African American males can be gained through deliberate mentoring interventions. Accordingly, this agency can be enacted when these students are faced with ambiguity, curriculum practices, and/or learning environments that may not complement their strengths.

The studies presented above demonstrated the benefits of mentoring for African American males, but also highlight the need for more empirically designed evaluations that can yield more conclusive evidence about mentoring. Considering the elusive and sometimes delayed effects of mentoring, empirically designed studies may in fact produce mixed or unfavorable results. No matter what the outcome, empirical studies provide valuable guidance in the planning process for these types of interventions and also establish a better gauge of what to expect. The next section presents other

(39)

Program Evaluations

The education literature is well endowed with research suggesting that Youth Development Programs (YDPs) are effective regarding student achievement (Karcher, Davis, & Powell, 2002; Mitchell, Bush, & Bush, 2002; Pagan & Edwards-Wilson, 2003). Most evidence for the success of YDPs is gained through program evaluations. Program evaluations can be defined as an informal or formal process used to inform decision-making through “systematic collection and analysis of information about alternatives” (Schalock, 1988, p. 3). Likewise, Roth et al. (1999) synthesized the results of several major youth development program evaluations throughout the United States. The following section presents the results of some of the findings.

The first program evaluation assessed the effectiveness of the Big Brothers/Big Sisters of America (BB/BS) program, a major mentoring program in the United States. A longitudinal analysis was conducted using random assignment of BB/BS locations over an 18-month period. Accordingly, eight sites were selected, comprised of 487 BB/BS participants and 472 control participants not in BB/BS. Results of the study found that non BB/BS participants achieved lower grades, skipped twice as many days of school, and felt less competent about school work (Roth et al., 1999).

The Quantum Opportunities Program (QOP) evaluation was another program evaluation presented. The QOP is a community-based program for youth receiving public assistance. Participants are required to complete 750 hours of education, personal

development and service activities over multiple years. A five-year longitudinal analysis of 4 sites was conducted using 82 randomly assigned experimental and control

(40)

increased. Participants were also significantly more likely to graduate from high school, participate in post-secondary education, and have fewer children. Although academic skills increased, there were no significant differences in grades in high school (Roth et al., 1999).

The ADEPT Project is an afterschool program that provided homework assistance and esteem-building activities for two hours per week throughout the school year.

Comparison groups were established using 540 program participants and 296 comparison participants. Results showed that there were no significant effects for self-esteem or behavior. Overall, results for standardized testing were mixed. No significant difference in overall scores for standardized testing was found between the two groups; but the average increase in standardized testing was significantly higher for program participants (Roth et al., 1999).

Another program evaluation presented by Roth et al. (1999) assessed the

(41)

This final section presents the results of a analysis of several previous meta-analyses and reviews of program evaluations using experimental and quasi-experimental techniques, conducted by the National Research Council and Institute of Medicine (NRC&IM). Although this study had several objectives, the primary objective was to determine whether or not YDPs were successful or not. Overall, the NRC&IM found that YDPs were effective regarding mental health, interpersonal skills, coping strategies, behavior, motivation and educational achievement. Additionally, the NRC&IM

discovered that person-centered rather than environmental-centered programs seemed to be most effective (Eccles & Gootman, 2002). Interestingly, the results of this study suggested that short-term programs [less than 9 months] can be just as effective as programs lasting 9 months to two years, thereby contradicting the research presented in the previous paragraph.

(42)

CHAPTER III METHODOLOGY

In this study, the primary group of interest was African American males participating in a district wide school-sponsored mentoring program known as Helping Hands. Reflecting on the research presented in the Chapter II regarding student achievement and standardized testing, variables of interest included socioeconomic levels and dis/ability level. The findings of this study can be used to help establish realistic expectations from interventions similar to the one presented. On that same note, these results can be used to inform program coordinators on how to proceed when developing mentoring programs or similar interventions designed to improve standardized test results. Specific research questions that were addressed in this study included:

1) Is there a significant difference in the overall achievement of African American males participating in the Helping Hands program compared to African American males who are not?

2) Is there significant interaction between socioeconomic status and Helping Hands as it pertains to academic achievement for African American males? 3) Is there significant interaction between exceptionality and Helping Hands as it pertains to academic achievement for African American males?

(43)

5) Is there a significant difference in the overall achievement of African American males participating in the Helping Hands program compared to African American males who are not, controlling for length of time in the mentoring program?

6) Is there significant interaction between socioeconomic status and Helping Hands as it pertains to academic achievement for African

American males, controlling for length of time in the mentoring program? 7) Is there significant interaction between exceptionality and Helping Hands as it pertains to academic achievement for African American males, controlling for length of time in the mentoring program?

8) Is there significant growth in the achievement of African American males participating in the Helping Hands program compared to African American males who are not, controlling for length of time in the mentoring program?

Participants

(44)

During the 2001-02 school year, the ethnic makeup in Wake County was approximately 62% European American, 26% African American, 6% Hispanic, 4% Asian, 2% Multiracial and less than 1% American Indian. Wake County employs an aggressive busing and student assignment program to diversify its schools through magnet programs. Overall, Wake County Schools is a high achieving school district with more than 91% of their students at or above grade level in reading and math, as measured by state End-of-Grade (EOG) exams in 2004.

Although Wake County Schools is a high achieving district, there are

achievement gaps among disaggregated groups. Specifically, in 2004, approximately 97% of European American students were at or above grade level in reading and math, compared to approximately 80% of African Americans. Further, 96% of students who did not receive free or reduced lunch were at or above grade level, compared to

approximately 78% of those who did receive free or reduced lunch. Females also

outperformed males by roughly 3%, with roughly 93% of females at or above grade level and 90% of males at or above grade level. While there are a number of achievement gaps in the district, the achievement of African American males will be the focus of this study.

In order to complete this study, Helping Hands (HH), a mentoring program for African American males in middle and elementary schools throughout Wake County, was used to evaluate its effects on student achievement. The purpose of the Helping Hands is to provide after-school mentoring activities for African American males in the Wake County School System. The Helping Hands program has three overarching goals:

(45)

2) To boost the self image of the African American male student. 3) To foster school success of African American male students.

The Helping Hands program was created in the late 1980s by, Robert E. Bridges, former Superintendent of Wake County Schools. Bridges conducted research1 on the academic achievement of African American males and found several themes that hinder their development. These themes were: lack of touchable [direct access] role models, fluctuating relationships with females, low involvement in school, and poor academic skills development. Therefore, Bridges decided it was necessary to reach out to the schools and communities for help with this matter.

Specifically, Bridges sought African American males working as teachers, counselors and administrators to serve as positive role models for African American males in grades 3-8. Mentors selected for the program are required to attend an initial 2-hour mentor training and given a resource guide to address four major areas throughout the school year. The two hour training is led by the director and consists of becoming familiar with the resource guide through lecture and question-answer format. The

resource guide includes suggestions and guidance for Positive Role Modeling, Improving School Success, Understanding and Improving Peer Relations, and Developing

Leadership Potential. Thereafter, mentors are assigned 5-10 mentees from a particular middle or elementary school and are contracted to spend 2 hours per week with the mentees, usually at a weekly after school session. Weekly contact with mentees is required for the full academic year.

(46)

interactions with students at school, while some mentees hear about the mentoring program from other participants and request to join the program. While there is an application process that must be completed by parents, once potential mentees are

identified, there are no rigid guidelines for determining who qualifies for the intervention. The most common factor of not being accepted into Helping Hands is related to the program’s policy of assigning no more than 10 mentees to a mentor, coupled with low numbers of African American males employed in schools.

The major aim of the program is to identify African American males who are likely performing below their academic and/or leadership potential. Throughout the year, the mentees participating in the program are required to learn poems that promote

positive self-esteem, along with practicing organizational techniques, study skills, and public speaking. Furthermore, mentees participate in coordinated activities throughout the school year with mentees from other schools. These activities include oratorical contests, trivia pursuit, athletic competitions, and various fieldtrips. Some of the fieldtrips include visits to colleges, camping and other enriching activities, depending on funding.

As mentioned previously, goal 3 of Helping Hands is “to foster school success of African American males”. This school success includes, but is not limited to school attendance, discipline, and academic performance. Since accountability within schools has increased, improving End-of-Grade Test results have become a major component of fostering school success for the participants.

Procedures

(47)

End-of-Grade test scores for all three years. Thus, depending on the number of mentees at each grade level with complete records, a stratified random sample of the same number of African American males with corresponding grade levels was selected as the control group. Grade level was chosen as the selection criteria to ensure fidelity in comparisons. If grade level was not used as a safeguard, results could be misleading. For example, if the control group had a total of 100 7th graders, and the intervention group had a total of 100 8th graders, the intervention groups would necessarily score higher, based on the progressive scoring scales of the End-of-Grade tests.

To further explain the selection criteria for the control group, an example is provided. If there were 150 total mentees in Helping Hands at Wake County Schools in 2004, 75 in 3rd grade and 75 in 4th grade, a similar list of all African American males at the same schools were generated. From this list, a systematic random sample of 150 students from this list, sorted by grade level, was selected using a random number generator. The sample was stratified-random, rather than random, to ensure that the frequency of grade-level was equally represented in both groups. Since, this procedure was a three-year study, this selection technique was completed for each of the academic years.

(48)

those students participating in Helping Hands (N=224) with a comparison group (N=224) for the school year 2002-03. Finally, the third cohort included those students participating in Helping Hands (N=338), with a control group (N=338) for the school year 2003-04. The sampling frame for the control group was a database of African American males not in Helping Hands who attended the same schools as participants in Helping Hands.

Since the literature in Chapter II, suggested that socioeconomic levels and dis/ability levels influence achievement, these two variables were included as

independent variables in this study. Socioeconomic levels were be determined by student lunch status (free, reduced or full) and therefore coded using two levels free/reduced and not free/reduced. Dis/ability level was determined using archival records from the school district providing designated exceptional labels for all participants. This category is referred to as Exceptionality and was coded using three categories: Not-identified,

Identified with a disability, and Academically Gifted. Appendix B provides a detailed list of codes used for this study, while Appendices C-E show a sample of the data from each of the cohorts.

Data Analysis

(49)

[See Appendices F-K]. After visually inspecting the histograms, only one set of mean presented extreme violations. This set of means, “Growth in reading scores for Year 2” were therefore excluded from any subsequent analyses.

After examining conformity to assumptions, the multivariate analysis of variance (MANOVA) procedure using the General Linear Models (GLM) function in the SPSS 12 software was used to assess differences in EOG test scores in reading and math. GLM was used because cells, such as the number of exceptional students in each group, were unbalanced. Univariate Analysis of Variance of EOG scores were conducted after the MANOVA procedures to assess differences on specific tests for reading and math. Post-hoc analyses, using Tukey’s studentized range, were also conducted to unpack omnibus results of the MANOVA and Univariate analytic procedures. EOG tests in reading and math were used as dependent variables.

The primary categorical variable of interest was whether or not the student participated in the mentoring program. Since the literature cited in the previous section suggests that exceptionality and socioeconomic status affect academic performance, these two variables were also used as independent variables. In particular, exceptionality and socioeconomic factors were used as independent variables to examine any interaction effects between these factors and the intervention. In addition, main effects of these two variables were also examined. The variable for the intervention was designated as

Helping Hands, followed by a two digit number to represent the year under investigation. This intervention variable was coded using 0 (did not participate in HH) and 1

(50)

socioeconomic status was coded using two levels 0 (not receiving free/reduced lunch) and 1 (receiving free/reduced lunch).

First, a MANOVA procedure was used to assess whether or not there was an overall difference in the achievement of African American males who participated in Helping Hands in the 2001-02 school-year, compared to those of like demographics [race, gender, and grade] who did not. Next, using these same groups, a MANOVA was conducted assessing whether or not there was a significant difference in the growth levels of African Americans participating in the program, compared to those of like

(51)

CHAPTER IV RESULTS

This section presents results of this study in three phases. Since the study was conducted using data from a three year period, results are presented by year. For each year, frequency tables and descriptions are provided first. Descriptive statistics, specifying means and standard deviations for EOG test scores and mean growth levels are presented next. Thereafter, the Multivariate statistics for the analysis of mean EOG scores in reading and math are provided, followed by Univariate results. Post-hoc tests results are also provided for significant F statistics. Since mean EOG test scores in reading and math were assessed, in addition to mean growth scores in reading and math, the previously mentioned results are repeated for the growth analyses. Thus each year contains two sets of results: 1) for overall performance in reading and math and 2) for overall growth in reading and math. After results for each year are presented, a discussion of these results by research question is provided.

Year 1 Results

(52)

(N=320) was 152.38, SD = 12.03. In addition, the overall mean score in math for Year 1 (N=320) was 257.28, SD = 9.17.

Table 2 (N=320)

Table 3 shows the mean scores in reading and math by HH Group and Control Group. Table 3 also shows that although the HH Group did not perform better overall in reading or math, but they did grow more than the Control Group in math. The standard deviations for the Control Group were slightly higher for the HH Group in reading and math. Since the sample size in this data was large, histograms were constructed [see Appendices F-G] to visually inspect the data for extreme violations in. As shown, these histograms do not represent extreme violations of the normality assumption.

Frequency Table for Exceptionality and SES (Year 1) HH Group Control Group Exceptionality

Not Identified with Disability 113 115 Identified with Disability 46 34

Academically Gifted 1 11

SES

Not Free/Reduced 55 89

(53)

Table 3 (N=320)

Year 1

Mean Reading and Math Scores For HH Group and Control Group Numeric Scores

HH Group Control Group

Subject M SD n M SD n

Reading 151.27 11.39 160 153.48 12.71 160

Math 256.54 8.72 160 258.03 9.58 160

Reading Growth 4.80 10.50 160 5.24 12.30 160 Math Growth 6.09 5.93 160 5.36 6.77 160

In order to determine if the Helping Hands Group differed in achievement from the control group, a MANOVA with GLM using Wilk’s Lambda was conducted. As shown in Table 4, no significant interactions were found. There were however significant main effects found for Exceptionality and SES. Table 5 shows the unique variance for each content level test at the Univariate level. For Exceptionality, Tukey’s Post-Hoc analysis shows that there was a significant difference in reading between all three groups of exceptional students, with the academically gifted group having a mean difference of +10.29 (SE = 3.33) and +16.84 (SE = 3.48) over the non-identified group and identified [with disability] group, respectively. For math, Tukey’s Post-Hoc analysis shows that the

academically group performed 14.30 (SE = 2.43) points better than the group not identified with a disability, while also performing 19.10 (SE = 2.54) points better than the group having a disability.

(54)

Table 4 (N=320)

Year 1

Multivariate Analysis of Variance of Mean Scores in Reading and Math

Source df F η2 p

Helping Hands 2002 (H) 2 1.86 .01 .16 Exceptionality (E) 4 3.13** .08 .01**

SES (S) 2 5.14** .03 .01**

H x E 4 1.83 .01 .12

H x S 2 .58 .00 .56

E x S 4 .67 .00 .61

H x E x S 2 .46 .00 .63

**p < .01.

Table 5 (N=320)

Year 1

Univariate Analysis of Variance of Mean Scores in Reading and Math

Source Dependent Variable df SS F p Helping Hands (H) Reading 1 86.17 .68 .41

Math 1 250.95 3.72 .06 Exceptionality (E) Reading 2 3247.41 12.88 .00**

Math 2 3056.94 22.64 .00**

SES (S) Reading 1 510.98 4.05 .05*

Math 1 629.33 9.32 .00**

H x E Reading 2 511.02 2.03 .13

Math 2 369.06 2.73 .07

H x S Reading 1 122.03 .97 .33

Math 1 40.12 .60 .44

E x S Reading 2 14.25 .06 .95

(55)

As stated previously, growth was calculated by using the residual of the test scores in reading and math from the previous year and the current year. This residual was used a dependent variable in the MANOVA model for this analysis. Table 6 shows that there were no significant interactions or main effects in growth for Year 1. Results at the Univariate level support these same results [See Table 7].

Table 6 (N=320)

Year 1

Multivariate Analysis of Variance of Residual Growth in Reading and Math

Source df F η2 p

Helping Hands 2002 (H) 2 .15 .00 .86 Exceptionality (E) 4 .24 .00 .92

SES (S) 2 .19 .00 .83

H x E 4 .35 .00 .84

H x S 2 2.58 .02 .10

E x S 4 .07 .00 .99

H x E x S 2 .60 .00 .55

Table 7 (N=320)

Year 1

Univariate Analysis of Variance of Residual Growth Scores in Reading and Math

Source Dependent Variable df SS F p

Helping Hands (H) Reading 1 23.88 .18 .67

Math 1 8.74 .21 .64

Exceptionality (E) Reading 2 50.34 .19 .83

Math 2 34.18 .42 .66

SES (S) Reading 1 7.59 .06 .81

Math 1 9.16 .22 .64

H x E Reading 2 126.73 .48 .62

Math 2 9.04 .11 .90

H x S Reading 1 .04 .00 .99

Math 1 174.16 4.27 .04

E x S Reading 2 21.84 .08 .92

Math 2 6.71 .08 .92

H x E x S Reading 1 121.65 .92 .34

(56)

Year 2 Results

Table 8 shows that the frequencies between the Helping Hands group and the Control Group are somewhat similar throughout. Again, as in the first year, the number of academically gifted students in the complete data set was rather small. Likewise, academically gifted students in the Control Group outnumbered those in the Helping Hands Group. Similar to Year 1, the Helping Hands Group also contained more free/reduced lunch students and fewer students who paid full price for their lunch. Overall, frequencies in the different cells are unequal, but the sample size in each group (N/2=224) is large. As with Year 1, the general linear models procedure will be used to account for these unbalanced cells.

Table 8 (N=448)

The overall mean score in reading for Year 2 (N=448) was 255.80, SD = 8.77. In addition, the overall mean score in math for Year 2 (N=448) was 261.92, SD = 8.96. Table 9 shows that although the Helping Hands Group did not perform better overall in reading or math, they grew more than the Control Group in math. Consistent with Year 1,

Frequency Table for Exceptionality and SES HH Group Control Group Exceptionality

Not Identified 178 169

Identified w/ Disability 42 37

AG 4 18

SES

Not Free/Reduced 83 116

(57)

the overall standard deviations for the Control Group were slightly higher. However, as shown by the histograms in Appendices H-I, there was only one extreme violation. Specifically, the Growth in Reading histogram demonstrates a violation in that the standard deviations are somewhat bimodal. Consequently, this set of “growth” means were excluded from the analyses.

Table 9 (N=448)

Year 2

Mean Reading and Math Scores For HH Group and Control Group Numeric Scores

HH Group Control Group

Subject M SD n M SD n

Reading 254.46 8.47 224 257.14 8.87 224

Math 261.25 8.22 224 262.59 9.62 224

Reading Growth n/a* n/a n/a n/a n/a n/a Math Growth 6.03 5.99 224 5.49 5.11 224 * Reading Growth was not calculated because the reading scale was changed between the 2000-01 and 2001-02 school-years.

(58)

over the non-identified group and identified [with disability] group, respectively. Similarly, Tukey’s Post-Hoc analysis shows that the AG group scored 15.54 (SE = 1.77) points higher in math than the group not identified with a disability, and 19.82 (SE = 1.94) points better than the group having a disability.

Table 10 (N=448)

Year 2

Multivariate Analysis of Variance of Mean Scores in Reading and Math

Source df F η2 p

HH03 (H) 2 .19 .00 .82 Exceptionality (E) 4 15.08** .07 .00**

SES (S) 2 .52 .00 .60 H x E 4 1.90 .00 .11 H x S 2 .59 .00 .56 E x S 4 .67 .00 .61 H x E x S 2 .99 .00 .41 **p < .01.

Table 11 (N=448)

Year 2

Univariate Analysis of Variance of Mean Scores in Reading and Math

Source Dependent Variable df SS F p Helping Hands (H) Reading 1 .60 .01 .92

Math 1 9.14 .14 .71

SES (S) Reading 1 25.33 .41 .52

Math 1 66.42 1.03 .31 Exceptionality (E) Reading 2 1511.47 24.64 .00**

Math 2 1711.25 26.51 .00**

H x S Reading 1 65.98 1.08 .30

Math 1 57.42 .89 .35

H x E Reading 2 214.29 3.49 .03*

Math 2 46.12 .71 .49

S x E Reading 2 65.86 1.07 .34

(59)

In year 2, reading scales were modified2, resulting in inconsistent deviations between years and thus were not included in this analysis. Since reading growth was not calculated for Year 2, an analysis of variance (ANOVA) was conducted for math. The overall model was significant, F(11, 436) = 2.80, p < .01, η2 = .07. But once again, Exceptionality was the only significant main effect for growth in Year 2 [See Table 12] accounting for most of this variance.

Table 12 (N=448)

Year 2

Univariate Analysis of Variance of Residual Growth in Math

Source df SS F p

HH03 (H) 1 14.00 .47 .49

Exceptionality (E) 4 66.35 8.39** .00** SES (S) 1 497.75 2.24 .14

H x E 2 34.03 .65 .52

H x S 1 38.72 1.15 .28

E x S 2 7.16 .12 .89

H x E x S 2 56.06 .95 .39

**p < .01.

Year 3 Results

Figure

Table 3 shows the mean scores in reading and math by HH Group and Control
Table 4 (N=320)
Table 9 shows that although the Helping Hands Group did not perform better overall in
Table 11 shows the unique variance for each content level test at the Univariate
+2

References

Outline

Related documents

The use of sodium polyacrylate in concrete as a super absorbent polymer has promising potential to increase numerous concrete properties, including concrete

The paper assessed the challenges facing the successful operations of Public Procurement Act 2007 and the result showed that the size and complexity of public procurement,

To address some of the aforementioned gaps, the re- search team initiated an investigation using Concept Mapping method to gather perspectives from a multidis- ciplinary group

We suggest the following: 1) The urothelium can be selec- tively removed by a gentle swabbing motion that leaves the deeper suburothelial layers intact; 2) Removal of the

Outcomes of laparoscopic radical nephrectomy in the setting of vena caval and renal vein thrombus: seven-year experience. Sundaram CP, Rehman J, Landman J, Oh J: Hand

Considering only women who received abortion care from public facility where there is shortage contraception supplies in study conducted in two of the region and failing to

The aim of the study was to assess the presence of pathogenic microbes on treatment tables in one outpatient teaching clinic and determine a simple behavioral model for

It was decided that with the presence of such significant red flag signs that she should undergo advanced imaging, in this case an MRI, that revealed an underlying malignancy, which