For many, it is simple common sense to fail a child who does not pass the academic requirements for promotion to the next grade. For others, making children accountable for their academic performance is part of an overall strategy to raise the quality of education. Responding to the public and political call for educational reform and accountability, Georgia and Texas, Baltimore and Chicago along with many other school districts have decided to end the practice of social promotion in hopes of improving the quality of education (Eisner, 2000)

Defining social promotion is somewhat difficult because the meaning behind the practice has become so infused with professional, political and academic agendas that it seems impossible to reach common ground. On one side are those who deride the practice as promoting a student from one grade to the next regardless of academic achievement; they claim that this is a policy that short-changes the child, teacher, school and society (DiMaria, 1999). On the other side of the debate are those who point to the abundance of research that overwhelmingly suggests that keeping a child with his or her peer group is the best insurance for high school graduation (Darling-Hammond, 1998). Countering this plethora of research is the assumption that a high school diploma has little value if the student is simply passed along and given a degree without meeting specific criteria for graduation (Eisner, 2000). As evidence of the rift in the stakeholders' views, the two major teacher organizations cannot agree on the subject. The more conservative American Federation of Teachers (AFT) shudders at the thought that school districts ignore policies and laws that ban the practice of social promotion by stating that this rampant disregard creates a huge class of ill-prepared and unmotivated students (AFT, 2001). While conceding that social promotion without intervention is deleterious to a learner, the National Education Association (NEA) is on record as stating that retaining students is even more pernicious (NEA, 2001).

Examining Georgia's Decision to End Social Promotion

As a top priority for his education reform agenda, Governor Roy Barnes urged the legislature to end social promotion during the State of the State address in February 2001. Governor Barnes proclaimed, "the time has come to end social promotion in our schools" (Barnes, 2001). Reasoning that social promotion is unfair to teachers, Governor Barnes charged the legislature with passing a bill that would require every student to pass an exit examination before being promoted to the next grade (Barnes, 2001). Using the rationale that schools and teachers are held accountable for competency, the Governor insisted that no student be promoted to the next grade level until proficiency in the subject matter has been assessed with a criterion based standardized test (Barnes, 2001). In Governor Barnes' words,

Now, nobody wants to have to hold a child back in school. It is difficult for them to be separated from their peers. But if some children are still behind even after we have taken every step available to give them extra help--after school programs, alternative programs, special reading programs and so on--we owe it to them to make this difficult choice... . But mostly, we should do it in fairness to those students who are passing through our system today without learning what they need to know. By promoting a child who is not really ready, we say, 'It's OK if you don't learn.' Well, I say, it is not okay (Barnes, 2001).

On March 21, 2001, the legislature of Georgia complied with the wishes of the Governor by passing into law a bill which mandated that students in grades third, fifth, and eighth must pass a standardized examination to move up to the next grade, beginning with third graders in 2004. Children who enter third grade in 2003 would be required to pass a state reading test, while those matriculating the fifth grade in 2004 and eighth grade in 2005 would be required to pass both a state reading and mathematics exit examination (State Board of Education [SBOE], 2001).

A second chance to pass the test is allowed if the child fails to pass the test on the first attempt. If the child should fail a second time, a grade placement committee is convened "composed of the principal or the principal's designee, the student's parent or guardian, and the teacher of the subject of a test on which the student failed to perform satisfactorily" (SBOE, 2001). It is the initial charge of the grade placement committee to provide some sort of accelerated instruction to prepare the child for the third test. After three attempts, the official code directs the school to retain the student. At this point in the process, the parent(s) or legal guardian may appeal to the grade placement committee to permit the child to move up to the next grade level. Citing from Official Code 20-2-283, "The grade placement committee may decide in favor of a student's promotion only if the committee concludes, using standards adopted by the local board of education, that if promoted and given accelerated instruction, the student is likely to perform at grade level. A student may not be promoted on the basis of the grade placement committee's decision unless that decision is unanimous" (SBOE, 2001).

Teachers Views on Social Promotion

Governor Barnes' contention that teachers are critical of social promotion may have some credibility when the research is examined. Tomchin and Impara (1992) published a study showing eighty-two percent of elementary school teachers believed that retention helps children prevent future failure and seventy percent thought that the threat of failure motivates children to succeed. A whopping ninety-eight percent stated that they would never rule out the decision to fail a child (Tomchin & Impara, 1992). DiMaria (1999), in a study of New York City teachers, found similar results. In this 1999 study, sixty percent of teachers felt that students should never be socially promoted with thirty percent reporting that the primary grades were the best times to retain in grade.

Clearly, an overwhelming majority of teachers feel that social promotion frustrates children by burdening them with schoolwork that is too advanced for them to comprehend. Teachers believe that this burden is an imposition, one that makes teaching much more difficult because it forces them to deal with the under-prepared while trying to teach those who are prepared (Thompson & Cunningham, 2000). The prevailing view of teachers is that instruction is easier and more effective when variability of academic competence within the class is reduced (Foster, 1993).

Both of the teacher assumptions about student and teacher frustration, however, are not borne out by the educational research. It may be that teachers are not aware of the preponderance of retention research as evidenced by their reliance on anecdotal accounts from colleagues. Anecdotal reports by teachers often suggest that children benefit from retention, yet because of the decision to retain, there is no opportunity to see how well the children might have progressed had they been promoted (NAECS, 2000).

Mary Lee Smith wrote in Flunking Grades that teachers tend to access practical knowledge rather than formal knowledge. Practical knowledge is the sort of knowing that begins with personal experiences followed by future action based on these personal experiences. For example, one teacher remarked in Smith's study, "when my own son was retained, it was because he was too young for his age, and the next year he was a real leader in his class, and we never regretted that decision; and ever since then I have recommended that parents of young children in my class take the same step" (Smith, 1989, p. 133)

Based on clinical interviews, Smith suggested that the educational philosophies of teachers fall into distinct categories and that these beliefs are directly related to their opinions on retention. One category Smith identified was designated Nativists, teachers who believe that the physiological maturation of ability develops over time in stages. Predictably, Nativist teachers feel that children should not be exposed to developmentally inappropriate instruction. If a teacher was not a Nativist, Smith found that teachers could be grouped in three additional ways. 1) Remediationists are teachers who are active instructional and resource managers, 2) Diagnostic Prescriptors believe that deficiencies, such as auditory memory and visual-motor problems, can be corrected when identified with specific instruction and 3) Interactionists, teachers who feel that successful teaching begins with the prior knowledge and interests that the child possesses (Smith, 1989).

Smith found that the most likely to retain are Nativists because they are prone to see physical size and chronological age as reasons to hold children back. This is somewhat of a confounding finding given that Nativists' beliefs are congruent with some widely held theories of child development (Smith, 1989). While Nativists are more likely to retain students than the others, all teachers agreed in Smith's study that retention is beneficial both in the short as well as the long term. Through her interviews Smith recorded anecdotes such as the yearning to put the child at the top of the class and many claims to the effect that there are no stigmas attached to retention if the teacher and parents handle it well. Often Smith heard about the disasters that occurred when children are socially promoted and very few teachers named any negative effects of retention. All stated that it is best to err on the side of retention and if any harm was done, its effects are temporary. Perhaps the most disturbing finding Smith reported is that teachers often discounted the child's feelings of disappointment, failure and confusion or reluctantly acknowledged that if any emotional harm was done, the positives far outweigh the negatives. No one in Smith's research responded that social promotion was beneficial. Smith suggested that instructional efficiency pre-empts the child's best interest when she concluded, "The teacher is a self-interested theoretician. Though couched in the rhetoric of pupil benefits, her beliefs about retention are, perhaps unconsciously, conditioned by a wish for a more homogeneous and trouble free class" (Smith, 1989, p. 149).

Contradictions Between Research and Practice

A preponderance of control-group studies, structured to measure the comparison between retained students and students recommended for retention but promoted anyway, come down clearly on the side of promotion. What these studies show is that students who are recommended for retention, but are nonetheless promoted to the next grade, end up doing just as well as or, in many cases, perform better academically than non-promoted peers (Foster, 1993; NAECS, 2000).

While school performance is usually the focus of the debate, the most pernicious effect of retention is that the decision to fail a child usually results in dire social consequences. Children who have been retained demonstrate more social regression, display more behavior problems, suffer retention-related stress, and more frequently drop out of high school. (Grissom & Shepard, 1989; Frymier, 1997; NAECS, 2000).

While it may seem unfair to some, keeping a child with his or her peer group is the best decision almost all of the time. When grade retention is used as a solution for poor performance, it is assumed that the problem resides in the child's learning ability. But, this is rarely the case (Darling-Hammond, 1998; Frymier, 1997).

Shedding light on why children do not do well in school, studies show that the reasons for poor performance usually stem from non-academic factors such as a seriously ill parent or the death of a sibling or maybe a parent lost his or her job last year (Frymier, 1997). In addition, many students who failed a grade had been in an accident or were seriously ill during the year (Frymier, 1997). Now that the 2000 census has been published, data show that many non-English speaking children have become part of our schools (U.S. Census, 2001). Research tells us that if English is not spoken at home, a child is twice as likely to be retained during his or her schooling (Foster, 1993). Another suggestive statistic is that over half of the students who were retained in grade came from a broken home where moving from town to town was a frequent experience (Frymier, 1997). Concomitant with these factors is what research has coined as "retention bias," a tendency to retain a higher proportion of males, those with small physical stature, poor children and minority students (Foster, 1993; Frymier, 1997; Miller, 2001).

Because repeating a grade is a highly visible act, one that separates a student from his age peers, what is most disturbing about failing a child is what happens to them afterwards. Rather than accepting failure, children perceive the decision to repeat a grade as a punishment for something out of their control, a perception that discourages them from completing school (Foster, 1993). It is well documented that students who are held back do worse in the long run compared to students who are promoted, in part because they give up on themselves as learners (Denton, 2001). Even small children perceive that failing a grade is a serious social stigma. Stigmatizing children lowers their self-esteem, a psychological albatross that often results in a teen pregnancy or drug and alcohol use later on (Darling-Hammond, 1998; Foster, 1993). A review of sixty-six studies conducted between 1990 and 1997 found that sixty-five of them showed retention to be ineffective or harmful (Denton, 2001). In another 1997 study of twenty-three risk factors for school failure, students who fail a grade have many more problems, in every risk area, than those students who were promoted to stay with their peers (Denton, 2001; Frymier, 1997; Owens & Magliaro, 1998).

Standardized Tests and Consequences of Failure

In preparation for the implementation of the legislation that bans the practice of social promotion, the state of Georgia has administered for the second year a Criterion Referenced Competency Test (CRCT), commonly referred to as the State Curriculum Test, as a way to determine how many children might be retained in grade. While some reading scores have improved on average, about twenty-five percent of fourth, sixth and eighth grade children flunked the test (Georgia Department of Education, 2001). Looking at the best results, eighteen percent of fourth grade students failed the reading component. In the worst performing category, forty-one percent of eighth grade children failed to achieve a passing grade in mathematics (Georgia Department of Education, 2001). What is alarming is that, beginning in 2003, promotion to the next grade will be based on the results of this test, meaning that one in five, maybe more, students will be retained in grade. While there is some clamoring to revise the CRCT so that more students will pass, using a standardized test to make the critical decision to pass or fail a student has questionable validity (Salzer, 2001). The technical complexity of performance-based standardized tests, coupled with the fact that performance based tests are relatively new assessment and evaluation techniques, means that tests like the CRCT require constant revision in their early stages (Elmore, Abelman & Furhman, 1996). These changes are evidence that there are fundamental flaws in performance based tests, changes that result in improvement in some 2001 CRCT scores over 2000 CRCT scores.

Even though the practice of standardized testing is hard pressed to show that it can produce real gains in student learning, state policymakers count on test revisions to improve scores. With standardized testing, researchers have found that scores will initially be low and then rise for several years before leveling off. This upward trend caused by the "saw tooth effect" is due to teacher-led test preparation rather than to student achievement (Miller, 2001). Predictably, policymakers tout the spike in test scores, usually within the second year of test introduction, as proof that accountability measures are working.

While it is likely that the scores will get better because of this tinkering with the testing techniques and test preparation of students, there is no reliable evidence which suggests that performance-based tests will ever be perfected. Because the primary purpose of a standardized test is to gather data from a very large group of test takers as a way to evaluate if the overall curriculum needs to be improved, the CRCT should never be used to make a decision that affects an individual student (Miller, 2001). It is very important to note that a standardized test score does not reliably measure what an individual child actually knows because children are not consistent test takers. Even if the test was administered several times, the problem remains: snapshots cannot show a child's full range of capabilities (Kohn, 2000; Miller, 2001).

Challenging the Claims of Success in Chicago, Baltimore, and Texas.

With over fifty years of research showing that grade level retention provides no academic advantages to students, the practice of retention persists and is on the rise nationwide (Owings & Magliaro, 1998). The Consortium for Policy Research in Education (1990) reported that by the ninth grade, approximately fifty percent of all U.S. school students have been retained. If the goal of retention is to allow students more time to develop adequate academic skills so that they will be successful in subsequent years, why do the follow-up data on implemented programs throughout the United States show evidence to the contrary?

Beginning in 1996, the Chicago public schools promoted only third, sixth, and eight graders who obtained the minimum score on the Iowa Test of Basic Skills. Initial studies purporting to show the success of Chicago's program revealed that students, especially those with the lowest prior test scores, showed impressive gains after a full year of intervention and intensive summer instruction. However, follow-up studies revealed that learning gains dissipated after three years resulting in an increased likelihood of school drop out (Thompson & Cunningham, 2000; Denton, 2001; Holmes, 1989). In Baltimore, a similar story unfolded; a study conducted at Johns Hopkins University found that the performance of students retained during elementary school did improve modestly during the year they repeated and for several years thereafter (Denton, 2001). Again, follow-up studies revealed that initial gains faded, with sixty-five percent of the retained students dropping out of school as compared to eighteen percent of all other students. For students who were held back more than once, the drop out rate soared to ninety-four percent (Denton, 2001). What unfolds as one examines the research is that retained children are, on average, worse off than those who are socially promoted (Holmes, 1989; Shepard et al., 1996).

As for the Texas model, the study claiming success had serious methodological shortcomings that limit its validity (Denton, 2001). Texas researchers reported in a 1999 study that the performance of retained third graders improved over those students who failed the Texas Assessment of Academic Skills test but were promoted to the next grade. The validity of the test comes into question due to the large disparity between the number of students who were retained (400 students) and the number who were promoted (35,000 students). The 400 students, one percent of the total, represented the extreme low end of the range of test scores, so any subsequent test scores had no place to go but up, a statistical phenomenon known as "regression to the mean." In addition to the test reporting flaws, Texas policymakers have also manipulated students to give the illusion of increased test scores. As reported by Haney (2000), school officials exclude poor test takers from the tenth grade TAAS by either retaining them in the ninth grade, classifying them as learning disabled, or encouraging them to leave school and pursue the GED. By employing these tactics, Texas schools can report apparent test score increases for the tenth grade students.

A Snowball's Chance in Georgia: the viability of choosing litigation as a strategy to stop the use of high stakes tests to determine promotion or retention.

Except for the mountains in the northern part of the state, it doesn't snow very often in Georgia and, when it does, chances are that the frozen precipitation will not last very long. A snowball's chance in Georgia has the life expectancy of a fruit fly, about a day. The same analogy holds true for using the courts to overturn the legislature's decision to use the CRCT to determine promotion and retention. While some cases across the nation have been won on the local level, almost all have been overturned at the appellate level, meaning that victory in the courts is short-lived. Appellate courts have overturned challenges based on two cases, United States v. Fordice in 1992 and Personnel Administrator v. Feeney in1979. Basically, these two cases frame the issue by deciding that "Placement testing, exit examinations, and achievement tests may be used to assist in the determination of classroom assignments and eligibility for graduation, provided that the test results are not a reflection of past racial segregation policies, the testing is accurate, and the results are open to public scrutiny" (Deskbook Encyclopedia of American School Law, 2002, p. 489). Because the Equal Protection Clause of the Fourteenth Amendment forbids schools from engaging in intentional discrimination on the basis of race, color, national origin or sex , the first legal litmus test is whether or not a test perpetuates or preserves illegal discrimination. What the courts deemed important in Larry P. v. Riles in1984 is that the State Department of Education had to foresee that the test would have a significant disproportionate impact by race. Second, the department of education has to have failed to show the validity of the test for minority children. Also, the test must cause a stigma and irreparable injury to the student. Moreover, Larry P. v. Riles demands proof that failing the test will result in effective educational opportunities for the child. (Heubert & Hauser, 1999).

While the courts have recognized that high stakes tests have an adverse effect on minority children, they consistently reject the argument that these injuries are caused intentionally by the state. Instead, courts find that the state has a substantial governmental interest in education and that high stakes tests are a legitimate way to hold students accountable (Heubert & Hauser, 1999). Even when presented with clear accounts of racial bias, courts have refused to find that high stakes tests violate Title VI saying that they do not intentionally effect a particular race in an adverse manner (Heubert & Hauser, 1999; Deskbook Encyclopedia of American School Law, 2002). The Supreme Court forbids any practice that, while appearing to be a fair, perpetuates or promotes the effects of prior illegal segregation. This may mean that it is unlawful for any child who has attended an illegally segregated school at anytime in her/his schooling may not be subjected to a high stakes test designed for promotion or retention. It is rare today that a child has attended such a school, yet it opens the possibility that the courts could scrutinize a test more closely if a state or school district has had a recent history of segregation or intentional discrimination (Heubert & Hauser, 1999).

The courts have almost uniformly dismissed claims of intentional discrimination and have steadfastly upheld that high stakes tests are rationally related to legitimate state interests (Deskbook Encyclopedia of American School Law, 2002).

While there is an abundance of research that shows that retention has deleterious effects, such as low self-esteem, negative attitudes toward school and a reduced chance at succeeding at school, the courts dismiss such reasoning as speculative. This view, that educational research is mere speculation, was evidenced through the case Erik v. by and through Catherine V.v. Causby North Carolina in 1997, a decision that upheld a school board's decision to fail children based on a standardized test by rejecting the argument that students suffer irreparable harm when retained in grade because any potential harm is based solely on speculation. In stark contrast to the educational research, the court rationalized the situation completely oppositely saying that because retention gives a child more time in school to catch up, the state is doing its job by providing more resources to those who need them (Deskbook Encyclopedia of American School Law, 2002).

Texas courts also viewed retention as a part of a remediation benefit in the case, GI Forum v. Texas Education Agency in 2000. In this case, retention was affirmed as a part of a larger remediation process that provided those students who failed any portion of the exam with extra instruction intended to help them overcome their deficiencies. The court accepted the State's position that school accountability and mandated remediation helped to address the effects of prior discrimination in Texas because the exam provided the state with an objective way to assess student mastery of the skills and knowledge. When the issue of racial and cultural bias was raised, the court ruled that the exam was not fundamentally unfair to minority students because it measured what it claimed to measure and what was tested was taught. Because the test was aligned with the curriculum, the court decided that it was a valid test that met accepted standards (Deskbook Encyclopedia of American School Law, 2002).

What the states are relying upon in court is a conservative legal viewpoint that legitimizes high stakes testing for promotion so long as the tests comply with generally accepted standards for its use. These generally accepted standards have two central principles: 1.) a test score, like any other source of information about a student, is subject to error. Therefore, high stakes decisions like promotion should not be made automatically on the basis of a single test score (Shepard &Smith 1987; Darling Hammond & Falk, 1995) and 2.) a student's test score on a test should be used only in conjunction with other information sources in making such an important decisions as promotion to the next grade (Heubert & Hauser, 1999, p. 126). The state is clearly skating on thin ice here given that the generally accepted practice standards among psychometricians do not support the use of standardized tests as stand alone instruments to determine grade level promotion. Take for instance, it is generally accepted practice to supplement test scores with other assessment measures such as those performed by the teacher in the classroom (Heubert & Hauser, 1999; AERA, 1985, 1998; Joint Committee on Testing Practices, 1988). Moreover, there is legal precedence that could disrupt the states' position demonstrated by the decision made in the United States v. Fordice in1992, a ruling that rejected the use of one test score for placement decisions (Heubert & Hauser, 1999).

A legal strategy that offers a glimmer of hope rests on the concept of disproportionality. What must be proven is that grade retention is disproportionate among protected minority groups when compared to whites and that this disproportionality will decrease if equally reliable alternative assessments are used (Heubert & Hauser, 1999). It is well documented that grade retention is disproportionate among blacks/hispanics when compared to whites by a margin of 2:1 (Heubert & Hauser, 1999). The data shows that by ages 10-11, ten percent more blacks and Hispanics are retained; by ages 15-17, forty to fifty percent more are retained and when students reach 15-17 years old fifty percent of blacks have fallen behind (Heubert & Hauser, 1999). There is a possibility that if the citizens of Georgia used disproportionality as a strategy, the state's decision to use the CRCT could be ruptured when it is shown that those adversely affected are disproportionately protected minorities. Yet, any exuberance must be tempered with a good dose of healthy cynicism.

In Texas, where the TAAS examination is used to determine promotion and retention, groups representing Texas minority students sued the state with the claim that the criterion referenced test discriminated against minority students in violation of the Due Process Clause of the 14th Amendment. What plaintiffs must show is that there is a preponderance of evidence that the policy of high stakes testing disproportionately has an adverse impact on a protected group, but this is not easy to determine. To prove adverse impact, proof must be presented via a study of the entire pool of test-takers that the success rate for members of a protected class is significantly lower than if a random sample was examined (Deskbook Encyclopedia of American School Law, 2002). Fortuitously for the children of rural declining Georgia, there is data available that has been compiled by the University of Georgia Department of Housing and Consumer Economics (2002) along with the State of Georgia's Office of Education Accountability (2002) that meets the requirement for a study of the entire pool of test-takers showing that the success rate for members of a protected class is significantly lower than if a random sample was examined.

Who Gets Hurt the Most: Relationships Between Poverty, Race and the Criterion Referenced Competency Test Results from Rural Declining Georgia.

As a way to illuminate just how pernicious a law such as this will be to the poorest among us, we have conducted a study of school systems in the thirty-nine counties categorized as "declining rural counties" in Georgia, commonly referred to as the "black belt", so named because of the large number of African Americans who reside in them.

Our methodology compared county by county demographic data compiled by the University of Georgia Department of Housing and Consumer Economics (University of Georgia Department of Housing and Consumer Economics, 2002) with the State of Georgia's Office of Education Accountability's (State of Georgia's Office of Education Accountability, 2002) statistics for each district. Because there is compelling evidence that family background is the primary determinate for school achievement (Shepard & Smith, 1989; Elmore, Abelman & Furhman, 1996; Clotfelter & Ladd, 1996), our study includes an analysis of eight socio-economic categories; 1) percentage of population that is African American, 2) per capita income, 3) children in poverty, 4) African Americans in poverty, 5) female headed families in poverty, 6) un-wed births, 7) percentage of population without a high school diploma, and 8) percentage of African Americans without a high school diploma. Given that family background is such an important predictor of success, it is critical to supplement the school lunch index, the common statistic used to determine poverty in schools, with multiple economic and cultural measures.

When we first embarked on this study, two objectives were foremost: 1) compare the data gathered from these thirty-nine rural declining counties with statewide data; 2) present descriptive statistics that illuminate the relationship between CRCT scores and multiple socio-economic data. But, after we looked closer at the numbers, we discovered that in many of these counties, the school district data did not match up with countywide data. After comparing the county population demographics with the school systems data, it became apparent that many of these school systems have a discernable racial imbalance. Because this discovery suggests that race will matter in the decision to fail a child in Georgia, this research was expanded to include a discussion about the future of the rural African American community once the CRTC is implemented.

These thirty-nine counties form a constellation of poverty that slashes through the southern region of the state of Georgia. Forming contiguous pockets of counties in rural decline, the constellation extends in a chain from the far southwest corner to the eastern part of the state. As a way to boost the clarity of the research, we have chosen to present this data through a geographic journey whose starting point begins in the most concentrated area of poverty in southwest Georgia. Traveling across the state, this study will explore those counties that make up the belt buckle, a band that traverses along the mid-section of the state from west to east, followed by a discussion of those counties that are located in the east.

Southwest Rural Declining Counties

These twelve counties are found huddled along the Alabama and Florida border in the farthest southwest corner of the State of Georgia framed by the Chattahooche River to the west, Albany, Georgia as the closest city to the east and Columbus, Georgia to the north. There are no major roads cutting through nor are there towns of any substantial population. While there may never be an occasion for many travelers to ever pay the folks here a visit, these twelve counties are home to 92,400 Georgians, of which, 14,080 are children in the public school system.

As Table 1 shows, these twelve counties have two to three times more African American citizens than the rest of the state, most of whom live in poverty. Because school children here are likely to be poor and living in a household headed by an unwed mother who dropped out of school, the prospect for academic success is bleak. With up to seventy percent of the African American population lacking a high school degree, academic role models are hard to come by. When Table 2 is examined, it becomes apparent that all of these counties have reported similar socio-economic data.

A child attending school in these counties would have a one in three, at best a one in five, chance of passing the the third grade to the fourth grade once the CRCT decides his or her fate, a probability much worse than the rest of the state (Table 3).

Passing on to the sixth grade will be even more difficult, given that your odds are about 50/50 that you will pass the CRCT. While the scores statewide are improving in the sixth grade CRCT, school's scores are getting worse, ever widening the gap between rich and poor. If a child is so fortunate as to make it to the eighth grade in 2006, chances are better than even that he or she will not go to high school the next year because they failed the mathematics portion of the CRCT. As for comparing their school to the rest of the state (Tables 3 & 4), their school is in a free falling spiral, dropping significantly behind an abysmal statewide percentage of failing students.

Percent Failing CRCT 4th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Southwest Counties	32%	31%	50%
Statewide	26%	26%	38%
Percent Change	+23%	+19%	+32%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Southwest Counties	37%	47%	46%
Statewide	24%	36%	31%
Percent Change	+54%	+31%	+48%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Southwest Counties	30%	44%	56%
Statewide	18%	32%	41%
Percent Change	+67%	+38%	+37%

Table 4
African American Southwest Georgia Rural Declining Counties

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Southwest Counties	37%	35%	57%
African American Statewide	37%	34%	52%
Percent Change	0	+3%	+10

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Southwest Counties	38%	47%	50%
African American Statewide	35%	49%	45%
Percent Change	+9%	-4%	+11%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Southwest Counties	36%	51%	64%
African American Statewide	27%	45%	58%
Percent Change	+33%	+13%	+10%

With the exception of one of the twelve counties, Webster County, the racial balance of the schools when compared to the general population is egregiously disproportional. Calhoun County's population is sixty percent African American, yet Calhoun County Schools have too few whites to report, meaning that forty percent of the white children in Calhoun County attend private schools or are home-schooled. Sixty percent of the fourth graders, forty-four percent of the sixth graders and sixty percent of the eighth graders in Calhoun County failed at least one CRCT content area test.

Terrell County is the same story, only worse. Terrell is also sixty percent African American with no significant white representation in the schools. Having the lowest per capita income of around $16,000, Terrell County will face the fact that sixty-two percent of the fourth graders, sixty percent of the sixth graders and sixty-eight percent of the eighth graders will fail their respective grades.

Quitman County, where forty-six percent of the population is African American, has no whites attending the one elementary public school there. In Quitman County, teachers and principals will face the daunting responsibility for carrying out the failure sentence for eighty-one percent of the fourth grade class and sixty-five percent of the sixth grade.

The same situation obtains in Randolph County, with ninety-three percent of their children on the free lunch program; teachers and principals there will be forced to fail sixty percent of the fourth, sixth and eighth grade students. While Clay County CRCT scores are not as low as the others, this all African American school system, with a per capita income of $17,000 and sixty-five percent of African Americans in the county without a high school degree, will fail twenty-nine percent in the fourth grade, fifty-five percent in the sixth grade and because forty-eight percent of the eighth grade did not meet the mathematics standards of the CRCT, they too, will fail. The remaining schools in the counties, Baker, Early, Miller, Mitchell, Seminole and Stewart, are also disproportionately African American when compared to the general population. Most of the schools in these counties are two-thirds African American with county data showing a range of one-third to one-half of the population as African American.

Mid-State Rural Declining Counties

Fourteen counties form a contiguous swath of land beginning in Talbot County, situated between Columbus and Macon Georgia, southward along Interstate 75 to the Florida border, where Clinch and Ware Counties envelope the great Okeefenokee Swamp. These mid-state counties are home for 168,276 Georgians, 28,854 of whom are children in the public schools.

Table 5 paints a picture of economic and social crisis with data that shows per capita income well below the state average, resulting in significantly more children in poverty. As with the southwestern counties, school children in the mid-state counties are likely to have a parent who is a poor, unwed African American mother without a high school diploma.

Table 5
Mid-State Rural Declining Counties
SES Data Compared to State

When each county is examined separately in Table 6, the relationship between race, poverty and educational attainment becomes clearer. Dooley, Macon, and Talbot, counties with the largest African American populations are the poorest; while Bleckley, Irwin, and Pulaski counties, with many fewer African Americans, are better off. These data suggest that this economic divide persists because of the lack of educational attainment among African Americans. When the column "African American No High School" is examined in Table 6, the data describe a population that has, for the most part, found it difficult to graduate from high school. Ten of these fourteen counties have anywhere from sixty to seventy percent of the African American population without a high school degree; the remaining four counties have fifty to sixty percent without a diploma.

Table 6
Mid-State Counties in Rural Decline Socio-Economic Status

The socio-economic divide between Whites and African Americans in the declining rural counties of Georgia will surely be exacerbated through the implementation of the CRCT mandates. Because of retention in the third, fifth and eighth grades, African American children will be systematically encouraged eventually to drop out of school, resulting in the inability to command wages that might lift them out of poverty. Supporting the assumption that income is proportional to test scores, Tables 6 & 7 suggest that there is a relationship between CRCT test scores and the income earned by African Americans. When Mid-State African American CRCT scores are compared to statewide figures, the data show that poorer African Americans living in rural declining counties do worse than those African Americans who live in counties with higher income levels. Further, there is evidence that African American students who live in places where more of the African American population has earned a high school diploma do better on the CRCT than counties with less educational attainment.

Table 7
African American Mid-State Georgia Rural Declining Counties

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Mid-State Counties	48%	44%	61%
African American Statewide	37%	34%	52%
Percent Change	+30%	+29%	+17%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Mid-State Counties	49%	59%	48%
African American Statewide	35%	49%	45%
Percent Change	+40%	+20%	+7%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Mid-State Counties	34%	49%	64%
African American Statewide Counties	27%	45%	58%
Percent Change	+26%	+9%	+10%

The poverty to failure equation is repeated in Table 8 when the aggregate Mid-State CRCT scores show much lower results than the statewide data.

Table 8
Mid-State Rural Declining Counties

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Mid-State Counties	36%	33%	45%
Statewide	26%	26%	38%
Percent Change	+36%	+27%	+18%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Mid-State Counties	33%	46%	34%
Statewide	24%	36%	31%
Percent Change	+37%	+28%	+10%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Mid-State Counties	25%	39%	48%
Statewide	18%	32%	41%
Percent Change	+39%	+22%	+17%

Eastern Georgia Rural Declining Counties

Consider the Eastern Rural Declining Counties. These fourteen counties stretch vertically southward from counties that lie northwest of Augusta to rural areas southwest of Savannah. While these counties have very similar socio-economic data commensurate with very low CRCT test scores (Tables 9 and 10), some data are particularly noteworthy. Screven County's eighty percent mathematics failure in the eighth grade for African American students means that eight out of ten African Americans will not go to high school once the CRCT becomes the arbiter for promotion. Hancock County's statistics show that eighty percent of the children born in 1999 in the county live in single parent households, meaning that the kindergarten class in 2004 would have eight out of ten children living with a single parent.

Table 9
Eastern Rural Declining Counties

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Mid-State Counties	38%	38%	49%
Statewide	26%	26%	38%
Percent Change	+46%	+46%	+29%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Eastern Counties	34%	47%	37%
Statewide	24%	36%	31%
Percent Change	+42%	+31%	+19%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Eastern Counties	27%	46%	51%
Statewide	18%	32%	41%
Percent Change	+50%	+44%	+24%

Table 10
African American Eastern Georgia Rural Declining Counties

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Eastern Counties	47%	45%	61%
African American Statewide	37%	34%	52%
Percent Change	+27%	+32%	+17%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Eastern Counties	44%	57%	47%
African American Statewide	35%	49%	45%
Percent Change	+25%	+16%	+4%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
African American Eastern Counties	35%	54%	63%
African American Statewide Counties	27%	45%	58%
Percent Change	+30%	+20%	+9%

When Tables 11 and 12 are compared to the southeastern and mid-state counties, there is a distinct outlier, Glascock County, a county with a small African American population of 8.3%, which is not consistent with the data from the other thirty-eight counties. Located in the center of a chain of five rural declining counties, Glascock County stands out as the only rural declining county that has SES and CRCT data better than, or comparable to, the State averages (Table 13). Because Glascock County's per capita income is in line with the other rural declining counties, the variable that confounds the repeated pattern of poverty and low CRCT scores is whiteness (Tables 14 &15). Not only will the vast majority of Glascock County students be promoted, about one third of the CRCT test takers actually exceeded the standards, a statistic not seen in any of the other thirty-eight county data. When the data from the other four counties in the chain are compared, Glascock County's relatively low unwed birth rate appears to reduce the number of children in poverty, suggesting that the "children in poverty" and "un-wed mothers" statistics may also be predictors for CRCT achievement.

Glascock County's segregation from its neighbors leads to another assumption germane to this research, that the CRCT Test creates a new kind of discrimination - one that hides behind the appearance of fair testing to mask persistent inequalities in the quality of education that rural African American children receive in Georgia (McNeil, 2000). Walter Haney, of Boston College's Center for the Study of Testing, warns that, "The consequences of standardized tests for Black and Hispanic students are clearly criminal from an educational point of view. It remains to be seen whether they are criminal under the United States Constitution" (McNeil, 2000, p. 231).

Table 11
Eastern Georgia Rural Declining Counties SES Data Compared to State

Table 12
Eastern Georgia Counties in Rural Decline Socio-Economic Status

Table 13
Glascock, Hancock, Jefferson, Taliaferro, and Warren Counties SES Data

Table 14
Glascock, Hancock, Jefferson, Taliaferro, and Warren Counties CRCT Data

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	21%	27%	30%
Hancock	33%	35%	62%
Jefferson	46%	43%	56%
Taliaferro	60%	57%	65%
Warren	49%	35%	75%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	23%	34%	20%
Hancock	40%	51%	41%
Jefferson	37%	51%	43%
Taliaferro	42%	75%	46%
Warren	60%	67%	66%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	11%	22%	19%
Hancock	25%	42%	66%
Jefferson	35%	53%	63%
Taliaferro	47%	59%	76%
Warren	19%	39%	63%

Table 15
African American Glascock, Hancock, Jefferson, Taliaferro, and Warren Counties CRCT Data

Percent Failing CRCT 4^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	Too Few to Report	Too Few to Report	Too Few to Report
Hancock	32%	34%	62%
Jefferson	54%	50%	65%
Taliaferro	61%	56%	67%
Warren	46%	34%	74%

Percent Failing CRCT 6^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	Too few to report	Too few to report	Too few to report
Hancock	40%	51%	41%
Jefferson	45%	58%	49%
Taliaferro	48%	81%	48%
Warren	60%	66%	65%

Percent Failing CRCT 8^th Grade

Content Area	Reading	English/ Language Arts	Mathematics
Glascock	Too few to report	Too few to report	Too few to report
Hancock	26%	43%	67%
Jefferson	36%	56%	67%
Taliaferro	50%	63%	81%
Warren	19%	39%	63%

Giving Up on Going to High School

Overwhelmingly, African Americans in Rural Declining Counties are at much greater risk of failing the fourth, sixth and eighth grade CRCT than African Americans who live in cities, suburbs or rural growth counties. Yet, those most at risk are eighth graders who attend all-Black schools in the rural declining counties. Because they failed one or more of the content area tests, chances are that most of the eighth graders in these ten counties will not go on to high school. Clearly, the worst performing category was mathematics with only forty-four percent eligible to move up to the ninth grade. If the law were effective today, Taliaferro County would send four students to high school leaving thirteen behind, Talbot County would send only eleven, holding back forty-four. Calhoun fails thirty-three of their fifty-seven eighth graders, Clay County retains half of their thirty-two children. In the larger counties, Hancock retains eighty of one hundred twenty-one as Terrell County keeps ninety-one of their one hundred thirty-three eighth graders in middle school. Ten of the thirty-nine Rural Declining Counties fall into the 'all-Black' school system category with six of these, Calhoun, Quitman, Clay, Randolph, Stewart and Terrell, being located in the Southwest section of the State. The others, Taliaferro, Hancock and Warren surround all white Glascock County in the east with Talbot County being the lone all-Black school district in the Mid-State region.

Because each of these counties has its own school district, these schools are not considered illegally segregated. Drawing school district lines by county does not, superficially at least, appear to be gerrymandering given that each school district corresponds to an established county. Yet, segregation is, nonetheless, the result and the children in these schools suffer all the doisadvantages of a segregated education.

Future Consequences

Being the most impoverished counties in the state, these Rural Declining school systems faced formidable challenges before the legislation to end social promotion was passed. With an average of twenty-five percent of their populations under the age of seventeen, these counties have large numbers of children who need enormous resources to overcome obstacles to academic success. What makes the "Declining Rural Counties" of Georgia's plight unique is that the children who attend schools in these counties will be denied de facto their property rights to a public education when they drop out of school as a result of the practice of grade retention. From the data presented here, most of these counties already have very high drop out rates. What percentage will the drop out rate reach when thirty-five to fifty percent of all fourth, sixth and eighth graders will be retained in grade? The question that ought to be asked is "Is this legislation really intended to improve education or is it a strategy to reduce the State's financial obligation to the rural poor?" It is clear that failing masses of poor children will not improve pedagogy because punishing children with retention does not change teaching. What we do know is that the association between retention and dropping out is noted consistently throughout educational research. Without a doubt, flunking children increases the risk of dropping out of school (Frymier, 1997). Because these thirty-nine counties are very poor, and the tax base available for public schools is small, the State of Georgia compensates for this revenue deficiency by making exceptionally large contributions to these counties. Thus, while not stated as policy, it cannot be ignored that the CRCT will most likely save the State a considerable amount of money by reducing the number of students in schools in these counties.

While we dispute claims that the CRCT is a valid instrument to determine if a child should be retained in grade, we do not dispute that the CRCT is ironically a very reliable measure of economic resources (Kohn, 2001). As legislators extol the virtues of achieving academic excellence by using a "fair" test, like the CRCT, to determine if a child passes or fails, some wonder if the real agenda is a subtle form of class warfare intended to institutionalize intergenerational immobility and social stratification (Ohanian, 1999), a kind of violence that leaves behind the children with the least resources (Spring, 2000). Given the correspondence between the economic system and the role that the institution of education plays in perpetuating the class stratification of society (Bowles & Gintis, 1976), this legislation guarantees that the social reproduction of the society in rural Georgia will be preserved. The claim that flunking these children is for their own good is unconscionable, considering the dire social and economic punishments that will be imposed upon them. Ohanian derided such accountability measures "as cynical as handing out menus to homeless people in the name of eradicating hunger" (Ohanian, 1999, p. 31).

Chelsea's Story: Putting a Face on the Data

We first met Chelsea at the garage where we have our car fixed; it was about 3:30 p.m. on a sweltering spring day in rural South Georgia. Her great-grandmother had just picked her up from school to bring her to the garage to finish out the workday. Great-grandmother sat down at her desk where she worked as the receptionist and bookkeeper for the family owned garage. As I waited for my car to be repaired, we both asked the light brown skinned six-year-old with big brown eyes and braided black hair if she enjoyed her day at school. Chelsea smiled broadly, "Yes!" she exuberantly replied as she showed us the cover of a book she brought home. From all appearances, Chelsea was a healthy, well-loved first grader. But, this has not always been the case for Chelsea. Just recently, her great-grandmother took legal measures to have Chelsea taken from her mother after a man who was living in the home broke Chelsea's arm while trying to spank her. After he broke Chelsea's arm, the boyfriend badly burned her baby brother because he was crying too loudly. This abuse by one of Chelsea's mother's boyfriends was one of many horrors that this beautiful little girl had to endure over the course of her short life. As great-grandmother tells it, her grandson had a short affair with Chelsea's mother, an affair that brought Chelsea into the world. After her grandson left Chelsea's mother, great-grandmother felt obligated to support Chelsea in any way that she could. The offer to give financial support and emotional support was a selfless act of love given that great-grandmothers' extended family is barely making ends meet. In addition to the financial hardship, it was a painful situation for great-grandmother to witness the cruelty inflicted on Chelsea. Chelsea's mother was involved with too many men and had too many substance addictions to care for her children. While great-grandmother reported to the social service system many times that Chelsea was being neglected and abused, Chelsea remained with her mother. It took an act of violence against the children for the system to intervene on Chelsea's behalf. Even after the occurrence of such crimes committed against helpless children, Chelsea must still comply with a visitation plan that includes overnight stays with her mother.

Behind in School. After a conversation with great-grandmother later in the Spring, we became aware that Chelsea did not do very well academically during the year. As a result of her poor academic performance, due to excessive absences in the beginning of the year, Chelsea's teacher recommended that she repeat the grade. Already retained once in kindergarten, Chelsea's great-grandmother was reluctant to hold her back again. great-grandmother felt that by Chelsea now being in a loving and secure home, the prospect that she will catch up to her peers is promising.

Chelsea's teacher is basing her recommendation to retain on the failing grades that Chelsea received during the school year. Yet, if the Iowa Test of Basic Skills and a psychological assessment conducted at the behest of the Department of Family Services are considered in the decision, it is clear that Chelsea's poor grades were the result of poor attendance rather than cognitive ability. While Chelsea was absent for most class time when she lived with her mother, her scores were not as low on the ITBS as one might expect, and in some linguistic areas, her scores were high. As a result of high scores in certain verbal skills, Chelsea scored a composite word analysis in the sixty-fifth percentile. Chelsea's overall math abilities were average, but her composite math score was low because of a very low score on the computation section of the test.

In support of the decision to promote, Chelsea scored at a grade level of 2.2 in reading and 2.0 in math on the Kaufman Test of Educational Achievement in a low stress testing environment. The Kaufman Test showed that Chelsea is beginning to read independently and can add and subtract simple numbers. As an indicator of her mental ability, the Weschler Intelligence test showed that Chelsea is a child of average intelligence. The psychologist who observed Chelsea noted that she was a cheerful child with no signs of depression. The drawings she drew for the psychologist showed a loving family situated around a sturdy tree and a solid house. The psychologist concluded that Chelsea was now in a loving family that could be trusted for support.

The assessment went on the note that Chelsea is not a behavioral problem in school, she has many friends and is openly curious about new things. She is articulate, attentive, with good concentration and perseverance. Chelsea does not have any violent tendencies and, it is worth noting, that in light of her own physical abuse, she is quoted as saying that it is wrong to hurt anyone else. Having a birthday in October of 1993 makes Chelsea one of the oldest in her first grade class. If her great-grandmother had not insisted that Chelsea be promoted, holding her back again may have compounded her risk for dropping out of school later in life. Chelsea is progressing very well in the second grade and her chances for success are bright. Yet, another story could have unfolded if a standardized test was the arbiter for her future.

Preventing Failure with Changes in Education Policy

The fact that many school systems are overwhelmed by the increased number of under prepared students and use social promotion as a necessity, the public backlash should not be directed to testing and retention as the answers. Retention policies should be highly suspect given the lack of demonstrated effectiveness and prevalent bias against certain groups of children. Focusing public attention on standardized testing directs attention away from input issues that can substantially effect quality education and places the blame instead on students and teachers. Because the negative effects of failure on children's achievement, motivation, self-concept, and graduation rates are so well known, one of the most important decisions in a young person's life should not be based on the outcome of a standardized test score alone. Rather than use high-stakes testing, schools can employ less costly strategies that are proven to support children's achievement, thus avoiding the social promotion/retention issue altogether.

Smaller class sizes, especially in the primary grades, are frequently cited in the literature as promoting effective learning. "When class size goes down, learning goes up. It improves student achievement, particularly in the early grades and among students who are disadvantaged due to their socioeconomic background" (DiMaria, 1999, p. 6). In small classes, students who may be a risk can easily be identified and therefore receive additional support throughout the year (DiMaria, 1999).

If children are grouped in un-graded or mixed-age classrooms groupings, classrooms where the age span is greater than one year, children have the opportunity to "progress according to their individual rate of learning and development without being compelled to meet age-related achievement expectations" (DiMaria, 1999, p. 6). When learning takes place in un-graded or mixed-aged classrooms, children will be able to advance to the next developmental level at their own pace without the restriction of grade-level labeling (DiMaria, 1999, p.7).

We also advocate that schools should provide full-time kindergarten (NAECS, 2000) as well as offer alternative educational settings such as preschool, after school and summer school programs to assist those students that are lagging behind in certain academic areas (Thompson & Cunningham, 2000). These programs are critical for students who are economically disadvantaged and for whom English is a second language.

When children are assessed, promotion decisions should be based on multiple assessments, not on a single test or a single administration of a test (Miller, 2000). And, when standardized testing is employed, the results are best used to identify problems so that swift remediation and curricular changes can be quickly implemented. (Miller, 2001). More preventative measures and less punishment in the form of grade retention should guide education policy decisions to give extra assistance to children found to be lagging behind (Thompson & Cunningham, 2000).

Considering the Cost of Monetary and Human Capital

Most politicians would argue that the cost of programs to improve teaching would be too expensive to implement. However, to get a clearer picture of the short and long term cost that will be incurred after the tougher retention policy takes effect, we must look at two factors, one a short term and the other a long term societal cost. In the short term, district operating budgets will be challenged by an increase in expenses when one considers that the cost of retaining a child for one year increases the educational cost for that child by eight percent (Foster, 1993). Longer-term costs are associated with the expense associated with the inter-generational poverty that will be perpetuated as retention encourages more students to drop out of school. According to a study completed by Grissom and Shepard (1989), an annual retention increase of five to seven percent will result in a cumulative increase of twenty percent to the present drop out rate.

Proponents of retention might counter these two points by speculating that holding children back will improve their academic achievement with the reasoning that this improvement in academic skill will keep them in school. Yet, research paints a completely opposite picture with evidence that being overage in school plays a larger role in the decision to leave school than does academic achievement. Even when retained children do better academically, they drop out anyway. On average, the drop out rate is thirteen percent higher for over-aged children than the drop out rate for normal aged children (Grissom, & Shepard, 1989). Initially, retention advocates will boast that the drop out rate has been lowered as a result of the get tough accountability policy. But keep in mind that the first few graduating classes will have all the at-risk students removed. It is also important to look beyond high school drop out rates because many of the students who were retained will leave school before they enter high school.

After an analysis of the CRCT results, it is apparent that the rate of retention will rise dramatically, precipitating a proportional increase in per student costs. To be fair, schools should only be held accountable for factors they can control, and, therefore, the decision to retain or promote a student should be made at the local level where socio-economic factors can be taken into account (Clotfelter & Ladd, 1996). In addition, the state of Georgia will also have to bear the burden of future societal costs if and when the student drops out of school. Is it not more cost effective to seek long-term lasting cures for poor student performance while students are in school rather than deal with the consequences later?

The costs do not stop here. As we discussed earlier, school systems need to be aware of the legal ramifications of retention and the potential for litigation. Not only could parents bring litigation that schools did not provide adequate educational resources for their children to succeed (D. W. Albritten, Executive Director, Georgia Association of Educators, personal communication, November 24, 2001), they could also challenge unfair school policies.

Recommendations

Not only is there a preponderance of evidence that there is no academic benefit from retention, such practices also appear to be harmful to the social and emotional development of children. Since retention policies do not address the needs of under-prepared children, it is recommended that the limited resources of school systems be redirected toward the above listed alternative programs so that education is more responsive to what is best for children, not for institutions, politicians, or professionals. In the state of Georgia, several of these alternatives, such as full-time preschool and kindergarten, and smaller class sizes, have already been proposed or implemented. Instead of using high-stakes testing, Georgia's educational system needs to provide a prevention and remediation program during the school year rather than wait until after the student fails the criterion test.

Moreover, education reform decisions need to be grounded in research rather than influenced by public demand for more stringent educational standards. If higher academic standards are the goal, then reform programs must speak to the reasons why student do poorly and offer meaningful assistance to improve learning outcomes. The educational community can no longer afford to ignore the consequences of policies and practices which disproportionately assign the burden of responsibility to the child rather than to the program, a reproach that places the child at risk of failure by generating apathy toward school and personal demoralization.

Only when the stakeholders abandon the slogan that greater retention means more rigorous standards can schools begin to meet the pressing needs of children. By taking alternative paths away from retention, schools become more accountable to the holistic needs of the child.

References

American Educational Research Association. (1985). Standards for educational and psychological testing. Washington, D.C.: American Psychological Association.

American Educational Research Association. (1998). Draft standards for educational and psychological testing. Washington, D.C.: American Psychological Association.

American Federation of Teachers (AFT). Eliminating social promotion. Retrieved November 12, 2001, from http://www.aft.org/edissues/socialpromotion/eliminat.htm

Barnes, R. Governor Barnes' 2001 Education reform initiative. Retrieved September 21, 2001, from http://www.ganet.org/governor/2001_ed_remarks.html

Berliner, D.C. & Biddle, B.J. (1995). The manufactured crisis: Myths, fraud and attack on the America's public schools. New York: Longman.

Bloom, B.S. (1956). Taxonomy of educational objectives: The classification of educational goals. New York, NY: Longmans.

Bowles, S., & Gintis, H. (1976). Schooling in Capitalist America. Education reform and the contradictions of economic life. New York, NY: Basic Books.

Chall, J. (2000). The academic achievement challenge: What really works in the classroom? New York: Guilford Press.

Clotfelter, C. T, & Ladd, H. F. (1996). In H. F. Ladd (Ed.), Holding schools accountable: Performanced based reform in education (pp. 23-64). Washington, D.C.: The Brookings Institution.

Coleman, J., Campbell, J., Wood, A., Weinfeld, F., & York, R. (1966). Equality of educational opportunity. Washington, D.C.: U.S. Department of Health, Education and Welfare, Office of Education.

Consortium for Policy Research in Education. (1990). Repeating grades in school: Current practice and research evidence. (Report No. RB-04-1/90). CPRE Policy Briefs (ERIC Document Reproduction Service No. ED323585).

Cruickshank, D. R., Jenkins, D. B., & Metcalf, K. K. (2002). The act of teaching. New York: McGraw-Hill Higher Education.

Darling-Hammond, L. (1998, November). Avoiding both grade retention and social promotion. The Education Digest, 64, 48-53.

Darling-Hammond, L., & Faulk, B. (1995). Using standards and assessments to support student learning: Alternatives to grade retention. In Report to the Chancellor's Committee on Grade Transition Standards. National Center for Restructuring Education, Schools and Teaching. New York: Teachers College Press, Columbia University.

Denton, D. (2001, January). Finding alternatives to failure: Can states end social promotion and reduce retention rates? (Report No. UD 034 039). Atlanta, GA: Southern Regional Education Board. (ERIC Document Reproduction Service No. ED451268).

Deskbook encyclopedia of American school law. (2002). Birmingham: Oakwood Legal & Business Publishing.

DiMaria, M. J. (1999). Issues of social promotion. (Report No. PS 028 203). New York, NY: New York City Board of Education. (ERIC Document Reproduction Service, No. ED437208).

Doll, W., Jr. (1993). A post-modern perspective on curriculum. New York, NY: Teachers College Press.

Eisner, C. (2000). Ending social promotion: Early lessons learned. A report on early lessons learned in the efforts to end social promotion in the Nation's public schools. (Report No. UD 033 891). Washington, DC: U.S. Department of Education. (ERIC Document Reproduction Service No. ED448234).

Ellmore, R.F., Abelmann, C.H., & Furhman, S. (1996). The new accountability in State education reform: Policy, practice and performance. In H. F. Ladd (Ed.), Holding schools accountable: Performance-based reform in education (pp. 65-98). Washington, D.C.: The Brookings Institute.

Ellwein, M. C., & Glass, G. V (1989). Ending social promotion in Waterford: Appearances and reality. In L. Shepard, & M.L. Smith (Eds.), Flunking grades: Research and policies on retention (pp. 151-173). Philadelphia, PA: Falmer Press.

Foster, J.E. (1993, Fall). Retaining children in grade. Childhood Education, 70, 38-43.

Frymier, J. (1997, February/March). Characteristics of students retained in grade. The High School Journal, 80, 184-190.

Georgia Department of Education. Education Reform Initiative. Retrieved August 8, 2001, from http://www.doe.k12.ga.us/sla/ret/General-CRCT.html

Georgia State Board of Education (SBOE). Official Code 20-2-283. Retrieved November 12, 2001, from http://www.ganet.state.ga.us/services/ocode.htm

Grissom, J.B., & Shepard, L.A. (1989). Repeating and dropping out of school. In L. Shepard, & M.L. Smith (Eds.), Flunking grades: Research and policies on retention (pp. 34-63). Philadelphia, PA: Falmer Press.

Haney, W. (2000). The myth of the Texas miracle in education. Education Policy Analysis Archives, 8(41). Available at http://epaa.asu.edu/epaa/v8n41/.

Heubert, J. & Hauser, R. (Eds.) (1999). High stakes: Testing for tracking, promotion, and graduation. Committee on Appropriate Test Use. Washington, DC: National Academy Press.

Holmes, C.T. (1989). Grade level retention effects: A meta-analysis of research studies. In L. Shepard, & M.L. Smith (Eds.), Flunking Grades: Research and Policies on Retention (pp. 16-33). Philadelphia, PA: Falmer Press.

Kohn, A. (2000). The case against standardized testing: Raising scores, ruining the schools. Portsmouth, NH: Heinemann.

Ladd, H. F. (1996). Holding schools accountable: Performanced based reform in education. Washington, D.C.: The Brookings Institution.

Malone, G., & Bowser, P. (1998, March). Debate: Can retention be good for a student? NEA Today, 16, 43-45.

Miller, D. W. (2001, March). Scholars say high-stakes tests deserve a failing grade. The Chronicle of Higher Education, 47, A14-A16.

Miller, J. (1983). The educational spectrum: Orientations to curriculum. New York, NY: Longman.

National Association of Early Childhood Specialists in State Departments of Education (NAECS) (2000). Still! Unacceptable trends in kindergarten entry and placement. A position statement. (Report No. PS 028 611). Chicago, IL: ERIC Clearinghouse. (ERIC Document Reproduction Service No. ED445775).

National Education Association (NEA). NEA Today Online. Retrieved November 12, 2001, from http://www.nea.org/neatoday/0003/presview.html

McNeil, L. M. (2000). Contradictions of reform: The educational costs of standardized testing. New York: Routledge.

Nieto, S. (1992). Racism, discrimination, and expectations of students' acheivement. Affirming Diversity. New York, NY: Longman.

Ohanian,S. (1999). One size fits few: The folly of educational standards. Portsmouth, NH: Heinemann.

Owings, W.A., & Magliaro, S. (1998, September). Grade retention: A history of failure. Educational Leadership, 56, 86-88.

Paratore, J. R., & McCormack, R.L. (2000). Responding to research in grouping: Flexible grouping in the middle grades. In K. Wood, & T. Dickinson (Eds.), Promoting literacy in grades 4-9 (pp. 402-420). Boston, MA: Allyn and Bacon.

Peterson, P. L. (1989). Alternatives to student retention: New images of the learner, the teacher and classroom learning. In L. Shepard, & M. L. Smith (Eds.), Flunking grades: Research and policies on retention (pp. 174-201). Philadelphia, PA: Falmer Press.

Reardon, S. (1996, April). Eighth grade minimum competency testing and early high school dropout patterns. Paper presented at the annual meeting of the American Education Research Association. New York, NY.

Salzer, J. (2001, August 16). Tests show student gains, but many still behind. The Atlanta Constitution, pp. A1, C1, C4.

Shepard, L.A., & Smith, M.L. (1989). Flunking grades: Research and policies on retention. Philadelphia, PA: Falmer Press.

Shepard, L.A., Smith, M.L. & Marion, S.F. (1996). Failed evidence on grade retention. Psychology in the Schools, 33 (3), 251-261.

Smith, M.L. (1989). Teachers' beliefs about retention. In L. Shepard, & M.L. Smith (Eds.), Flunking grades: Research and policies on retention (pp. 132-150). Philadelphia, PA: Falmer Press.

Spring, J. (2000). American education (9th ed.). Boston, MA: McGraw-Hill Higher Education.

State of Georgia Office of Educational Assessment. Georgia's report card. Retrieved December 27, 2001, from http://www.ga-oea.org

Thompson, C.L., & Cunningham, E.K. (2000). Retention and social promotion: Research and implications for policy. (Report No. UD 033 924). New York, NY: ERIC Clearinghouse on Urban Education. (ERIC Document Reproduction Service No. ED449241).

Tomchin, E.M., & Impara, J.C. (1992, Spring). Unraveling teachers' beliefs about grade retention. American Educational Research Journal, 29, 199-223.

United States Bureau of the Census. School enrollment 2000. Retrieved November 12, 2001, from http://www.census.gov/population/www/socdemo/school.html

University of Georgia, Department of Housing and Consumer Economics. Georgia facts. Retrieved January 5, 2002, from http://www.ga-facts.net

Walberg, H., & Fowler, W. (1987). Expenditure and size efficiency for public school districts. Educational Researcher, 16, 5-13.

About the Authors

Donald R. Livingston, Ed.D.
Email: dlivingston@lagrange.edu

Don Livingston is an Assistant Professor in the Education Department at LaGrange College in LaGrange, Georgia. Don completed his doctorate at Georgia Southern University in Curriculum Studies in 2000.

Sharon M. Livingston
Email: livingston_Sharon@hotmail.com

Sharon is a doctoral fellow at Georgia State University in Atlanta, Georgia where she is pursuing the Ph. D. in Educational Policy Studies.

Their research interests include student performance, retention policy and teacher quality. They are presently engaged in research that examines the relationships between teacher educational levels and student achievement in high poverty rural counties in Georgia.

General questions about appropriateness of topics or particular articles may be addressed to the Editor, Gene V Glass, glass@asu.edu or reach him at College of Education, Arizona State University, Tempe, AZ 85287-2411. The Commentary Editor is Casey D. Cobb: casey.cobb@unh.edu .