STUDENT MOBILITY AND STUDENT ACHIEVEMENT TONY THERON WILCHER. (Under the direction of Dr. C. Thomas Holmes)

Transcription

1 STUDENT MOBILITY AND STUDENT ACHIEVEMENT by TONY THERON WILCHER (Under the direction of Dr. C. Thomas Holmes) ABSTRACT This study examined the relationship between student mobility and student achievement at a middle school in a metropolitan school district in Georgia. Student mobility has been linked to low student achievement. Many educators believe that student mobility is the inevitable result of students moving between residences. Student mobility not only impacts students who change schools, but also classrooms that have to adjust to a changing student enrollment. The researcher defined student mobility as the number of times each student moved to a different school, not counting the natural transitions between elementary, middle, and high school. Students with no moves were considered non-mobile, 1 to 2 moves moderately mobile, 3 or more moves were identified as highly mobile. Student achievement was defined as the student s performance on the ITBS in reading and math. Data were obtained on eighth grade students in the areas of reading and mathematics. The eighth grade students were given the Iowa Test of Basic Skills, a norm referenced measure of student achievement. A three way ANOVA was conducted on reading and mathematical scores by the factors of gender, race, and mobility. There were three research questions. (1) Are there statistically significant differences in scores on the Iowa Test of Basic Skills (ITBS) among levels of race

2 and levels of mobility. (2) Are there statistically significant differences among levels of mobility and gender? (3) Are there statistically significant differences among levels of race and gender? The data revealed that there were no statistically significant interactions among any levels of the three independent variables (race, gender, and mobility). However, there were statistically significant main effects for race, gender, and mobility. White students achieved higher scores than minorities in both reading and mathematics. Non-mobile students obtained higher scores in reading and mathematics as opposed to moderately mobile and highly mobile students. There were significant differences in the area of reading, as females obtained higher scores than males.

3 STUDENT MOBILITY AND STUDENT ACHIEVEMENT by TONY THERON WILCHER B.S., Georgia College & State University, 1992 M.Ed., The University of Georgia, 1993 Ed.S., The University of Georgia, 1994 A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree DOCTOR OF EDUCATION ATHENS, GEORGIA 2005

4 2005 Tony Theron Wilcher All Rights Reserved

5 STUDENT MOBILITY AND STUDENT ACHIEVEMENT by TONY THERON WILCHER Major Professor: C. Thomas Holmes Committee: Sally J. Zepeda C. Kenneth Tanner Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia December 2005

6 DEDICATION This dissertation is dedicated to my faithful companion, Yolanda, and my parents, Callie and Earnestbell Wilcher. Thank you so much for your unconditional love and support throughout the years. Yolanda, you have been a constant source of inspiration by just being willing to understand the many days of study throughout my graduate school years. Thanks also goes to my Heavenly Father who provided the wisdom, health, and guidance, both before and after I gave my life to him on January 3, That day is a day that will forever burn in my heart. Without you, I am nothing, but with you, I can do all things. Glory be to God, my Lord and Savior Jesus Christ. iv

7 ACKNOWLEDGMENTS This journey has been a long but sweet odyssey, and I would like to thank many people who assisted me along the way. First my wife (Yolanda), children (Brooke and Christina), as well as countless educators, among them are Connie and Christine, and my immediate and extended family. Special thanks also are reserved for my advisor and chair of my committee, Dr. C. Thomas Holmes and the other professors on my committee, Dr. Sally J. Zepeda and Dr. C. Kenneth Tanner. Thank you all for your patience and guidance. v

8 TABLE OF CONTENTS Page CHAPTER 1 INTRODUCTION...1 Purpose...5 Justification for the Study...6 Research Hypothesis...7 Research Questions...8 Definition of Terms...8 Limitations of the Study...8 Overview of the Study...9 Overview of Research Procedures REVIEW OF RELATED LITERATURE...10 Mobility Statistics...10 Reasons for Family Mobility...14 Mobility and Achievement...19 Early Research on Student Mobility and Achievement...19 Contemporary Research on Student Mobility and Achievement...23 Military Mobility and Student Achievement...29 Race and Gender on Student Mobility...33 vi

9 Possible Gender Differences in Achievement Confounding Variable...34 Socioeconomic Status and Mobility METHODOLOGY...39 Research Context...39 Selection of Research Participants...41 Data Collection Procedures...41 Data Collection Instrument...43 Data Analysis FINDINGS AND ANALYSIS OF DATA...47 Characteristics of the Sample...47 Description of the Treatment...48 Analysis of the Data SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS...58 Summary...58 Discussion...61 Implications for Policy...62 Implications for Practice...63 Factors to Consider for Future Research...66 Implications for Future Research in Student Mobility...77 REFERENCES...79 vii

10 CHAPTER 1 INTRODUCTION The task of educating students that move within and between school systems seems to be more challenging given the ever expanding metropolitan Atlanta area. Student mobility is easily accomplished given the close geographic distances between metro counties and a vast interstate system. The broad metropolitan Atlanta area consists of 13 counties, and with families on the move, family mobility translates into increased student mobility. High student mobility is not new to urban metropolitan areas (Straits, 1987). However, there is increasing accountability of educators to raise student achievement with both mobile and non-mobile students. In some instances high stakes tests are the instruments of choice used to measure student achievement. Although not everyone agrees with the use of high stakes testing (Perrone, 1991; Stalker, 2001; Wenning, 2003), high stakes tests and the significance placed upon them has the potential to change the manner in which mobile students are educated. Federal, state, and local education agencies have developed standardized tests to measure student achievement. For example, the United States Department of Education uses students test performance from the National Assessment of Education Progress (NAEP), which tests a sample of students in selected grades in schools throughout the United States. Georgia uses the Criterion Reference Competency Test (CRCT) and Iowa Test of Basic Skills (ITBS) as a measure of student achievement, while some school districts develop and use local measures of student achievement. 1

11 These tests are designed to measure students success on learning specific academic standards. The results of students performance on such tests have become perhaps the most important factor for holding local schools accountable in meeting state and federal standards (National Center for Educational Accountability, 2002). In fact, many states have tied accreditation or ranking of schools to standardized test results. In Georgia, schools can be labeled as making adequate yearly progress or needs improvement based on the performance of students on standardized test. Schools that do not meet specified benchmarks over a defined period of time may endure a wide range of sanctions aimed at improving poor student achievement. In some cases the sanctions are significant, such as changes in curriculum or school reconstitution. In Georgia, the first accountable school year report card for grades K-8 was issued during Fall 2004, and financial awards were distributed between July 1, 2004 and June 30, The emphasis on performing well on the CRCT and other high stakes test seems to involve several layers of educators, ranging from the central office, to the local building administration, and teachers. Increased expectations are also coming from the federal, No Child Left Behind Act (Dougherty, 2002), which links federal government funding of education with state accountability standards. The No Child Left Behind Act (NCLB) of 2001 was signed into law on January 8 th, 2002, by President George W. Bush. The act is a continued reauthorization of the original Elementary and Secondary Education Act (ESEA) of 1965, which was a part of the late President Lyndon B. Johnson s war on poverty. Title I was also a part of the war on poverty that included educational services to students from low socioeconomic backgrounds. The aim of the original ESEA, the forerunner to No Child Left Behind, was to level the educational landscape for 2

12 students in poverty in an era when human rights were present in many legislative and executive acts (e.g., Civil Rights Act of 1964, Title IX, etc.). There are several major provisions of the NCLB, some of them are: Schools and school systems must test students in grades 3-8. School must show improvement each year. This is called adequate yearly Progress. If a school does not meet improvement goals after two years, it is designated as a Needs Improvement school. Low-income parents of students in Needs Improvement schools also can request supplemental educational services, with transportation provided, and paid for by the school district. Parents of students in Needs Improvement schools also can send their child to another school with transportation provided by the school district. Schools must hire highly qualified staff and must tell parents about the qualifications of the staff. All students must be performing on grade level within 12 years of the signed act. Ninety-five percent of students in a school must be tested, as well as 95% of students in certain subpopulations, which includes race, gender, economically disadvantaged, students with disabilities, and Limited English Proficient students. (Wenning, 2003) School accountability policies such as NCLB apply to all students; however, initial policies were focused on Title I schools that did not meet federal standards for improvement over an established amount of time. Local school districts are highlighting the importance of student achievement. The focus must be thorough given that if accountability standards are not met then 3

13 changes in the school operations are mandated and school funds could be drawn. Although the estimated amount of federal funds the government provides to states is around 7%, that percentage represents nearly $12 billion, an amount most states need to supplement soaring state educational costs. In such an educational environment that emphasizes accountability for student achievement, all factors that may impede student achievement should be considered, in particular student mobility and its relationship to student achievement. A report by the Educational Testing Service (Barton, 2003) indicated that student mobility is one of 14 factors that can have a significant influence on student achievement. The factors reported included low birth weight, nutrition, lead poisoning, rigor of school curriculum, teacher preparation, teacher experience and attendance, class size, availability of classroom technology, school safety, reading to young children, TV watching, parent availability and support, parent participation, and student mobility. Both the NCLB and Georgia s A+ Education Act vaguely mention student mobility and its potential impact on student achievement. NCLB has a provision that allows local school districts to discount the test results of students that are mobile. NCLB allows the provision for students who do not remain in one school for the same academic year because the student may change schools. However, where the student changes school is important. If a student moves to another school within the same school district, then the school district counts the student towards the school district s accountability standards for NCLB. If the student moves out of the school district then the state counts the student in the state's accountability standards pertaining to NCLB. Of note, each state is allowed to create a definition of full academic year. In some states, full academic year could mean from the first day of school up to the first day before administration of the state s assessment of student achievement. In the case of the 4

14 state of Georgia, students must be enrolled on the first official student count day (known as FTE), which is typically held in October and March. If students arrive after that FTE day then they will not have to be counted by the local school in the student achievement totals regarding meeting adequate yearly progress. With such a broad scope for exclusion, it is possible that the mobile student that moves several times in a year may become lost in the system in terms of who is accountable for his or her academic achievement. One must remember that the student achievement of students who are highly mobile are judged with the same standards as students who do not move throughout their educational program. Georgia s Office of Student Achievement (OSA) evaluates student achievement in the areas of race, gender, economically disadvantaged (i.e., paid lunch students and free or reduced lunch students), LEP, and students with disabilities. However, regarding the area of student mobility, the OSA has not addressed the potential negative interaction of student mobility and student achievement. There is evidence that student mobility is negatively associated with performance on standardized tests (Demie, 2002; Kerbow, 1996; Mao, Mellor, & Whitsett, 1998;). Purpose The purpose of this study was to determine whether student mobility had a negative relationship on student achievement in reading and mathematics for eighth grade students. Student achievement was measured using the Iowa Test of Basic Skills (ITBS). Each student included in the study s sample was identified as being highly mobile, moderately mobile, or nonmobile (stable). The students were grouped according to their race (Black, Hispanic, and White), and gender (male, female). For example, there was a group of students who were in a category 5

15 known as highly mobile, Black, and male; a group that was moderately mobile, White, and female. Justification for the Study The increased use of standardized testing as the primary means to evaluate student achievement has redefined the priorities placed on schools. Both federal and state departments and boards of education are using students performance on high stakes standardized tests as a means of determining academic achievement (American Educational Research Association, 2004; United States Department of Education Office of Civil Rights, 2000). In addition, financial rewards and in rare cases, job security, are tied to student performance on high stakes tests (Loschert, 2000). The influence that student mobility may have on the success of schools in promoting academic achievement has not been factored into the equation. The critical question is whether student mobility negatively influences student achievement? If student mobility hinders student achievement, then school achievement may be affected in like manner. If local schools do not meet accountability standards then significant consequences may follow. With so much at risk, a thorough investigation of student mobility and student achievement should be conducted because it may provide answers for schools who struggle to educate mobile students. Georgia s newly formed Office of Student Achievement is the state accountability agency holding schools accountable for increasing student achievement in several demographic areas, such as gender (male/female), race (Black/White/Hispanic), socioeconomic status (paid lunch students and free or reduced lunch students), and other areas. Student mobility and its relationship with student achievement should be investigated, as the study could yield valuable information to assist educators when working with certain groups of students. 6

16 To illustrate, suppose the research mentioned in this paper indicated that student mobility negatively impacted student achievement for Hispanic males more than any other group. The information could be useful to English for Speakers of Other Languages (ESOL) teachers and principals to help promote student achievement with this particular group of students. In addition, if Hispanics make up a sizeable portion of your student population, for example 20% or more, it may be beneficial to understand the relationship between student mobility and student achievement for Hispanic students. The hope of generating solutions to improve student achievement for mobile Hispanic students could become a planning goal in a school s strategic plan. General and vague assumptions should be avoided in an environment where specific accountability standards have been issued to local schools by both federal and state education agencies. Research Hypothesis Null Hypothesis 1: There are no statistically significant main effects on (a) mobility, (b) race, and (c) gender. Null Hypothesis 2: There are no statistically significant two way interactions involving (a) mobility x gender, (b) race x gender, and (c) race x mobility. Null Hypothesis 3: There is no statistically significant three way interaction involving mobility x race x gender. 7

17 Research Questions 1. Are there statistically significant differences in scores on the Iowa Test of Basic Skills (ITBS) among levels of race (White, Black, or Hispanic) and levels of mobility (high, moderate, non-mobile)? 2. Are there statistically significant differences among levels of mobility and gender? 3. Are there statistically significant differences among levels of race and gender? Definition of Terms 1) Reading achievement the percentile score for the total reading score. 2) Mathematical achievement the percentile score of the total mathematical score. 3) Mobility Student movement from one school to another, within or outside of the school system. Students moving three or more times during their educational program will be defined as highly mobile, one or two times moderately mobile, no moves as non-mobile. Limitations of the Study The following limitations of this study should be noted: 1. Only one grade was used in this study, therefore similar findings may not be found with different grade levels. 2. The study did not randomize participants in this study; therefore the results may not be generalizable to other schools or students. 3. The study was limited to two schools and one school system in Georgia, which limits the ability to generalize the findings. 4. The lone measure of student achievement used in this study was the Iowa Test of Basic Skills. 8

18 Overview of the Study This investigation was organized into five chapters. Chapter 1 introduced the problem of the study, presented the research hypothesis and questions, defined terms, and listed the study limitations. Chapter 2 contained a review of the literature on student mobility and achievement. Chapter 3 explained the methodology used in this study, including a description of the research context, selection of participants, data collection procedures, data collection instrument, and description of statistics used for analysis. Chapter 4 presented the findings of the research and data analysis. Chapter 5 provided the summary, conclusions, and recommendations for future research. Overview of Research Procedures The participants of this study were eighth grade students attending two public middle schools during the school year Each student s achievement in reading and mathematics was measured by viewing their ITBS scores collected from permanent records. Permission for conducting this study was granted by the local school system s Director of Assessment and Evaluation. The students were not chosen at random, but the sample included all eighth grade students who had ITBS test scores available. In addition to the collection of ITBS results, information regarding the race, gender, and number of schools attended prior to was collected. The number of schools attended provided data indicating how mobile a student had been prior to the year. This quantitative study used a 3 (mobility - stable, moderate, high) X 2 (gender) X 3 (race - White, Black, Hispanic) research design. The independent variables were mobility level - measured by number of schools attended, race, and gender. The dependent variables were reading and mathematical achievement on the ITBS. 9

19 CHAPTER 2 REVIEW OF RELATED LITERATURE The purpose of this literature review was to provide a comprehensive review of the subject of student mobility and how it influences student achievement. A particular emphasis was placed upon studies related to the aforementioned research questions. Cook et al. (1992) suggests that useful literature reviews clarify the topic and include studies that are not biased in one direction or another. Useful literature reviews feature a wide range of persons, settings, and times; a heterogeneous representation of both the treatment and effect; and a broad range of potential explanatory variables. Mobility Statistics Mobility is a way of life for many Americans. The U.S. Census Bureau reported in 2003 that over 40.1 million Americans, approximately 14% of the population, changed residences. The majority of Americans moved within the same county, approximately 24 million, 8 million moved between counties, and around 8 million moved to another state (U.S. Census Bureau, 2003). The student mobility rates varied by the characteristics of the movers. Homeownership was one of the strongest predictors of whether a person moved. At least one-third of people occupying renter housing units moved during 2003, in contrast, only 1 in 14 people living in owner occupied units moved during People with income below the poverty level were more likely to have moved than those above the poverty level. Education accounted for only small differences in the moving rates, ranging from 11% of high school graduates moving to 13% of those with a bachelor s degree. Hispanics were more likely to have moved than Asians, 10

20 Blacks, or non-hispanic Whites. The addition of education did little to change that order, however, economic variables (employment, poverty status, etc.) appeared to have an effect. This was particularly true of Blacks who became just 14% more likely to move than non- Hispanic Whites, when Blacks had the same age, education, and economic characteristics as non- Hispanic Whites. Young adults had the highest moving rates. In 2003, about one-third of year olds moved in the previous year, more than twice the moving rate of all other age groups, however, people 55 and older were more likely to have moved to a different state than younger age groups. Life events such as start of a family and pursuit of employment may be a reason for younger adult movement, whereas retirement may be the impetus for older Americans movement (U.S. Census Bureau, 2003). A 1994 United States General Accounting Office (GAO) study on student mobility investigated factors involving residential movement and the corresponding relationship to student transience. The study was entitled Many Change Schools Frequently, Harming Their Education. The GAO investigation was conducted primarily in response to questions asked by Marcy Kaptur, an Ohio congressmen, who wanted to study the issue of why students move frequently from school to school. According to the GAO report, local school districts generally provide little additional help to assist mobile children and the Department of Education could play a role in helping mobile children receive appropriate educational services in a timely manner (p.2). The GAO study involved 15,000 third graders across the nation from 235 elementary schools. Parents, teachers, and school administrators completed surveys and questionnaires. The samples of students in the study were stratified by census region and three levels of mobility. 11

21 The GAO Report included the following findings: (1) Seventeen percent of all third graders have attended three or more schools since the start of first grade; 24% of third graders have attended two schools, while 59% remained at the same school since first grade. (2) Third graders below grade level in reading and mathematics were less likely to receive related Title I services if they changed schools frequently. Students who changed schools frequently were more likely to be below grade level in reading and mathematics than children who did not change as frequently. (3) Third graders who changed schools frequently were more likely than those who never changed schools to be below grade level in reading and mathematics, regardless of income. (4) Third graders who changed schools frequently were more likely than those who never changed schools to have repeated a grade, regardless of income. (5) Overall, children from low-income families were more likely to be low achievers than those from higher income families, regardless of frequency of school changes. (6) Third graders who changed school frequently were more than twice as likely to repeat a grade as those who never changed schools. (7) Children who changed schools within the district were slightly more likely to be below grade level in reading and mathematics than those who changed schools across districts. (pp.25-34) The GAO report produced several mobility statistics that seemed to establish a negative relationship between mobility and school success. Similar to the GAO findings, Long (1992) also found a relationship between family status and student mobility. He found that living 12

22 with both biological parents, independent of other variables, consistently lowered a child s lifetime number of moves and raised the duration of residence in the current residence. This was true for all families regardless of race. Family incomes under $15,000 were consistently associated with higher lifetime moves and lower duration in the current residence. Long s (1992) findings were not surprising given the challenges of finding affordable housing for families with incomes under $15,000. In contrast to Long s (1992) focus on family dynamics and mobility, Kerbow (1996) researched student performance in urban schools. In his research with Chicago public schools, Kerbow discovered that only 40% of Chicago students remained in the same school from 1 st through 6 th grade, not counting the logical transition between elementary and middle school. Kerbow further stated that some of the transient students matriculated in schools where only 50% of the students were still enrolled in the school during any given three-year span. According to Kerbow, 4 th grade students who switched schools were on average four months behind students who stayed in the same school. Students who moved four or more times were, by 6 th grade, a full year behind. Student mobility in the Chicago school system was overwhelmingly an internal process, where 82% of the transfers were moving from one city school to another. In some cases, Kerbow noted that the moves were in the same cluster of schools, sometimes within or near the same neighborhood. The schools shared similar racial, academic achievement, and income characteristics, so essentially, low achieving students would move from one low achieving school to another. The lack of diversity of schools may have been due to the polarized housing patterns that sometimes characterize large urban areas. Housing clusters may include residents with similar income, race, and education levels. 13

23 Based on the mobility statistics from the large GAO report, U.S. Census Bureau, and Kerbow s (1996) study of a large urban school system, the message seems clear that a number of U.S. children may attend several schools early in their educational experience. In addition, the consistent movement seems to be correlated with less than spectacular academic achievement. An important issue that seemed to emerge from the mobility statistics is a need to look at why students move. By understanding the underpinnings of student mobility, educational leaders may be in a better position to render assistance to students from families that are highly mobile. Reasons for Family Mobility The Minneapolis, Minnesota Public Schools conducted an investigation into why families move. The study became known as the Kids Mobility Project (1998) and it was a two part study. One part involved a quantitative emphasis that included a look at factors that may affect student performance at the elementary level. Factors such as socioeconomic levels, ethnicity, out-ofstate-births, family structure and attendance were researched. All were reported to have a strong relationship to reading achievement test scores and residential moves. Researchers looked at students who took the California Achievement Tests (CAT) in grades 1-6 during the year The sample size consisted of 6,098 students, roughly 25% of all 1 st -6 th graders at that time. The sample was consistent with the overall proportion of races in the Minneapolis Public Schools, which was 40.4% African American, 35.4% Caucasian, 8.3% American Indian, 11.8% Asian American, and 4.1% Hispanic. Mobility was researched from November 1994 until June 1995 (Kids Mobility Project, 1998). Quantitative Results Kids Mobility Project The analysis of the student performances on the CAT indicated that the more moves students made, the lower the average reading score. Students who moved three or more times 14

24 had an average reading score 20 points lower than students who did not move. The CAT reported scores as Normal Curve Equivalents, which had a mean of 50 and a standard deviation of 21, therefore, the difference of nearly 21 points was just short of one full standard deviation away from the mean. A finding of this magnitude is considered a significant difference. However, the question of whether student mobility alone was responsible for low academic achievement was not answered by the data recorded by researchers for the Kids Mobility Project. The reason being is that other factors alluded to in the study also were correlated to low academic performance on the CAT. For example, the following reading scores were reported for the other factors studied. Students eligible for free lunch had average scores 20 points lower than non free lunch students. Students living with both parents or with a parent and stepparent showed higher gains in scores over one year and had higher scores overall. Average reading scores for students born out-of-state were 10 points lower than Minnesota-born students and two points lower still for those born in other countries. Average scores of students of color were more than 20 points lower than scores for White students. Qualitative Results Kids Mobility Project The second part of the Kids Mobility Project used a qualitative approach, which among other things, involved interviewing 100 families newly enrolled in the Minneapolis Public Schools. Parents answered questionnaires and agreed to face-to-face interviews. The researchers were looking to answer the following questions: Do families recognize that frequent moves affect their children s school performance? Do families have a choice? To answer these 15

25 questions, researchers obtained data from the interviews and developed the following four categories: Coping Moves, Forced Moves, Upward Moves, and Lifestyle Moves. The data revealed that 59% of families were moving to cope with substandard conditions or not being able to find affordable housing. Three themes were highlighted in the coping moves category, problems with landlords, bad relationships, and substance abuse. In contrast, at least 21% of the interviewees indicated a need to move because of eviction, property condemnation, dangerous situations, and their own behavior. One parent stated, for me, moving was always tied with drugs and not having money to pay rent, then going to a shelter. Now that I m clean I m going to a shelter. Now that I m clean, I m going to stay put, give my kids a place to stay. I hope we can make it. Less than a quarter of the interviewees indicated that upward moves and lifestyle moves were responsible for mobility. Nine percent of the families viewed moving as a manner of living and lifestyle. The interviewees saw moving as a way to get away from the pressure of life and to generate a sense of change that could help with life s frustrations. One mother stated, I like to move, I m always on the go, seeing new people and places. When problems surface, and they always do, I ll just move again (p.11). Upward moves accounted for 11% of the interviewees. This group viewed moving as a positive experience, a chance for a better life. Families described their experience as trying to learn from the past and move on to the future (p.11). Similar to the Kids Mobility Project (1998), research by Warren-Sohlberg and Jason (1992) further underscored the importance of investigating the reason behind student mobility. They found that students that transfer schools because their family had to move as the result of a loss of housing and other household considerations, such as a divorce and financial difficulties, 16

26 had poorer academic performance than children who were transferring schools because they were leaving a violent neighborhood to go to a new neighborhood. The performance measure was the Wide Range Achievement Test. The upward mobile students scored superior on all subtests. Long (1992) proposed two hypotheses for why American children are so mobile. The first was that families with children have a minute number of obstacles to moving. This seems plausible since apartments and rental residences are readily available. Mao, Whitsett, and Mellor (1998) reported that many families with meager incomes move from apartment to apartment regularly to take advantage of marketing offers such as first month free or move in specials. When families move, children in those families also move, thus perpetuating the student mobility challenge. Neighborhoods and communities in America have included mobility as a structured part of life cycle stages that involve child bearing and rearing. The desire to move to a neighborhood with safe streets and good schools is a goal for most families, no matter the income or race. This notion is supported by Schneider, Stahl, and Struyk (1985), who reported that the residential mobility rate of Americans rises with income, while those in other western countries, such as West Germany, do not. Higher incomes enable homeowners to move more often, as they trade up to larger more expensive homes. Other reasons for frequent moves include divorce, family tragedy or death, military obligations, or seasonal job opportunities. Seasonal jobs (e.g., fishing, timber, construction, tourism, etc.) require employees to move at times that may not be opportune for children, such as during the school year (Lash & Kirkpatrick, 1990). Migrant workers are often affected by midyear moves that coincide with the planting and reaping season for crops. Migrant families 17

27 move not according to the standard academic calendar but according to the seasonal calendar. The harvest season may be in the spring or summer and warm weather climates such as Florida and California attract migrants who work for many months of the year, but not the complete year. Families that move because of a high paying job promotion seem to be insulated from the affects of frequent moves. Stroh s (1990) investigation of children s adjustment after a corporate move by their parents indicated a positive relationship between pre-move and post-move adjustment. In other words, children who were well adjusted prior to a move continued to be well acclimated after a move. Stroh reasoned that for the children, environmental changes did not negatively alter their overall adjustment. Stroh indicated that parental support and adjustment, especially from the father, were found to be instrumental in the overall and post-move adjustment of children. A related finding was noted in a New York State Department of Education (1992) study. In New York City public schools, families moved for several reasons, primarily, better housing, upward mobility, seeking higher income, military duties, and migrant work. The study revealed that families from poor housing areas moved more frequently than those in average and good housing situations, while students living with one parent or with no natural parents moved from school to school more frequently than do children with two parents. In summary, students move for a variety of reasons. Many of the reasons stem from a breakdown in the family make-up. Moving in many respects is a way of life and a coping mechanism for life s trials and tribulations. Upward mobility has also been recognized as a factor in student mobility. Whether upward or downward mobility, family and student movement is a part of the American way of life. From an educational perspective, a concern might be how much 18

28 of an impediment does mobility have on the achievement of the students that move for various reasons? Mobility and Achievement The relationship between student mobility and achievement has been researched extensively during the past 40 years. Researchers have realized the importance of students remaining in a constant learning environment in order to learn the skills necessary to succeed in school. For contemporary researchers, increased accountability in America s public schools has added another incentive to the quest to investigate factors that influence student achievement. Student mobility has been identified as a potential factor that affects student achievement. The findings of some studies indicate that mobility is associated with lower student achievement. Biernat and Jax (2000) indicated that frequent relocations interrupts regular attendance, continuity of lesson content, and the fostering of relationships between student and teachers. In addition, high student mobility has a slowing effect on basic skills acquisition. Biernat and Jax hypothesized that high mobility may create a long-term risk of school failure and dropout. Highly mobile students are also more likely to be retained in a grade than children who do not change schools (Fowler-Finn, 2001). However, despite a number of studies indicating a negative relationship between student mobility and student achievement, not all studies concur with this premise, especially earlier studies in the 1960s and 1970s. In addition, some contemporary research findings by Adduci (1990), Paredes (1993), and Sanderson (2003) acknowledge that mobility alone may not cause lower student achievement. Early Research on Student Mobility and Achievement Numerous studies were done in the 1950s, 1960s, and 1970s that examined the effect of mobility on student achievement. The findings of many of these early studies indicated that 19

29 mobility had no effect on achievement (Black, 1975; Bollenbacher, 1962; Cramer & Dorsey, 1970; Gilchrist, 1968; Moore, 1966); however, some researchers concluded that mobility negatively impacted achievement (Corbett & Wesolowski, 1966; Frankel & Forlano, 1967; Frazier, 1970; Levine, 1966; Morris & Mariana, 1967), and a few indicated that student mobility had a positive impact (Snipes, 1966; Whalen & Fried, 1973; Wise, 1971). In Whalen and Fried s study, they concluded that mobility had a differential effect on students depending upon their I.Q. Their study was conducted in Livermore, CA during the school year. Whalen and Fried studied 874 students and indicated that students with high Intelligence Quotients (I.Q.) and high mobility actually improved in the area of achievement, while low IQ students with high mobility had lowered achievement. Whalen and Fried used the General Vocabulary Test as the dependent measure of analysis. The psychometric properties (i.e., validity and reliability) of the General Vocabulary Test are unknown, which cast some doubt about the credibility of the findings produced by Whalen and Fried s investigation. Snipes (1966) reported a positive relationship between student mobility and academic achievement for elementary students attending public school in Georgia. In his research at the University of Georgia, Snipes used a sample size of 464 sixth grade students who were given reading tests from the California Achievement Tests, Form W. The analysis of variance results indicated that students who were identified as mobile achieved higher scores in reading vocabulary and reading comprehension. Upon this finding, Snipes concluded that number of moves did not detract from reading achievement, but rather strengthened it. Evans (1966) supported Snipes (1966) findings with a study on the academic achievement of 97 fifth and sixth grade students. Unlike Snipes, Evans used numerical grades in the participants of Arithmetic, Reading, Science, and Social Studies. The numerical grades were 20

30 converted from letter grades. Evans reported that the mobile group significantly outscored the non-mobile group in achievement. Of note, there were no significant differences in the standard deviations between the mobile and non-mobile group according to data reported by Evans (1966). In addition, only small differences were indicated in the averaged correlation scores. However, Evans concluded that mobility did not have a negative effect upon mobile students. According to Evans, mobile student s higher scores indicated that they showed an ability to adjust and outscore the non-mobile group. Carpenter (1971) also investigated the impact of student mobility and student achievement. Her dissertation involved fourth grade students in Colorado. Like her predecessors, Carpenter looked at reading achievement as the dependent measure. She administered the Stanford Diagnostic Reading Inventory Reading Test, Level I, Form W. Raw mean scores for the mobile and non-mobile students were computed and comparisons were made using a t-test. The results indicated that there were no significant differences in reading achievement between the mobile and non-mobile students. Hefner (1994) questioned the findings of many of the early studies. He reported that many of the studies reported during the early years, especially 1950s and 1960s, dealt with financially well off, healthy, White families. Given the level of segregation in public schools during that time, Hefner seemed to have provided a plausible explanation for the research findings. A review of sample demographics in the early studies seemed to support Hefner s contention. Very few samples included minority participants, however, some that did include minority representation reached similar conclusions. Research by Black (1972) and Harris (1973) were two studies that involved minority participants. 21

31 Harris (1973) studied inner-city Blacks and Hispanic students in Detroit. He investigated both reading and mathematics, using 406 students in Title I schools. Harris found that there were no significant relationships between reading and arithmetic levels with respect to race, gender, parent s education level, number of moves, and teacher attitude. Harris was one of the first studies to look at other variables and their role in student mobility and achievement. Harris used Chi-square values as his primary tool of analysis. Black (1972) also studied minority students from a Title I school. Black s studies preceded Harris study and indicated that there were no significant relationship between mobility and pupil achievement. He studied 210 students with reading achievement being the dependent variable. Black used the analysis of covariance to adjust the mean achievement scores from the student s performance on the California Comprehensive Reading Test. The analysis of variance was used to analyze the adjusted mean scores from the reading achievement tests. Black found that there were no significant differences at the.05 level of significance between the mobile and non-mobile students. Contemporary researchers (Adduci, 1990; Fernandez, 1987; Hefner 1994; Ingersoll, 1988; & Mehana, 1995) disputed the positive relationships purported in the literature. For example, Lacey and Blane (1979) cited a study done by Bourke and Naylor in which they reviewed 28 mobility studies. Bourke and Naylor indicated that they found mixed results. Eleven studies showed student mobility had no effect on achievement, 12 showed mobility lowered student achievement, and 5 indicated higher achievement. There appeared to be no major consistency in the early research data on student mobility and achievement. 22

32 Contemporary Research on Student Mobility and Achievement A review of several early studies indicated that there was little consensus on the relationship of student mobility and achievement. However, there also appear to be studies conducted in the 1980s, 1990s, and into the decade of 2000 that were equally contradictory. Although few contemporary studies find a positive relationship, some did not support a negative effect. Zamudio (2004) researched the relationship between student mobility and academic achievement. The study took place in Arizona and involved a sample of 487 third grade students and 490 fifth grade students. Zamudio found a statistically significant negative relationship between mobility and academic achievement. Of interest, the negative finding was most profound for students considered to be upper middle class or high SES. A statistically significant effect on both reading and mathematics achievement was discovered for high SES students in both third and fifth grade. Zamudio (2004) further analyzed the data by performing multiple regression tests. The results indicated that mobility was not significantly related to the achievement of students in the low SES group. However, low SES students did have a mobility rate higher than the high SES student. Zamudio noted that lower SES students tended to be Hispanic and were taught by teachers with less experience and education. One plausible explanation for Zamudio s findings may be that the initial achievement level of the lower SES students may have been lower at the outset than the higher SES students; thus the potential to observe a drop in academic achievement was most notable for the high SES group. In other words, the high SES students may have had farther to fall, in terms of test scores. Zamudio reportedly controlled for both family background and school characteristics. 23

33 Although unusual, Zamudio s (2004) finding regarding high SES or upper middle class students being affected by mobility is consistent with the position of Wasserman (2001). In his research on student mobility in Canadian schools, Wasserman states that although socioeconomically disadvantaged students usually suffer the greater losses in achievement from changing schools, more advantaged students also face consequences (Wasserman, 2001, p.2). Benson, Haycraft, Steyaert, and Weigel (1979) also investigated the influence of student mobility on student achievement. They researched 1,007 sixth grade students in Colorado. After reviewing mobility information from their permanent records, the researchers found that approximately one-fifth of sixth grade students remained in the same school after enrolling in kindergarten. Conversely, four out of five sixth grade students did not remain in the same school. The impact on student achievement was that the students who were more transient had lower achievement percentiles on the Stanford Achievement Test - Reading section. Mehana (1995) reported results similar to Benson et al (1979). In contrast to Benson et al. s study, Mehana used grades as the dependent measure. Mehana focused on 988 Black urban students in the sixth grade. He found that 58% of the students changed schools a minimum of one time, not counting moves between elementary and middle school. Mehana reported that students who achieved an A had not made more than two school transfers. In contrast, students that had transferred greater than two times had averages of D and F. Mehana s study supported research by Levine, Corbett, and Wesolowski (1966), who also found that the more often a student changed school, the poorer the grades on his or her report card. Two studies conducted in the Austin Independent School District by Paredes (1993) also found a negative relationship between student mobility and achievement. Paredes indicated that student mobility alone was not the only factor influencing student achievement. One of the 24

34 studies explored four demographic factors and their influence on student achievement The four variables investigated were (1) student mobility (i.e., number of school moves), (2) income status- an estimate of parent s earnings, (3) ethnicity of the student, and (4) grade level of the student. Student mobility was defined as the number of withdrawals and reenrollment of each student. Students were placed into four categories; (1) stable over time, (2) moved during the current year of attendance in the school but not during the previous year, (3) did not move during the current year, (4) moved one or more times during the current year. Parents income was determined by applications for government-funded free and reduced lunch programs, a common way of categorizing students based on income status. The categories were free or reduced and not eligible for free or reduced prices. Ethnic categories included American Indian, Oriental, Black, White/other, and Hispanic. Students were studied in Elementary, Middle, and High Schools. The years of the study were Student achievement was measured with the Iowa Test of Basic Skills for second grade; and grades 3-6 were assessed with the now defunct Norm Assessment Program for Texas (NAPT). Paredes (1993) researched student s records as far back as 13 years. The research results indicated that students with higher numbers of moves had lower mean grade equivalents. Paredes also found negative results between student mobility and ethnicity, income status, and grade level. In each grade level from 2-8, Paredes found that mean grade equivalents were lowered with each student move, with grade 8 being the only exception. Paredes concluded that it was clear that there is some relationship between student mobility and student achievement, however, he did not report that student mobility alone caused low academic achievement. Paredes indicated that student mobility is one factor in student lives that can negatively affect learning (p.17). 25