. ‘L .. .2.. v1.21 3.”? .LIBRARY Michigan State University This is to certify that the dissertation entitled CORRELATES OF ACADEMIC ACHIEVEMENT AMONG AFRICAN AMERICAN BOYS presented by TIA RENEE WALTON RAMSEY has been accepted towards fulfillment , of the requirements for the PhD. degree in Family and Child Ecologx Major Pip/feésor’s Signature (P/zr/o 7 f / Date MSU is an affirmative-action, equal-opportunity employer ..-.-._._._.-._.-.-.-.-.-.-.-.-.-.-.-.—.-.-.-._._._.-.-._.-.-.-.-....-._.-._.-.-.-.-.-.-.-._.-.-.-.-.-.-.-._.—.-.-._.-.-.-.- PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/07 p:/CIRC/DateDue.indd‘p.1 CORRELATES OF ACADEMIC ACHIEVEMENT AMONG AFRICAN AMERICAN BOYS By Tia Renee Walton Ramsey A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of family and Child Ecology 2007 ABSTRACT CORRELATES OF ACHIEVEMENT AMONG AFRICAN AMERICAN BOYS By Tia Renee Walton Ramsey This dissertation investigated the extent to which neigthrhood context (safety/quality), school context (perceptions of school climate), church (religious involvement), peers (negative peer characteristics), extracurricular activities (involvement in extracurricular activities), and family characteristics/processes (income, family structure, father involvement, home environment, parent involvement in school, parents educational expectations of the child) were related to academic achievement among African American males. Achievement levels in math and reading were divided into three groups (high, medium, low) for comparison. Two models were used as frameworks for investigating the significance of these contexts/processes on academic achievement in math and reading across the multiple contexts. Bronfenbrenner’s ecological model and Clark’s model of school success and failure were. used to assess the factors that proved to be significant in the achievement of African American males. This study used a subset of data from the Child Development Supplement II (CDS-II) which is one research component of the Panel Study of Income Dynamics, (PSID), a longitudinal study of a representative sample of US. individuals and families. Data for this study was pulled from 228 African American boys ages 12-18 and their parent/caregiver. Many of the predictors of achievement were found to correlate significantly with each other; however, only two of the eleven predictors of achievement were found to be significant in the final regression analyses. Family structure (one parent/two parent) was proven to be a significant predictor for math achievement. Parent level of educational expectations was shown to be a significant predictor of achievement for both math and reading. To Sterling, Evan and Isaiah who are my greatest inspirations iv ACKNOWLEDGEMENTS This has been a long journey. There have been a lot of ups and downs throughout this process. It has only been my strong belief in God and the power of prayer that has gotten me through it all. I would like to thank my committee members for hanging in there with me through it all. I would especially like to thank Dr. Tom Luster, my advisor for not giving up on me. TABLE OF CONTENTS LIST OF TABLES ................................................................................ viii LIST OF FIGURES ................................................................................. ix CHAPTER 1: INTRODUCTION .................................................................................... 1 Purpose of the Study ................................................................................. 2 Objectives of the Study ............................................................................. 4 Theoretical Model ................................................................................. 4 Conceptual and Operational Definitions .......................................................... 9 Rationale for the Present Study ................................................................ 11 CHAPTER 2: ' REVIEW OF THE LITERATURE ............................................................... 14 African American Males and Achievement ..................................................... 14 Factors Influencing High Achievement of African American Males.. ....................... 14 Factors Influencing Low Achievement of African American Males. . . . . . . . . . . . . . ........16 Contextual Process Factors Influencing Academic Achievement ............................. 20 Neighborhood ................................................................................... 21 School ............................................................................................ 21 Church/Religious Involvement ................................................................ 22 Peers ................................................................................................ 23 Extracurricular Activities ...................................................................... 25 Family Characteristics and Processes ........................................................ 26 Research Hypotheses ........................................................................... 29 CHAPTER 3: METHODOLOGY ..................................................................................... 34 Research Questions ................................................................................. 34 Assumptions ........................................................................................... 34 Sample ............................................................................................. 35 Research Design and Procedure .................................................................. 36 Research Variables and Measures .............................................................. 37 Reliability Analysis ofVanables43 Data Analyses43 CHAPTER 4: RESULTS ............................................................................................. 46 Descriptive Statistics ............................................................................ 46 Correlations Among Variables .................................................................. 49 Correlations Among Predictor Variables ...................................................... 50 Correlations Between Predictor variables and Outcome Variables ........................ 56 vi One-way-AN OVA ................................................................................. 56 Multiple Regression ............................................................................. 62 Post Hoc Analyses .............................................................................. 65 Review of Research Hypothesis ............................................................... 66 CHAPTER 5 DISCUSSION AND CONCLUSIONS .......................................................... 72 Summary of Findings ............................................................................ 72 Limitations ofthe Study/Implications for Future Research......................................82 Conclusions ....................................................................................... 86 APPENDIX A: Items fiom measures used to assess each construct ......................... 88 BIBLIOGRAPHY ................................................................................... 91 vii LIST OF TABLES Table 1: Clark’s Five Categories of Activities that Shape Success or Failure of the Learning Process ................................................................................. 8 Table 2: Six Contexts and Factors for the Outcome of Academic Achievement ......... 11 Table 3: Reliability Measures of Variables Using Cronbach’s Alpha ...................... 43 Table 4: Sample Demographic Characteristics ................................................ 48 Table 5: Descriptive Summary of Predictor Variables ........................................ 49 Table 6: Zero-Order Correlations Among the Predictor Variables for Achievement. . .51 Table 7: Correlations Between the Predictor Variables and Achievement for Reading and Math .................................................................................... 52 Table 8: Crosstabs/Chi Square Analyses for Family Structure and Achievement ......... 57 Table 9: One-way ANOVA: A Comparison of the Three Achievement Groups (Low, Medium, High) on the Continuous Predictor Variables for Math ............................. 58 Table 10: One-way AN OVA: A Comparison of the Three Achievement Groups (Low, Medium, High) on the Continuous Predictor Variables for Reading ......................... 60 Table 11: Multiple Regression Analysis: Predictors of Achievement Standardized Beta Coefficients .......................................................................................... 63 Table 12: Multiple Regression Analysis: Predictors of Reading and Achievement. . . ..63 Table 13: Multiple Regression Analysis: Predictors of Math and Achievement... ......64 viii LIST OF FIGURES Figure 1: Theoretical Model ..................................................................... 5 ix Chapter 1 Introduction Thirty-two percent of African American males between the ages of 20-35 are in contact with the criminal justice system compared to only 6% of White males and 8% of the Latino male population (Winbush, 2001). It is projected that by 2020, two out of every three Afi'ican American males will be involved in the criminal justice system (Kunjufu, 2001). The leading cause of death among Black males ages 15-24 is homicide. Fifty-four percent of the deaths among Black males ages 15-24 are due to homicide compared t012.2% for White males (National Vital Statistics Reports, 2000). These statistics shed light on the social plight that many African American males are faced with in America. As African American male youth strive to succeed in school, one must understand the social pressures that surround them. Some youth will beat these negative odds, while others will not. For those African American youth who fail in school, there is a very good chance that they may succumb to this plight. It is the goal of this study to help society understand the strengths and weakness that African American male youth possess within themselves, their families, and their community. Why are so many African American male students failing in school? Why are many other African American male students succeeding in school? Several key statistics indicate that many African American males are not faring well in school. One out of every ten African American male high school graduates enrolls in college (Kunjufu, 2002). African American male youth are considered to be a “lost generation” plagued by low academic achievement (Kunjufu, 2002). Findings indicate that African American males attend college less than African American women and far less than White males. The national drop-out rate for Afiican American males continues to increase and has reached a high of 49.6% within the last decade (US Census, 2000). These disturbing statistics have cast perceptions of hopelessness, helplessness and educational despair by American society toward African American males. Most studies investigating the academic achievement of African American male students have focused on poor outcomes. Very few studies emphasize the success of African American male students. There is a need to expand the existing knowledge of the factors that influence the academic success and failure of Afi’ican American boys. Moreover, it is important to consider achievement from an ecological perspective and to explore potential influences in the multiple contexts that are likely to influence the 7 development of African American boys. There is a need for fiuther research looking at factors which affect the academic success of African American male students. There has been a great deal of research investigating African American children and academic achievement. However, despite the educational dilemma of African American boys, there has been little research on factors affecting their academic success (Banks, 1998; Boykin, 1983; Delpit, 1988; Fordham and Ogbu, 1986; Hale-Benson, 1986; Shade, 1982; Willis, 1989). Purpose of the Study Some African American male students are succeeding in the classroom. What factors have allowed them to succeed? This study identified the factors that influenced the academic achievement of high-achieving Black males and low-achieving Black males. In identifying these key factors, the study provided new information on how the academic achievement levels of African American males can be improved. This study also helped us to understand the supportive structures across multiple contexts that allow some African American male students to succeed in the classroom. Six contexts were assessed along with several factors that are believed to be key areas associated with the academic achievement of Black males. The six contexts are: neighborhood, school, church, peers, extracurricular activities, and family. Black males ages 12-18 years old were examined along with responses regarding their behavior from their parents and caregivers. The data source is the Panel Study for Income Dynamics (PSID) Child Development Supplement Wave II (CDII). . The purpose of this study was to investigate the factors associated with the six contexts that were believed to be related to the academic achievement of high, medium and low-achieving African American male students. Identifying these key factors is intended to shed light on a unique area of research that has yet to be uncovered and could contribute to improving the academic performance of Afi'ican American males. This study attempted to determine the extent to which neighborhood context (safety/quality), school context (child’s perceptions of school climate), church context (child’s religious involvement), peer context (negative peer characteristics), extracurricular context (child’s involvement in extracurricular activities), and family context (income, family structure, father involvement, home environment, parent involvement in school, parents’ educational expectations) were related to academic achievement among African American males (See Figure 1). This study focused specifically on African American males because they are currently significantly lagging behind African American females and all other ethnic groups in achievement (Kleinfield, 1999; Pollitt, 1999; Herbert, 1998; Young, 2001). Previous research has identified some factors that are related to the academic achievement of African American males; these factors were examined in this study. However, there is surprisingly little research on this important topic. Objectives of the Study The objectives of this study are: 1. To understand the differences in the contexts of low, medium and high-achieving African American males. 2. To determine which factors are significant within the various contexts of the lives of African American males for successful achievement. 3. To examine the additive effects of variables across human ecological systems in order to shed light on the academic success or failure of African American males. Theoretical Model This study incorporated components of Bronfenbrenner’s model of human development and Clark’s model of school success and failure (see figure 1). These two theoretical models allowed us to better understand two things. First, Bronfenbrenner’s model helped us understand how the overall development of the African American male child may be influenced by several interrelated aspects of his environment. Second, Clark’s model showed us how academic success or failure can be predicted by the amount of constructive learning experiences from the environment. Both models emphasize the importance of contextual factors on a child’s development. Bronfenbrenner’s ecology of human development was the primary theory underlying this study. He suggests that research on human development should incorporate an awareness of the environmental systems within which people are fiinctioning. Families and children live in multiple physical and social environments. Figure 1. CORRELATES OF ACHIEVEMENT AMONG AFRICAN AMERICAN BOYS WITH VARIABLES:AN ECOLOGICAL MODEL Neighborhood Context -safetquuality School Context perceptions of school climate Church -religious involvement Achievement -Reading Peers - -Math -negative peer characteristics Extracurricular Activities -invoIvement in extracurricular activities Family Characteristics and Processes -income -family stmcture (single parent vs. 2 parents) -home environment -parent involvement in school -parents' educational expectations -father involvement Bronfenbrenner’s (1979, 1986) model of human ecology focuses on the individual and his or her relationship with the environment. This model provided the basis for understanding how the family, environment and society impacted the development of the child. Bronfenbrenner divided the environment into four levels or systems: microsystem, mesosystem, exosystem, and macrosystem. The developing child moves in and out of various contexts as it interacts with the environment. In order. to gain a full picture of the factors contributing to successful and poor outcomes of African American boys in school, one must review each of the ecological systems as well as individual characteristics that affect their success. The microsystem consists of immediate contexts that individuals experience in various degrees depending on their stage of development. As the child matures and is exposed to different settings and individuals, this level becomes more complex. At this level, children are influenced by others, and they also influence others within their microsystems (Shaffer, 1994). This study examined variables from the following microsystems: neighborhood, peers, school, and family. It also examined individual differences in the child’s social, community and religious involvement. The interconnections among microsystems are referred to as mesosystem. Bronfenbrenner asserts that children’s development is likely to be enhanced by supportive links among microsystems (Shaffer, 1994). This study examined several interconnections that affected the academic success and failure of Afiican American males. These interconnections include parental involvement in school and child participation in extracurricular activities at church, school and in the neighborhood. The exosystem consists of settings that children typically do not experience directly but may still affect their development (Kohn, 1977). This study investigated settings that may influence the academic success and failure of African American males such as the church and the community. The child’s religious involvement could be influenced indirectly by programs developed and endorsed by the church advisory board. The macrosystem refers to the values, beliefs, rules, customs, and ideology that determine how children should be treated, what they are taught, and which goals are most important. These beliefs and values differ across cultures and can significantly influence the types of experiences children face in their homes, schools, neighborhoods, and other contexts that directly or indirectly influence them. In this study, the beliefs, values and ideology of African American males and their parents will be studied. These beliefs, values, and ideology may influence the type of home environment, level of parent educational expectations of the child, the degree of parental involvement in school and the child’s level of religious involvement. This study examined the additive effects of variables across these human ecological systems in order to shed light on the academic success and failure of African American males. Clark’s (1990) model for the academic success and failure of children suggests that social circumstances do not have an independent effect on one’s life achievements. Their effect is mediated by other behavioral and attitudinal variables. These variables are sometimes referred to as cultural and process factors or developmental-ecological factors (Ogbu.l986). Clark believes that achievement is influenced by the child’s interpersonal communication in everyday life, and this communication occurs in a variety of ecological or environmental contexts that help them learn and practice the prerequisite skills needed for high achievement. These contexts include: homes, schools, neighborhoods, and other community settings (church, recreation centers, libraries, museums, and playgrounds). Clark believes that children who do constructive learning activities outside of school. are more likely to be high achievers. While those who are not stimulated by these types of constructive learning activities outside the home are more likely to be low achievers. He suggests that there is a “mental workout” that takes place as children are engaged in various mentally stimulating activities which lead to more successful students. Clark discusses five categories of activities that shape success or failure in the learning process (see Table l). Table I. Clark’s Five Categories of Activities that Shape Success or Failure of the Learning Process Categories Activities Professimigl Tutoring progrgms, study groups, book clubs Out of School Homewglg, musical lessons, volunteer erk, and apprenticeships Leisure Board games, reading a book, organized youth programs, scouts_ Recreational Watching television, listening to music, biking, jogging, group sports, “hanging out” Health Maintenance Exercise, eating, daydreaming, sleeping, grooming Clark’s five categories of learning activities are: (1) Professionally guided, formal learning activities (tutoring programs, study groups, book clubs); (2) Deliberate out of school learning and work activities (homework, musical lessons, volunteer work, and apprenticeships); (3) High—yield leisure activities (board games, reading a book, organized youth activities, scouts); (4) Recreational activities (watching television, listening to music, biking, jogging, group sports, “hanging out”); (5) Health maintenance activities (going to church, exercise, eating, daydreaming, sleeping, grooming). Clark’s model was useful in this study because there were several factors being investigated which fell under one of the five categories. Neighborhood safety/quality can affect whether a child stays in the house to read a book or plays baseball at the local park. Positive or negative perceptions of school climate can affect a child’s school attendance or participation in school activities. Clark considers church involvement and extracurricular activities as “constructive learning” activities. Peer and family characteristics can be associated with the degree to which students are involved in leisurely activities inside or outside of the home. As a result, the factors that determine the academic success or failure of African American male students can be easily discussed using this model. A limitation to Clark’s model is that it does not address the effects of the child’s perceptions and beliefs on his academic success or failure. Children bring family characteristics and processes (income, family structure, parenting practices) to the table which influences how they see the world. These personal characteristics may affect their religious involvement, peer relationships, involvement in extracurricular activities, and academic achievement. Parents may also bring personal beliefs on their involvement in school and their educational expectations of their child that may influence a child’s success or failure in the classroom.iBronfenbrenner’s model addresses Clark’s theoretical limitation in that the child’s development is believed to be influenced by the personal characteristics of the child, family, peers, and any other individuals that are a part of the child’s environment. Conceptual and Operational Definitions The next section provides the conceptual and operational definitions for the variables in this study. Achievement level of students. The achievement level of students is determined by the student’s test scores in math and reading. For some analyses, the scores were divided into three levels (high, medium, low) and for other analyses achievement is treated as a continuous variable. The scores on the Woodcock Johnson Revised Tests of Achievement (WI-R) were used. . Neighborhood context. Neighborhood context is the parent’s perceptions of neighborhood safety and quality. The parent’s score on the residential stability subscale was used. School context. School context is the student’s perceptions of the school climate and connectedness. The student’s score on the subscale measuring school climate was i used. Church Church context is the student’s perceptions of religion and involvement in various religious activities. The student’s score on the religious subscale was derived from the Child Development Supplement Child Interview. Peers. Peer context is the involvement of peers in negative or risky behaviors. It was measured by the student’s score on subscale measuring negative peer characteristics: disobedience, gang involvement, do dangerous things, trouble in school, fights with other kids, and drink alcohol regularly. Extracurricular activities. Extracurricular activities are the student’s involvement in academic, athletic, and volunteer activities. It was measured by the student’s score on the subscale measuring the student’s involvement in various activities. This subscale is derived from the Child Development Supplement Child Interview. Family characteristics/processes. Family context is divided into characteristics and processes. The student’s family characteristics are: household income, family structure, and home environment. Income is total family income based on parent report for the year 2002. Family structure is whether or not the student comes from a single or two-parent home. Home environment is measured by the parents’ score on the “Home Observation for Measurement of the Environment-Short Form” (HOME-SF). The 10 students family processes are: parent involvement in school, father involvement, and parent’s educational expectations for the child. Parent involvement in school is the level of parent participation at the child’s school. Father involvement is the level at which fathers participate in their son’s lives. Parents’ educational expectations for child are the parents’ perceptions of the level of anticipated success they have for their child. These family processes are measured by three subscales derived from the Child Development Supplement Primary Caregiver-Child Interview. Three separate scores were obtained. Table 2. Six Contexts/F actors for the Outcome of Academic Ac'mevemcz: Context/F actors Instruments Neigiborhood Safety/Quality Pmt/Caregiver interview Family Contexts Income Parent/Caregiver interview Single / Two—Parent Home Child interview Home Environment HOME-SF Processes Parent Involvement in School Parent Educational Expectations Parent/Caregiver interview Parent/Caregiver interview Father Involvement Child interview Church Religious Involvement Child interview School School Climate Child interview Peers Peer Negative Behavior Child interview Extratcurricular Involvement in activities Child Interview Outcomelair'ables Student Achievement Reading Woodcock Johnson Math Woodcock Johnson Rationale for the Present Study There has been a great deal of research investigating African American children and academic achievement; however, despite the educational dilemma of Afiican American 11 males, there has been little research on factors affecting the academic achievement of African American males. For several decades, social scientists and educational researchers have contributed numerous theories to explain variations in the academic performance of Black youth. The body of scholarly literature in the 19405 and 19503 presented ethnic self-hatred theories suggesting that the reason for academic failure in African American youth was due to low self esteem based on negative perceptions of their ethnicity, while researchers in the 19603 focused on the cultural deprivation theory in explaining the fi'ustrating cycle of failure in the Black school experience suggesting that academic failure was due to inferior aspects of the Black culture (Deutsch, 1967; Lewis, 1969; Reissman, 1962). In the 19703, researchers offered theories centered on parenting styles, family stability, and single parenting which viewed African American families as weak and disadvantaged , and in the 19808, researchers continued to search for answers by examining social forces, such as discrimination, prejudice, low socioeconomic status, and the introduction of cultural differences which encouraged society to start looking at issues outside of the Afiican American family for answers to academic failure (Boykin, 1983; Fordham & Ogbu, 1986; Gay, 1993; Hale-Benson, 1986; Shade, 1982; Willis, 1989). In the 1990s, cultural variables in the classroom rose to the forefront among researchers as factors greatly influencing the achievement of Black boys focusing on the strengths that they bring to the classroom (Banks, 1998; Delpit, 1994; Hale, 1994; Hare & Hare, 1991; Irvine, 1990; Ross & Jackson, 1991). Currently, the limited research available on the academic achievement of Black males focuses primarily on problems or poor outcomes. These factors include: the disproportionate placement of Black males in special education, suspension, and 12 expulsion (Dunbar,1999; Harry & Anderson, 1999; Herbert, 1998; Rodney, Rodney, & Laxley, 1999; Spencer, Beale, Dupree, & Swanson, 1998; Spencer & Oates, 1999; Townsend, 2000). Regardless of the various directions research has taken during each time period, there is one common theme shared by all. Black males have been plagued by underachievement for decades, and this educational dilemma continues today. However, despite the negative trend, Luster and McAdoo (1994) believe that it is imperative that researchers not stereotype Afiican American children by focusing on low- achievement and value their academic diversity. The focus should be shifted to answer the question of why so many African American students do well in school while others struggle. This study differed from earlier studies on the achievement of African American males in that it compared factors related to high, medium, and low-achieving Afiican American males. It utilized data from the last wave of data collected from year 2002 obtained from a longitudinal study which began in 1968. Also, this study applied an ecological view of the relationships among variables assessed by the child and the parent across multiple contexts (home, neighborhood, school, and church). Finally, the objectives of this study were addressed. This study helped us understand the differences in the contexts of low, medium and high achieving Afiican American males. The data determined which factors were significant within the various contexts of the lives of African American males for successful achievement. Also, this study examined the additive effects of variables across human ecological systems in order to shed light on the academic success or failure of Afiican American males. 13 Chapter 2 Review of Literature The purpose of this section is to discuss two important aspects of the literature for this study. First, this chapter presents significant research findings regarding the academic success and failure of Afiican American males. Second, this chapter presents a review of the studies that examined factors shown to influence the academic achievement of African American males. These factors were grouped into six primary categories: neighborhood context, School context, church, peers, extracurricular activities, and family. Studies were discussed to present significant findings pertaining to several factors in order to justify their inclusion in this study. A fi'ican American Males and Achievement The following studies provide a foundation for the understanding of African American males and achievement. This section reflects the discovery that there is very little research on Afiican American males and academic achievement, particularly research that investigates their academic success. As a result, significant studies are presented that show factors affecting the academic achievement of African American children with key findings extrapolated to suggest how they relate to Afi'ican American males in the classroom. Factors affecting the high and low achievement of African American males will be discussed. Factors Influencing High Achievement among African American Males Some studies have shown that there are several factors that contribute to the academic success of African American males. Ladson-Billings (1994) found that the 14 performance of Afiican Americans is influenced to a large degree by the social support and encouragement they received from teachers. Noguera (2000) also believes that successful achievement patterns in African American male youth can be accomplished if there is a willingness to provide the resources and support necessary to create the conditions that nurture academic success. In a survey of 537 seniors at an academic magnet high school, responses from Black males to the question “my teachers support me . and care about my success in their class” was at 42% for “strongly disagree” and 38% for “disagree”. These findings suggest that many African American males do not feel supported by teachers which may lead to a negative classroom climate that may affect their level of achievement. In the area of parents’ educational expectations, studies have shown that African American males have high expectations that may have been instilled into them by parents. Noguera (2000) studied high school students in Northern California and found that 90% of the Black male students “agreed” or “strongly agreed” that education is important and that they want to go to college. This finding suggests that African American male youth have positive attitudes about education and have high educational aspirations. Are those positive attitudes about education and high educational aspirations influenced by their parents? Luster and McAdoo (1996) looked at long-term effects of African American children’s participation in the Perry Preschool project. This study documented a positive role that African American families played in the lives of their children. Even though all of the families in this study were low SES, many parents helped their children to overcome obstacles associated with low SES and obtain higher levels of education than 15 they themselves had. This finding shows that an African American male student’s level of education can be influenced by their family’s positive or negative perceptions of achievement. Herbert (1998) conducted a qualitative study on two high achieving African American male high school students. He found several characteristics that contributed to their success such as: a strong sense of self, strong aspirations, supportive family members, positive role models, involvement in extracurricular activities, and an appreciation for diverse cultural backgrounds. These findings provide insight into the necessary factors that will assist Afiican American male youth in becoming high academic achievers. Factors Influencing Low Achievement of African American Males The underachievement of Black youth is still a popular topic of research; however, the underachievement of Black boys specifically is still limited (Dunbar, 1999; Harry & Anderson, 1999; Herbert, 1998; Rodney, Rodney, & Laxley, 1999; Spencer, Beale, Dupree, & Swanson, 1998; Spencer & Oates, 1999; Townsend, 2000). Although limited, there is still currently more research on the failure of Afiican American male students rather than on their success. The lack of research on the achievement of Black boys is puzzling given the recent statistics that show serious problems in their educational attainment (Herbert, 1998; Kleinfield, 1999; Pollitt, 1999, Young, 2001). Much of the current literature on the low achievement of African American males focuses on the disproportionate placements in special education programs and very little attention is given to identifying the behaviors that actually cause this dilemma. As a result, there is a major gap in the literature. 16 African American children comprise 17% of the US student population but constitute almost 40% of the students placed in a range of special education categories. If an Afiican American child is placed in special education, 80% of the time the child will be male. Only 3% of Afiican American children are placed in gifted and talented programs (Kunjufu, 2002; US. Department of Education, 2000). Nationally, Black boys score lower than any other group on standardized tests and are three .times more likely than their White American counterparts to be misplaced in special education or classes for slow learners (Rodney, Rodney, & Laxley, 1999). They are retained in grade more often than their White classmates, particularly at the elementary levels (Rodney, et al., 1999). The drop-out rate for Black boys is disproportionately higher than that of other ethnic groups (Rodney, et al., 1999). They are also suspended or expelled from school at a higher rate than any other ethnic group (Rodney et al.1999). These statistics are important; however, they only provide a description of a portion of the problem faced by low-achieving Afiican American males. Current literature does not provide explanations as to why the dispr0portionate placement and poor performance of many Afiican American males in the classroom occurs. A study by Dunbar (1999) examined the rates of school retention, suspension, expulsion, academic achievement, grade attainment, school attendance and participation in extra-curricular activities by Black male students. When compared to students from all other ethnic groups, Black male students showed the highest school drop-out rate, lowest high school graduation rate, low achievement expectations by teachers, lowest enrollment in gified programs, and highest rate of special education placement. Black males accounted for 43% of the total New Orleans public school population but were 17 disproportionately represented in non promotions (57.5%), suspensions (65%), and expulsions (80%) (Townsend, 2000). The use of suspension and expulsion with Black students has wide-ranging consequences. Among the most obvious is the denial of access to learning opportunities that occurs when students are not in school. Students who receive out of school suspensions or expulsions typically are not provided opportunities to continue their schoolwork (Dunbar, 1999). Suspended and expelled children and youth are at greater risk for encountering the legal system. When asked to inform teachers concerning effective strategies for making school a better place for them, they invariably suggested that teachers get to know them and what their lives are like when they are not in a school setting (Townsend, 2000). These findings suggest the need for more research in the areas of teacher’s academic expectations. Also, the perceptions of African American male students’ regarding school climate should be investigated further. White-Hood (1994) suggested that gender is a critical component in teacher’s perceptions of the firture outlook of Afiican American adolescents. Teachers believed that African American males lacked future goals, interests, and had no opportunity to break through barriers. Teachers in this study also felt that African American males tended to exhibit fewer academic aspirations, are less likely to complete high school, and are at greater risk of dropping-out compared to females. These findings suggest the need to further explore teacher’s expectations toward the academic success or failure of African American male students to determine whether or not their expectations affect the performance of students. Spencer et a1. (1998) believe that the transitional nature of adolescence is potentially difficult for Black males because they must struggle not only to address the normative 18 tasks of adolescence, but also to develop a healthy self-identity in the face of a dominant culture that does not expect them to succeed. Parents are vital in helping their children develop a positive self-identity. This study will investigate how parenting practices and parent’s educational expectations of African American male students relate to their achievement. Among African American youth educational research has shown a consistent gender gap in achievement. African American girls appear to be faring much better than African American boys in school. The gender gap in higher education is significant among Blacks. From 1977 to 1997, the number of bachelor’s degrees awarded rose by 3% for Black men but by 77% for Black women; among 1996-1997 college graduates, Black women outnumbered men almost 2 to 1 (Young, 2001). Black college women (62%) vastly outnumber Black college men (37%) (Kleinfield, 1999; Pollitt, 1999). What characteristics must Afiican American males possess in order to combat the threat of academic failure? Honora (2002) examined the connection between future expectations and school achievement among low-income, urban Afi'ican American adolescents. Higher achieving adolescent girls expressed more future goals and expectations and considered more long- term goals than higher achieving boys. High achieving girls set more goals related to college, employment, marriage, and family planning. High achieving boys set more goal related to sports and leisure I Polite and Davis (1999) presented educational experiences discussed by several authors that were believed to be most critical to the betterment of Afiican American men. The authors proposed that these critical experiences affect educational attainment, job 19 procurement, quality of life, and responsible fatherhood. They frequently found that the reasons for the alienation and poor academic achievement of some Afiican American males is that they perceive most school experiences as irrelevant to their masculine identity and development. Polite and Davis (1999) also argued that schools and teachers impose a feminine culture on males that unintentionally produces oppositional behavior. Many African American male students are in schools where the majority of their teachers are female. Female teachers may encourage their male students to exhibit behaviors similar to females and reward them for being less aggressive and less active in the classroom. This study investigated the behavior of Afiican American male students and their perceptions about school. Contextual and Process Factors Influencing Academic Success A The purpose of this study was to identify the factors that may influence the academic achievement of Afiican American boys. An ecological approach was used to assess the many contextual and process factors across various settings. This section presents a review of several studies that examined the six main contexts that have been shown to influence the academic success and failure of African American males. Each main context has factors that provide more specific information. The six main contexts and their factors are as follows: neighborhood context (safety / quality), school context (perceptions of school climate), church (religious involvement), peers (negative peer characteristics), extracurricular activities (involvement in extracurricular activities), and family (income, family structure, father involvement, home environment, parent involvement in school, parents’ educational expectations). This study examined the additive effects of variables across human ecological systems in order to shed light on the 20 academic success or failure of African American males. Key findings from research have led to the development of several hypotheses that are centered on the six main contexts that this study investigated. Neighborhood Context Research has shown that neighborhood safety and quality may influence the academic achievement of African American boys. German-Smith and Tolan (1998) evaluated the relations between exposure to violence, family relationship characteristics and parenting practices, and aggression and depression symptoms among 243 inner-city Black and Latino boys aged 11-15 years. Results showed that exposure to community violence was related to increases in aggressive behavior and depression over a one year period. Ainsworth (2002) found that neighborhood characteristics predicted educational outcomes in his study using the National Educational Longitudinal Study of 1988 linked to 1990 census information. Wacqant (1966) argued that students from disadvantaged neighborhoods were more likely to attend inferior schools that do not focus on teaching and learning which may affect the academic performance level of its students. These findings suggest the need for more research in the areas of neighborhood safety and how it may influence the academic achievement of African American males. School Context Student perceptions of the school climate may influence the academic achievement of African American males. It is important to determine how the student’s connectedness to school affects his level of academic success. Herbert (1998) found that Black males exerted less effort academically and maintained more negative attitudes about school than females. They also found school less relevant and less personally meaningful than did 21 females and were more pessimistic about social factors. Another study showed that Black boys had more positive beliefs about their abilities and expectancies for success in math and science than did girls, and boys valued science more than did the girls (McClendon & Wigfield, 1998). However, research also showed that Black boys lagged behind girls in verbal skills and believed that schools are hostile toward them (Kleinfield, 1999). If Black boys have negative perceptions of the school environment; how does this affect their level academic achievement? These findings suggest the need for investigating Black male students’ beliefs and perceptions about school climate and how it may affect their academic success. Martin, Cotton and McCarrahesby (1996) examined adolescents’ emotional and behavioral perceptions concerning the presence of guns in their school. Three hundred and seventy-six middle school students were interviewed for the study. Thirty—six percent of the students felt afraid someone would hurt or attack them at school. Boys were more likely than girls to feel afraid. A nationwide survey of US. students showed that 8% of girls and 30% of boys carried weapons to school (Martin, et al., 1996). Racial differences of the boys were not assessed in this study. 'Afiican American male student’s have concerns about their school safety and classroom climate. These findings imply that more research needs to be conducted in order to understand how classroom safety may influence the academic success or failure of Afiican American male students. Church/Religious Involvement Many churches provide supportive services and activities for youth. Churches and religious organizations are becoming more recognized as partners in steering youth away from crime and drug use and toward academic achievement (Freeman, 1986; Johnson et 22 al.., 2000). A study by the Morehouse Research Institute (1999) found that religious involvement contributed more to the academic progress of adolescents from low-income neighborhoods than those from higher-income neighborhoods. Religious involvement was also associated with higher parental educational expectations and successful degree completion. Church attendance may promote characteristics such as: diligence, commitment, and routine which are helpful skills that can be utilized for school success. Many churches and community organizations have stepped in to provide academic support and adult mentors outside of school through programs such as Simba and Omega Boys Club (Watson & Smitherman, 1996). This program showed that church attendance affected parent educational expectations and achievement. Evidence suggests that high- achieving African American adolescents are more likely to have particular religious values (Ginsburg & Hanson, 1986). Gary and Brown (1991) found in their study that as levels of religious involvement increased, educational attainment among African- Americans students also increased. These findings suggest the need to understand the influence of religion/church attendance on the academic development of African American male students. Peers Context One important factor that may influence the academic achievement in African American males is their relationship with peers. The effects of negative peer behaviors on the high and low academic achievement of Afiican American males were explored in this study. Osbourne (1997) compared Black, White, and Latino male students and found that Black male students experienced an “emotional withdrawal” from academics at a greater 23 rate than do Whites and Latinos. Osbourne also found that Black boys exhibiting intellectual behavior in the classroom are often teased by peers who view it as too feminine in nature. This perception is due to the cultural emphasis that African American males have on macho behavior. However, even though the grades of these Black boys plummeted, their self-esteem remained the same over four years and in some cases increased. Osboume’s explanation for this finding was that academic self-esteem is one facet of overall global self-esteem. Therefore, even though a boy’s self-esteem may have gone down in the area of academics, the rest of his self-esteem could remain intact. In the book, The Warrior Method by Raymond Winbush (2001), psychologist Claude Steele (1992) found that powerful stereotypes over “being smart” in the Afiican American community triggered anxiety and academic withdrawal on the part of African American boys in the late elementary and middle school grades. African American boys in these grades were ridiculed and deemed “less masculine” for earning good grades. These boys often struggle between making good grades and being accepted by fiiends who see high grades as “acting white.” Steele believes that the sharp decline in Black boys’ academic achievement is due to factors associated with their transition into puberty. There is peer pressure to perform athletically and a stigma is attached to a Black boy who excels academically. Sociometric surveys were completed at third grade for a low SES, urban sample of Black youth and delinquency data were gathered for grades 6, 8 and 10. For boys, the additive effect of childhood peer pressure rejections and aggression predicted more serious delinquency (Miller-Johnson, et al., 1999). Fordham and Ogbu (1998) suggest that Afiican American youth believe that working hard for good grades is “acting white.” They have defined “acting white” as having proper 24 speech, good grades, and attending educational extra-curricular activities. These findings suggest that there is something about the way that Afiican American males perceive academic achievement that may interfere with their desire to suCceed in school. This , study investigated whether negative peer behavior is related to achievement among African American males. Maryshow (1992) found that African American students rejected high achievers who achieved in Eurocentric mainstream cultural modes such as competition and individualism. Those high achievers who exhibited behaviors rooted in the African culture such as communalism (cooperation) and verve (preference for variability) were rated significantly more acceptable by their peers. These key findings suggest the need to investigate the effects of peers on the academic success or failure of African American male students. Extracurricular Activities Many studies have found that participation in extracurricular activities is linked to higher academic success (Clark, 1990; Eccles et al., 2003; Marsh & Kleitman, 2002; Scales, Benson, Leffert, & Blyth, 2000). Eccles (2003) used data from the Michigan Study of Adolescent Life Transitions (MSALT), which is a longitudinal study that followed 6th graders in 1983 in 10 school districts, to assess the effects of involvement in extracurricular activities. The data showed that the child’s involvement in extracurricular activities predicted the completion of more years of education and college graduation. This study also found that the total number of clubs and activities predicted greater attachment to school. Lancer (2002) suggests that an alternative route to success for African American boys is through athletics. This study showed that low-achieving urban 25 Black males perceive professional sports as their ticket to success. Unfortunately, this ideology has a low success rate and does not always promote academic excellence. How does the child’s participation in extracurricular activities influence the academic success or failure of African American students? Family Characteristics/Processes A number of family characteristics/processes may influence the academic achievement of Afiican American males. (Amos, 1991; Rodney, et al., 1999; Luster & McAdoo, 1996; Batey, 1999; Lingxin, 1997; Pruitt & Corbin, 1999; Morehouse Research Institute and Institute for American Values, 1999; McLanahan & Sandefur, 1993; McLanahan & Garfinkel, 1999). These factors include income level, family structure (single/two-parent home), father involvement, home environment, parent involvement in school and parent’s educational expectations. Father involvement is linked to the academic achievement of children. Compared to children with both parents at home, children who live apart from their fathers are five times as likely to be poor, do poorly in school, and are twice as likely to drop out of school (Lingxin & Brinton, 1997). Father absence leads to additional challenges for African American boys, such as poverty, lack of male role models, and poor self-identity development (Pruitt & Corbin, 1999). Four out of every 10 children in the United States will go to sleep in homes where their fathers do not live (Morehouse Research Institute and Institute for American Values, 1999). Before they reach the age of 18, more than half of American children are likely to spend at least a significant portion of their childhood living apart from fathers. In 1999, nearly 3 million children were living in father-absent homes in neighborhoods in which a 26 majority of families with children were headed by single mothers (Morehouse Research Institute and Institute for American Values, 1999). About 4.5 million US. children in 1990 resided in predominately fatherless neighborhoods, in ‘which more than half of all families with children were headed by single mothers. Eighty percent were Black. Of all the African American babies born in 1996, 70% were born to unmarried mothers. These trends pose significant threats to African American boys and to our nation (Morehouse Research Institute and Institute for American Values, 1999). There is compelling evidence that children raised by single parents generally do not fare as well as children raised by two married parents. Controlling for race, neighborhood characteristics and mothers’ education and cognitive ability, research stated that boys raised in single parent homes are twice as likely to commit a crime leading to incarceration (McLanahan & Sandefur, 1993; McLanahan & Garfinkel, 1999) Many African American boys are significantly challenged with poor achievement, crime involvement, and violence. Living in a two-parent home might provide more guidance and emotional security to Black boys as they face these challenges. The effect of father absence is a very important area that may provide additional insight on the development of African American boys living in single and two parent homes (McCabe, Clark & Barnett, 1999; Pruitt & Corbin, 1999; Rodney, et al., 1999). Assessment of this area was not limited to the presence or absence of a biological father. This study investigated the extent to which African American fathers were involved in their son’s lives. Batey (1999) examined factors that led to successful Afiican American boys and revealed that these boys had positive relationships with their families, friends, and other adults in their lives. They were socially competent and effective problem 27 solvers, who were able to negotiate through a web of , adversity at their school and in their neighborhood. Fathers or father figures appear to play a more significant role than mothers during the adolescent phase of African American boys (Pruitt & Corbin, 1999). The availability of fathers plays an especially important role in the development of positive self-esteem in African American boys. The lack of male role models for African American boys at home, as well as in the educational system could be a negative factor in promoting educational attainment (Rodney, et al., 1999). McCabe et a1. (1999) examined the relations among family protective factors, stressful events, and behavior adjustment of African American 6th graders. This study found that father involvement did not appear to be a major predictor of child adjustment; however, it predicted the acting out behaviors of students. Father involvement in the child’s life and firm limit setting at home may be ‘ particularly important for encouraging rule compliance and positive classroom behavior. These findings suggest the need for more research regarding the effects of single parenting, father absence, and poverty on the academic success or failure of African American males. Parental involvement in school and educational expectations of the African American boys are important areas of research. Maroon (1999) found that parental involvement had an important effect on several academic outcomes: GPA, math/science, verbal ability, work habits, and social development. Mandara (2006) found in a review of research articles on family functioning and male academic achievement that when African American parents were actively involved in their son’s academic efforts by monitoring homework and other academic pursuits and create a constant and positive dialogue with teachers and school officials, they increase the odds of their son’s 28 succeeding in school. Luster and McAdoo (1996) looked at the longitudinal effects of participation in the Perry Preschool project and found that Afiican American male students spent more time on homework if their parents had high educational expectations. Time spent on homework at age 15 was positively related to educational attainment. This finding reveals the importance of further investigating high parent expectations towards the educational achievement of African American boys. I In conclusion, it is clear to see that there are several factors that influence the academic success and failure of African American male students based on the numerous studies that have been discussed. It was quite challenging to find studies that focused primarily on African American males and achievement. Many of the studies discussed factors relevant to Afiican American children but not specifically to Afiican American males. It was even more challenging to find studies that focused primarily on African American male students and academic success. Most of the studies presented findings that examined the academic failure of African American male students. It is the hope of this study that further insight may be provided by solely focusing on Afiican American male students; and identifying several factors that may influence their academic success and failure. Research Hypotheses Based on the research objectives, the following hypotheses were tested. 1. What is the relationship between neighborhood safety/quality and the achievement levels of Afiican American boys? Ho 1: There is no difference in how the parents of high, medium, and low -achieving Afiican American males rate their neighborhood safety/quality level. 29 Ha 1: Parents of high-achieving African American males are likely to rate their neighborhoods as safer places to live than the parents of low-achieving males. 2. What is the relationship between the child’s perception of the school environment and the level of achievement? Ho 2: There is no difference in how high, medium, and low-achieving African American male students rate'their school safety/learning climate. Ha 2: High-achieving African American male students are likely to perceive a better school/safety climate than their low-achieving peers. 3. What is the relationship between the child’s religious involvement and the achievement levels of Afiican American boys? Ho 3: There is no difference in how high, medium, and low-achieving Afiican American males’ religious involvement. Ha 3: High-achieving Afiican American males have more religious involvement than their low-achieving peers. 4. What is the relationship between the child’s negative peer characteristics and the achievement levels of Afiican American boys? Ho 4: There is no difference in the levels of negative peer behaviors of high, medium, and low-achieving African American male students. Ha 4: Low-achieving African American male students have peers who exhibit greater levels of negative behaviors than high-achieving male students. I 5. What is the relationship between the child’s amount of extracurricular activities and the achievement levels of African American boys? 30 Ho 5: There is no difference in the involvement of extracurricular activities of high, medium, and low-achieving African American males. Ha 5: High-achieving African American male students are more involved in extracurricular activities than low-achieving African American male students. 6. What is the relationship between family characteristics (income, family structure) and the achievement levels of Afiican American boys? 6a.What is the relationship between household family income and the achievement levels of African American boys? Ho 6a: There is no difference in the family income levels of high, medium, and low- achieving African American males. Ha 6a: The families of high-achieving African American males have higher incomes than the families of medium and low—achieving African American males. 6b.What is the relationship between family structure and the achievement levels of African American boys? Ho 6b: There is no difference in the achievement levels of high, medium, and low- achieving Afiican American male students from single and two-parent homes. Ha 6b: Afiican American male students from two-parent homes have higher achievement levels than African American male students from single-parent homes. 7. What is the relationship between family processes (home environment, parent involvement in school, parent educational expectations of the child and father involvement) and the achievement levels of African American boys? 31 7a.What is the relationship between the level of a supportive home environment and the achievement levels of African American boys? Ho 7a: There is no difference in the home environments of high, medium, and low- achieving African American male students. Ha 7a: The home environments of high-achieving African American male students are of better quality than the homes environments of medium and low-achieving African American male students. 7b.What is the relationship between the parent’s involvement in the child’s school and the achievement levels of African American boys? Ho 7b: There is no difference in the level of parental involvement in school for high, medium, and low-achieving African American male students. Ha 7b: High—achieving African American male students have parents who demonstrate greater levels of parental involvement in school than the parents of their medium and low-achieving peers. 7c.What is the relationship between the parental educational expectations of the child and their achievement levels? Ho 70: There is no difference in the parental educational expectations of high, medium, and low-achieving Afiican American male students. Ha 7c: The parents of high—achieving Afiican American male students have higher educational expectations than the parents of medium and low-achieving African American male students. 7d.What is the relationship between the level of father involvement in, the child’s life and the achievement levels of Afiican American boys? 32‘ Ho 7d: There is no difference in father involvement for high, medium, and low- achieving African American male students. Ha 7d: The fathers of high-achieving African American male students have greater interaction than those of medium and low-achieving African American male students. 33 Chapter 3 Method The primary objective of this study was to examine factors that distinguish between low, medium and high achieving Afiican American 12-18 year old boys. In addition, this study identified the relationships between all of the factors, as well as, the relationships between the factors and the outcome variables of reading and math. The methods used to achieve these objectives are presented in this chapter. The chapter is divided into the following sections: (a) research questions, (b) assumptions, (c) sample, ((1), research design and procedures, (e) measures, (f) data analysis plan Research Questions 1. In what ways do the contexts of high, medium and low-achieving African American male students differ (e. g., neighborhood, school, peer group)? 2. Do high, meditun, and low-achieving African American male students differ in terms of their involvement in activities outside the home (extracurricular activities, church)? 3. Do high, medium and low-achieving Afiican American male students differ in terms of family characteristics and processes? Assumptions The next section presents the underlying assumptions of this study. 1. One must consider more than a single predictor variable when studying issues regarding children and education. 34 2. The academic dilemma of African American boys must be examined across multiple ecological systems in order to provide a more comprehensive view. Therefore, using Bronfenbrenner’s ecological model is appropriate. 3. Afiican American boys appear to undergo a distinctive educational experience which is reflected by their lower levels of academic achievement compared to African American girls and males from other ethnic groups. 4. High, medium and low-achieving African American boys have differing experiences. Sample Data were used from the Child Development Supplement-II (CDS-II) which is one of the various ongoing research components of the Panel Study of Income Dynamic (PSID), a longitudinal study of a representative sample of US. individuals and families. Since 1968, the PSID has collected data on family composition changes, housing and food expenditures, marriage and fertility histories, employment, income, time spent in housework, health, consumption, wealth, and more. In 1997, the PSID administered a supplement to its original data set with more information on PSID parents and their 0-12 year-old children using the Child Development Supplement I (CDS-I). The purpose of the study was to provide researchers with an extensive, nationally representative, and longitudinal data base of children and their families with the goal of studying the intricacies of early human capital development. PSID re-contacted families in CDS-I that remained active in the PSID panel as of 2001 in order to create the Child Development Supplement-II (CDS-II). 35 The Panel Study for Income Dynamics (PSID) re-interviewed 2,021 families who provided data on 2,907 children and adolescents aged 5-18 years old. This second wave of data is called the Child Development Supplement II (CDS-II). The CD3 collected data from parents /caregivers, teachers, and 2,907 children via face to face interviews, computer interviews, telephone interviews and written surveys. The children selected for this project are African American male respondents aged 12- 18 years old. There are approximately 228 cases available for this study. The Child Development Supplement 11 consists of information obtained from the primary caregiver, child and the teacher via face to face computer assisted interviews and telephone interviews. Also, self- administered mail surveys were used. Several measures were used in order to gather information regarding the child. Research Design and Procedure 2 The Panel Study of Income Dynamics (PSID) re-interviewed 2,021 families from the wave I of the Child Development Supplement (CDS-I) to form the CDS-II. CDS-II collected data from parents / caregivers, teachers, and children through various methods. Parents / caregivers were given face to face interviews and self-administered mail surveys. Teachers were given telephone and self-administered mail surveys. Finally, children ages 5-18 years old were given face to face assessments to measure their academic ability. Children 8-10 years old were given computer-assisted personal interviews (CAPI) to answer questions regarding their social / emotional well being, school environment, family environment, and neighborhood environment. Also, children 10 years old and over were given audio computer assisted self-interviews 36 (ACASI) to complete on their own. Respondents were paid each time they completed a survey or an interview. Children ages 12-18 were used in this study. Using children ages 12-18 requires some caution due to the fact that it is a very ‘wide age range to put into one group. The developmental stages for a 12 year old and an 18 year old are not the same. It is important to review all of the questions in each construct to ensure that the developmental expectations are appropriate for each age group. Research Variables and Measures The Child Interviews and the Parent Interviews consisted of several questions each with various Likert scale formats. Many of the variables were assessed by more than one question; however, it was a challenge to group them into subscales due to the different Likert scale formats for the answers. When individual measures are discussed below, I note how I combined items to create subscales with total scores to measure each construct. The HOME-SF and the Woodcock-Johnson Revised Test of Achievement (WJ-R) are instruments that are widely used and have been shown to be valid and reliable (Baker & Mott, 1999; Caldwell and Bradley, 1984) See appendix A for a list of items from each measure that were used to assess the various constructs. The following items were selected from the Child Interviews to represent the following constructs: race/ethnicity, religious involvement, negative peer characteristics, school climate, and extracurricular activities. The interviews were conducted with children and their parents/primary caregivers in three different formats. The face-to-face format was used with children ages 5-18 to measure academic ability. Computer-assisted personal interviews (CAPI) were given to children ages 8-10 and the audio-computer- 37 assisted self-interview (ACASI) was given to adolescents 10-18 years old to measure family environment and neighborhood environment. Answers were recorded in several different formats such as “yes/no” and “never” =1 to “every day”=6. Predictor Variables School climate. School climate and connectedness was assessed with 4 items completed during the Child Interview. The items are: “Do you feel a part of school?”, “Do you feel close to people at school?”, “Do you feel happy at school?” and “Do you feel safe at school?” Answers were recorded on a 6—point scale ranging fiom “never” to “every day”. The total score was computed by summing all of the items. Higher scores indicate more positive levels of school climate. Religious involvement. Child’s religious involvement was assessed with 4 items completed during the Child Interview. The items are: “How important is religion to you?”, “How much comfort do you get from religion?”, “How often did you attend religious services?” and “How often did you participate in religious clubs or activities?” Two items had responses scored on a 4-point scale ranging from “not at all important” to “very important.” The other two items had responses on a 6-point scale ranging from “not at all” to “more than once a week”. The total score was computed by summing all of the items. Higher scores indicate higher levels of religiosity Negative peer group characteristics. Negative Peer behavior was assessed with 6 items’completed during the Child Interview. Sample items include: “How many of your fiiends do the following: encourage you to disobey your parents, are in gangs, encourage you to do dangerous things, get in trouble in school, etc.?” Responses were scored on a 5- point scale ranging from “none” to “almost all or all.” A total score was computed by 38 summing all of the items. Some items were recoded so that higher scores indicate more negative friends. Extracurricular activities. Extracurricular activities were assessed with 5 items completed during the child interview. Sample items include: “Were you a member of any athletic or sports teams at school in the last 12 months?” and “Were you a member of any groups in the community such as scouts or hobby clubs in the past 12 months?” Responses were scored by answering “yes=1” or “no=0.” A total score was computed by summing all of the items. Responses were recoded so that higher scores indicated higher levels of involvement in extracurricular activities. Parent/Caregiver Interviews Data were obtained during the parent / caregiver interview regarding neighborhood safety/quality, family income, single/two-parent home, absent parent interaction, home environment, parent involvement in school, and parent’s educational expectations for child. The interviews with a parent or caregiver also inquired about several additional topics that were not included in this study, such as the targeted child’s health, school enrollment, childcare issues, family activities, finances, chores, gender roles of parents, psychological stress, family conflict, and parent work schedule. Parent’s educational expectations for child Parent’s educational expectations for the child were assessed with one item completed during the Parent / Caregiver Interview. The item is: “How much schooling do you expect your child to achieve?” Responses were scored on an 8-point scale ranging from “11th grade or less” to “MD, LAW, PHD, or other doctoral degree.” Higher scores indicate higher educational expectations. 39 Home environment. The level of supportive home environment was measured using the HOME-SR. The Home Observation for Measurement of the Environment- Short Form from the Caldwell and Bradley (1984) HOME Inventory was used to measure cognitive stimulation and emotional support that parents / caregivers offered to their children. The measure included interviewer/observations of the home and neighborhood. It was divided into four parts: infant/toddler, (birth to age 3) early childhood (3-6 years old), middle childhood (6-10 years old), and early adolescent (10-18 years old). The short form of HOME is comprised of 26 items. Each item is scored as 0 (suggesting the absence of quality stimulation) and 1 (indicating the presence of quality stimulation). F our scores for each age module were included: total standardized score, total raw score, emotional support subscale raw score, and cognitive stimulation subscale raw score. The total standardized score was used in the analysis. Bradley and Caldwell (1979) reported inter-rater reliabilities fi'om six studies in the high .803 to low .903, and 6 month test- retest subscale correlations ranging from .45 to .87. Neighborhood safety/ quality. Neighborhood safety/ quality was assessed by two variables that were combined to formulate one score. Neighborhood quality was assessed with one item completed during the Parent / Caregiver Interview. The item is: “How would you rate your neighborhood as a place to raise children?” Responses were scored on a 5-point scale ranging from “excellent” to “poor”. The response scale was recoded so that higher scores indicate more positive perceptions of the neighborhood as a place to raise children. Neighborhood safety was assessed with one item completed during the Parent / caregiver Interview. The item is: “How safe is it to walk around alone in your neighborhood after dark?” Responses were scored on a 4-point scale ranging 40 from “completely safe” to “extremely dangerous.” The response scale was recoded so that higher scores indicate a safer neighborhood. The correlation for these two items was .629. Parent involvement. Parent involvement at school was assessed by eight items completed during the Parent / Caregiver Interview. Sample items include: “In the last 12 months, how many times have you —Had a conference with child’s teacher?” “. . ...Attended a school event?” This is a continuous variable with scores ranging from 0— 997 (#tirnes). High scores reflect greater levels of parent involvement in school. Missing data code equals 998. Father involvement. Father involvement was assessed by four items completed by the child’s parent/ caregiver during the Parent/Caregiver Interview. Sample items include:” How often does the father take you to extracurricular activities? ---help with homework? Scores were reported based on 5-point scale ranging from “never” to “all the time.” Higher scores reflect higher levels of father involvement. These items were completed by all of the children regardless of their family structure. (N =22 8) Demographic Variables Family Structure. Family structure was assessed by items selected from the Parent/Caregiver Interview. Does child live with his/her biological mother? Father? Yes=1 No=5. A new variable was created for the analysis. Children that answered yes to both questions were placed in the two-parent household status, while those who answered no to one were placed in the one parent status. Child race. Child race was assessed by the child from the Child Interview. 41 (Child)What do you call your race/ethnic group? 1= Afiican American 2= White 3= Hispanic 4= Asian/ Pacific Islander 5: American Indian or Alaskan Native 6= Multi- Racial Child sex. Child sex was assessed by the Summary Variables. (Interviewer) Sex of individual: Male =1 Female =2 Child age (at assessment). Child age was assessed at the Child Interview. How old are you --------- ? The age of the child was recorded in months. Income. Income was assessed by items selected from the demographic portion of the Panel Study for Income Dynamics (PSID). Family income was obtained from the Panel Study for Income Dynamics (PSID) by the primary caregiver who indicated the level of income by selecting salary ranges. Outcome Variable Student ’3 achievement. Achievement was measured using the Woodcock- Johnson Revised Test of Achievement (WI-R). The Woodcock-Johnson Psycho- Educational Battery-Revised offers a normed set of tests for assessing cognitive abilities and academic achievement (four subtests). The CDS-II used three subtests as a measure of reading and math achievement: the Letter-Word (matching pictures with words), the Passage Comprehension (comprehension and vocabulary skills using multiple-choice and fill in the blank), and the Applied Problems tests (solving practical math problems). These subscales can be used individually or combined to create scores. The language and reading scores were combined into one score. The tests were standardized based on age (mean = 100) so that data from children who differed in age could be used in the same analysis. ‘ 42 Reliability Analyses of Variables Several measures were created to assess the variables. These measures were selected from the various interviews conducted by the child and the parent/primary caregiver. The new measures assessed the following variables: neighborhood safety/quality, child’s perceptions of school, child’s religious involvement, negative peer characteristics, parent involvement in school, and father involvement. It was necessary to estimate the reliability of these measures. The reliabilities were computed by using Cronbach’s alpha. Measures with alphas .50 or greater were included in the study. Table 3 shows the alpha coefficients for the various measures created from multiple items. The variable scales had an acceptable level of internal consistency. Table 3. Reliability Measures of Variables Using Cronbach’s Alpha Measure Cronbach’s Alpha Neighborhood Safety / Quality .633 Child’s Perception of School .634 Religious Involvement .910 Negative Peer Characteristics .732 Parent Involvement in School .754 _F_ather Involvement .922 Two additional variables were also created from items in the interviews; however reliabilities were not computed for them. The child’s involvement in extra curricular activities is an index based on the total number of activities the child participated in; a child received a score of one for each item scored yes. It was assessed as a categorical variable (yes/no). Parent’s educational expectations were measured by a single question. Data Analyses Data analyses were done using the Statistical Package for the Social Sciences (SPSS). The data analyses involved four major components: descriptive statistics, 43 correlations, multiple regression analysis, and comparison of means through one-way ANOVAs Descriptive statistics were computed first. The child, parent, and family characteristics were described in the analysis. Descriptive statistics were used to determine the distributional characteristics of each of the variables. Frequency tables which included means, standard deviations and range were used to inspect the data and to provide demographic information on the sample. Descriptive statistics were used to describe the child’s age and the number of children per household. Parent characteristics, such as age, education and sex of head of household, were also assessed. Finally, family characteristics such as marital status and household income were measured. Math and reading scores were analyzed separately for this study. High, medium and low achieving Black boys were compared in this study. The subjects were divided into three groups (1/3) based on their performances on the Woodcock-Johnson Revised Test of Achievement (highest third, middle, lowest third). The initial plan was to use student grades; however, due to missing data from teachers, this variable could not be measured. Correlations among variables were computed. Correlations were calculated to determine the extent of associations among the predictor variables and the associations between the predictor variables and achievement (reading / math). Separate analyses were done for reading and math. One-way ANOVAs were used to compare the mean test scores of the high, medium and low achievers for all of the factors that are continuous variables: neighborhood context (safety / quality), school context (perceptions of school climate), church 44 (religious involvement), peers (negative peer behaviors), extracurricular activities (child’s involvement in extracurricular activities), and family characteristics / processes (income, home environment, absent parent interaction, parent involvement in school, parent’s educational expectations). For categorical variables, such as family structure, cross-tabulations were conducted and a chi-square statistic was calculated to test for associations between variables. Multiple regression analysis was used to determine which factors were uniquely predictive of group membership (high, medium, low achievement) when all predictor variables were included in the model. A second set of multiple regression analysis was conducted on the predictor variables shown to be significant in the ANOVAs. 45 Chapter 4 Results In this chapter, the results of the data analysis are reported. The chapter is divided into three areas. First, demographic characteristics of the sample, as well as, the descriptive data of all the demographic and predictor variables are reported. Second, the results of the data analysis are reported, including bivariate correlations, multiple regression analyses, and one-way AN OVAs. Finally, the results are reviewed with regard to the hypotheses of the study. Descriptive Statistics The first set of analysis included descriptive statistics of the African American boys and their parents/primary caregivers. The goal was to obtain data on several instruments that contained responses from both the child and the parent/primary caregiver. The data were divided into three main categories for the analysis of demographic characteristics: children, parent/primary caregiver, and family. The sample size for this study is 228 African American boys. The average age of the boys was 15 years, and the range in age was from 12 to 17 years. The average number of children born to the families was approximately two. Several aspects of the parent/primary caregiver were assessed. They included age, level of education, and sex of the head of household. Only biological and step parents were used in the analysis. Grand parents and foster parents as primary caregivers were not included. The mean age of the mothers/female primary caregiver was 38.61 years, with an age range of 29-58. The mean age of fathers/male primary caregivers was 40.23 46 years, with an age range of 31-61. The mean level of education for mothers/female primary caregiver was 13 years, with a range of 5-17 years. The mean level of education for fathers/male primary caregiver was also 13, with a range of 8-17 years. Approximately 48% of households were headed by females, while 52% were headed by males. Family characteristics were the last area of demographic variables measured. Marital status of the parent/primary caregiver and total household income were reported. Forty-five percent of the parent/primary caregivers were married, while the others were never married (24%), widowed (4%), divorced (18%), or separated (9%). Outliers were removed in order to conduct a more accurate analysis of total household income. Only income with ranges between $10,000- $150,000 was used in this study. Outliers were eliminated in order to obtain more accurate results. The mean total household income was $60,000 with a range from $10,615-S144,160. The goal was to include as many cases in the analyses as possible, while taking into consideration that the data were obtained from two different sources. Children and parent/primary caregiver were the sources that were utilized. Table 5 is a descriptive summary of the demographic characteristics of the sample. Table 6 is descriptive summary of all the predictor variables. 47 Table 4: Sample Demographic Characteristics Variable N=228 Child Characteristics Age (Wm) M 15.0 SD 1.8 Range (min-max) (12.1-17.4) # Children 1.95 Parental Characteristics Mother Father Age (yearS) M 38.61 40.23 SD 3.89 4.34 Range (min-max) (35-58) (37-61) Education # Years completed M 12.53 12.58 SD 20.47 21.14 Range (min-max) (5-17) (8-17) Sex of Head (%) 47.8 52.2 Family Characteristics Marital Status (%) Married 44.7 Never married 24.3 Widow 4.0 Divorced 18.1 Separated 8.8 Total Household Income M $60,000 SD ~ $31,048 Range $10,000 - $144,160 48 -..-#1 4.5-_— ~, . , "7'1. . ... Table 5. Descriptive Summary of Predictor Variables Variable N Mean SD Neighborhood Safety/Quality 224 6.32 1.70 Child Perception of School Climate 228 17.53 4.91 Child Religious Involvement 228 10.24 6.44 Negative Peer Characteristics 228 10.10 4.37 Child Extracurricular Activities 226 3.18 1.40 Child Home Environment (HOME-SF) 228 1.02 1.00 Parent School Involvement 228 18.14 69.87 Parent Educational Expectations of Child 228 4.50 5.00 Father Involvement 228 3.83 5.19 Descriptive analysis on all of the variables revealed missing data in some cases. VMany of the original variables had to be discarded due to significant missing data. Data from teachers would have been valuable; however, this area contained the most missing data. For example, one of the questions asked of the teachers was: In general, how adequate are the amounts of the supplies and materials listed for your classes? The number of missing classes for this variable was 227. Correlations Among Variables Pearson correlation matrices were computed to determine the relationship among the predictor variables for achievement. Two separate matrices were computed to describe the associations among the predictor variables (Table 6), and the associations between the predictor variables and the achievement variables of reading and math (Table 7). Polit and Beck (2004) indicate that strength of the correlation is dependent on the variables 49 W? 2: if.- "hr-.1 .. I being assessed: a correlation greater than 0.5 is considered high; a correlation of 0.3 to 0.5 is considered moderate; and a correlation of 0.1 to 0.3 is low. Correlations Among Predictor Variables An examination of the correlations among the predictor variables revealed several significant relationships. The following correlations are presented based upon the six main contexts of the study: neighborhood, school, church, peers, extracurricular activities, and family characteristics/processes. Neighborhood context. Although there were two significant correlations between the neighborhood safety/quality variable and other predictor variables, they were both low. As expected, there was a significant correlation between neighborhood safety/quality and the home environment; boys from safer neighborhoods also tended to have more supportive environments. Also, there was a significant correlation between neighborhood safety/quality and parent’s educational expectation level of the child; parents who lived in better neighborhoods tended to have higher educational expectations for their boys. 50 [hula ._|..If§_ I... flew? a. NGBOBQ 0030.308 >325 Em 33.08.. 553.8 >oEo033. m. GEE... ox§o§.oc.m1=