ࡱ> y{x@ Չjbjbqq "-l(((>4"tP,m d     ""R"t"  1>L   Spare the Rod, Suspend the Child?: Discipline Policy and High School Dropouts Ren R. Rocha, Texas A&M University Funding Provided By: The Carlos H. Cantu Hispanic Education and Opportunity Endowment and The Project for Equity, Representation, and Governance. Abstract Recent events have put a spotlight on the issue of school discipline. While some contend that harsher discipline policies improve student performance because they foster a safer educational environment, others argue that they impair student success and disproportionately target minorities. Using data from a 184 school districts in Texas, I examine the differing ways in which disciplinary actions influence Anglo, Latino, and African-American student achievement. Evidence provided by this study supports those who contend that suspensions are often used to push out students. Also, the results indicate that discipline policies have a much more potent impact among minorities than they do among Anglos. The finding is most consistent for African-Americans. Thus, the increased emphasis placed on harsher disciplinarily policies in recent years may produce negative consequences that policymakers do not intend. Gregory (1995) remarks, the song of American education has long been sung to the tune of the hickory stick. Indeed, recent events, such as highly publicized school shootings, have put a spotlight on the issue of school discipline. Most of this renewed focus has centered upon zero-tolerance polices, the goal of which is to curtail discipline problems by establishing severe consequences for student misconduct. These consequences frequently involve penalties such as suspension and expulsion. While the connection between expulsion and failure to graduate from high school is self-evident, Skiba and Peterson (1999) note that suspension is a moderate to strong predictor of a students dropping out of school. Similarly, Ekstrom (1986) finds that sophomores who dropout had suspension rates three times that of those who stayed in school. Furthermore, suspensions can be seen as an approach whereby problem students are encouraged to dropout. This is only further compounded by the fact that 24 States do not mandate districts provide alternative means of education for students who are expelled or suspended (Civil Rights Project, 2000). Given this, the extent to which disciplinary policies, such as zero tolerance, are equally applied across racial lines would appear to substantially affect the number of minorities who dropout of school. Initial research indicates that the application of punitive measures is not equitable (Claiborne, 1999; Fasko, 1995; Gregory, 1995; Keleher, 2000; Shaw, 1990). Latinos, as well as African-Americans, face a greater probability of being expelled or suspended than Anglos. However, inflated rates of suspension are not consistent across all school districts for Latinos. Rather, in some districts Latinos are disciplined in numbers roughly comparable to the percentage of the population they comprise. In other districts, they are disciplined in numbers as high as four times what one would expect given the demographics of the student body (Keleher, 2000). For their part, African-Americans make up 32% of all suspensions nationwide, despite only making up 17% all students attending public schools. This contrasts sharply with Anglos, who make up 63% of the public student population, yet account for only 50% of suspensions (Civil Rights Project, 2000). The racial disparity that exists in the administration of zero tolerance policies is somewhat surprising given that zero tolerance, at least in theory, is supposed to create a uniform standard to deal with student misconduct. The way in which the standard is applied, however, usually gives a large degree of discretion to administrators. This allows such factors as racial prejudice to influence the decision (Keleher, 2002). While at first zero tolerance was meant to apply only to the most egregious transgressions (i.e.: taking a gun onto school grounds), it has been reconstrued to apply to a broader scope of violations (Civil Rights Project, 2000). This expanded definition usually includes the possession of drugs or any type of weapon. Anecdotal evidence illustrating exactly how broadly zero tolerance has been interpreted abounds. Two demonstrative examples given by the Civil Rights Project consist of a six year-old African American who was suspended for having a toenail clipper on campus in Harrisburg, PA, and a ten year-old African-American suspended for engaging in drug related activity. In this case, drug related activity consisted of wearing one pant leg above the knee. The inequitable application of school discipline policies does not end with zero tolerance; it also extends to the administration of corporal punishment. Here, both racial and gender inequities are evident (Fasko, 1995; Gregory, 1995). When combining the influence of race and gender, the bias in the use of disciplinary measures becomes clearer. Gregory (1995) finds that African-American males are sixteen times more likely to be given corporal punishment than white females. Considering that such punishments can often be used to pushout minorities (Arnez, 1978), discrimination in their application translates into disparate dropout rates. There have been several rationalizations, other than racial prejudice, proposed to explain this apparent inequity. The simplest explanation concludes that minorities are disproportionately disciplined because they commit a disproportionate amount of offences. However, this does not explain why although minorities are suspended and expelled at a higher rate than Anglos, this trend does not hold for less serious forms of punishment (Skiba and Peterson, 1999). If minorities simply misbehaved more often, then this should be evident not just in their suspension and expulsion rates, but in the rate at which they are assigned less serious forms of punishment also. Yet, minorities are over-represented when it comes to serious offenses to a far greater extent than they are in regards to minor offences. This raises concerns that minorities might be receiving serious punishments for minor offenses. Another explanation suggests that socioeconomic status, not simply race, is the driving force behind this inconsistency. However, multivariate analysis has found that race remains a significant factor even when socioeconomic status is taken into account (Shi-Chang, 1982). Districts with a large minority population are also more likely to purse policies of zero tolerance. This, to some extent, explains the variation which exists at the national level (Civil Rights Project, 2000). Obviously, if minority districts follow tougher discipline guidelines, then this would increase the percentage of minorities disciplined relative to Anglos. That said, it does not explain the differences noted within districts. Why minority districts are apt to adopt stringent disciplinary policies is also an intriguing question for future investigation. This increased emphasis on disciplinary issues would seem to imply that delinquency has become an increasingly important concern for administrators. When questioned about the most critical problems facing them, however, administrators typically rank mundane problems, such as tardiness and absenteeism, higher than school safety (Skiba and Peterson, 1999). Skiba and Peterson cite an NCES statistics which show that schools without zero tolerance policies are actually less likely to report serious incidents of crime than schools with such polices. This, of course, could simply be because zero tolerance policies are only initiated in schools with existing discipline problems. Nevertheless, this knowledge raises concerns regarding the ability of zero tolerance policies to effectively prevent criminal behavior in schools. In order to develop a clearer understanding of the use of disciplinary policies, one must consider them within the context of second-generation discrimination. Second-generation discrimination can be briefly defined as the use of academic grouping and discipline in a discriminatory manner so that minorites students are separated from Anglos (Meier and Stewart 1991). Meier and Stewart develop a model, based on their political theory of second-generation discrimination, in order to predict the levels of corporal punishment, suspensions, and expulsions among Latino students. Meier and Stewart (1991) find district size, social class, and Latino resources do not affect levels of corporal punishment among Latinos, suggesting that corporal punishment is not necessarily being employed for discriminatory ends. A significant relationship does exist between corporal punishment and Latino representation. Moreover, levels of black corporal punishment are also a significant influence. Meier and Stewart remark that this finding is disconcerting, noting no logical reason exists why the Hispanic punishment ratio should be negatively related to the black punishment ratio. As with corporal punishment, there exists a negative, statically significant, relationship between black and Latino suspensions and expulsions. Likewise Meier and Stewarts analysis shows that social-class variables are not the predominate factors affecting suspensions. It is important to observe that, especially with regards to expulsions, the level of Latino representation on school boards greatly reduces the application of disciplinary measures on that group. While the preceding discussion has focused on the discriminatory use of disciplinary measures, the relationship between academic grouping and disciplinary policy must also be considered. Academic grouping refers to sorting students accordingly to ability, needs, or aspirations. Academic grouping includes ability grouping, curriculum tracking, special education, bilingual education, and compensatory education (Meier, Stewart, and England 1989). Meier and Stewart contended the logic of second-generation discrimination would imply a positive relationship between negative academic grouping and corporal punishment. Indeed, Meier and Stewart (1989) find this to be the case with major urban school districts containing large proportions of blacks. However, such a relationship for Latinos can only be described as uncertain. Meier and Stewart (1991) posit two explanations for this uncertainty. First, the discriminatory use of academic grouping may suffice, eliminating the need for inequitable applications of corporal punishment. On the other hand, the lack of a relationship between academic grouping and corporal punishment may simply imply that the two function independently of one another, but nonetheless continue to function. Empirical analysis also reveals a clearly discernable link between suspensions and academic grouping. Meier and Stewart also indicate that suspensions work indirectly to affect expulsion rates, with suspensions being significantly and positively related to expulsion. In sum, the literature suggests that disciplinary policies are not applied equally across racial lines. In turn, disciplinary policies can considerably influence high school dropout rates. While a few pieces do attempt to examine this using multivariate analysis, most rely on simple percentages and bivariate correlations. While these establish a covariant relationship, they does not definitively establish a casual relationship between the way in which discipline is administered and what students eventually dropout. Hence, the literature is in need of more multivariate analysis, in order to determine whether or not there exists a relationship between harsher disciplinary punishments and ethnicity. This study aims to study disciplinary policies, race, and educational outcomes. Data The data for the purpose of this study was gathered by the Office for Civil Rights (OCR) Elementary and Secondary School Survey in 2000. A subset of this sample containing school districts in Texas is used. The survey contains information data covering incidents of corporal punishments, out of school suspensions, expulsions, and expulsions issued under the auspices of zero tolerance policies. In order to study discipline policies in a multivariate context, the OCR data is supplemented with data from the Texas Education Agency. This data includes information on dropout rates, demographic variables, standardized test pass rates, and local resources available to school districts. Variables Dependent Variables Dropout Rates The Texas Education Agency gathers data annually pertaining to dropouts. This data is also available by ethnicity. The dropout rate takes a longitudinal snapshot of different cohorts of students. The data is self-reported and, as it is part of a greater accountability system, TEA acknowledges that there exists an incentive for districts to underreport their dropout rate. TEA attempts to curb this possible behavior via an auditing process, however concerns regarding the reliability of this measure remain ( HYPERLINK "http://www.tea.state.tx.us/research/pdfs/rider71.pdf" Dropout Study: A Report to the 77th Texas Legislature 2000). These self-reported dropout rates have a minimum value of zero, with a maximum of 6.6 percent. The mean dropout rate is 1.4. TASS Exam Pass Rate During the time frame in which data for this study was gathered, Texas used a high-stakes test known as the TAAS. Passage of this test was mandated if students were to receive their high school diplomas, resulting in pass rates developing a great deal of salience among the public. Although students were not permitted to graduate without first passing the exam, actual failure of the exam was only the 11th most citied reason for dropping-out (Report on Public School Dropouts 1999). Thus, although the pass rate on the exam is far from a measure of dropouts, it provides indicator district quality that is salient to both the public and elected officials. Independent Variables Disciplinary Measures For the purposes of this study, there will be four independent variables of main interest: the corporal punishment ratio, suspensions ratio, and expulsions ratio, and the zero tolerance expulsions ratio. OCR gathers data on the number of incidents of each of these categories by race. This is then divided by the number of students enrolled within the district of the appropriate racial/ethnic composition. For example, the Latino suspension ratio has a mean value of .05 (with values ranging from 0 to .22). A one-unit shift in this variable would be the equivalent of a district moving from having issued no suspensions of Latino students to issuing as many suspensions to Latinos as there are Latino students. Together, these variables should provide a good gauge of the severity or laxness of a given districts discipline policy. Control Measures As with all research in education policy, the effect of the independent variables of interest should be seen within the context of other factors that may be affecting school district performance. Accordingly, I make an effort to control for the resources available to districts. This is done by taking account of the percentage low-income students (defined by the percentage of students eligible for Texas free lunch program), revenue per pupil, teacher experience, and the amount of state aid received by a district. Although these measures are bound to suffer from some degree of multicollinearity, together they should provide an appropriate control for the resources districts have available. Analysis and Findings In order to determine whether or not a relationship exists between disciplinary policies and Latino dropout rates, I regress the Latino dropout rate, as reported by the Texas Education Agency, against three indicators of school disciplinary practices: ratios of Latino corporal punishment, out of school suspensions, and expulsions. The findings are displayed in table one. The results are telling. The arguments made by Arnez (1978) and Skiba and Peterson (1999) that suspensions can be used to push out students, thereby making them a good indicator dropouts, appears to be supported by these findings, with suspensions being positively related to the Latino dropout rate. Interestingly enough, incidents of corporal punishment and expulsions fail to reach conventionally accepted levels of statistical significance. In order to determine if this relationship is consistent among other racial/ethnic groups, I replicate the same equations for African-Americans and Anglos. The results of African Americans are identical to those for Latinos, with suspensions being the only measure of school discipline that is statistically related to the dropout rate. The results for Anglos tell a different story. Unlike Latinos and African-Americans, none of the independent variables achieve significance, indicating that Anglos are perhaps less sensitive to the severity of discipline policies than are racial/ethnic minorities. The overall explanatory power of these models is quite limited, especially for Anglos. Thus, while there certainly appears to be a relationship between suspensions and high school dropout rates, several other factors play a key role in determining dropout rates. In order to establish how the relationship between discipline policies and dropout rates might appear in a multivariate context, I expand the models presented in table one. [Table One About Here] Table Two presents the results of the expanded model. The model controls for the percentage of low-income students in a district, the average number of years of teacher experience, the amount of state aid received by the district, revenue per pupil, and, in the case of Latinos, the percentage of students involved in bilingual education programs. I expect state aid and the percentage of bilingual and low-income students to be positively related to the dropout rate. Meanwhile, greater amounts of revenue per pupil and teacher experience should both work to lower the dropout rate among Latinos. I also expect districts that exercise disciplinary measures more freely to suffer from higher dropout rates. The models are more or less consistent with these hypotheses. [Table Two About Here] Districts with a more impoverished student body and those with more bilingual students experience higher dropout rates among Latinos. It appears that greater amounts of state aid, however, can help to offset this slightly. Interestingly, the data suggests that corporal punishment is actually negatively related to dropout rates. Suspensions, expulsions, and expulsions issued under zero tolerance policies, on the other hand, are all insignificantly related to dropouts. In order to determine if these findings are representative of other racial/ethnic groups I repeat the analysis for Anglos and African-Americans. This time African-Americans tell a different story than Latinos. Here, the data show corporal punishment expulsion rates to be insignificant determinates of high school dropouts. Meanwhile, suspensions show a significantly positive relationship. Thus, African-Americans seem most vulnerable to the pushing out strategies described above. Anglos appear to be unaffected by the severity of discipline policies. One possible explanation is that Anglos disciplinary action is not designed to discourage educational attainment. Alternatively, Anglos could simply be demonstrating greater resiliency than other groups. Although this study is primarily concerned with dropout rates, one cannot ignore the preoccupation of many school districts with standardized testing (especially in Texas). Indeed, Meier and OToole (2003) find that superintendents typically list performance on state mandated high stakes testing to be their foremost goal. Accordingly, while discipline policies have been shown to have an influence on dropout rates, any possible relation with high stakes testing results is important and should be of interest to practitioners. Table three shows the impact of disciplinary policies on TAAS pass rates. The controls used in Table Two are kept, with one notable addition. I also control for the percentage of minority teachers, as past research has indicated that a more diverse teaching body benefits the performance of both Anglo and minority students (Meier, Wrinkle, and Polinard 1999). [Table Three About Here] Unlike dropouts, out of school suspensions adversely affect the Latino pass rate. Thus, it appears that depriving students of instruction is not an effective strategy for improving Latino performance on the TAAS exam. Also, corporal punishment and explosions both fail to reach significance at the .1 level. While the results described in table two suggest that harsher discipline policies might lower the dropout rate among Latinos, this evidence indicates that they do so at the risk of jeopardizing performance on standardized exams. Once again, this analysis is repeated for Anglos and African-Americans. As with Latinos, suspension was the only significant discipline variable for African-Americans, negatively influencing that groups TAAS pass rate. All other variables behaved in a manner consist with the results presented in table two. Likewise, the results for Anglos are analogous to those presented in table two, with none of the four discipline measures showing a significant relationship to Anglo performance on the TAAS exam. Also of interest is the positive impact of a diverse teaching body for Anglos (a relationship which is also present for Latinos and African-Americans). This evidence would seem to support the results of Meier, Wrinkle, and Polinard, who describe a similar relationship. The preceding discussion would not be complete without an understanding of what motivates school districts to actively engage in harsh or lenient disciplinary policies. Table four presents a model wherein the severity of discipline policies is hypothesized to be a function of social and economic resources, the percentage of minority teachers, as well as the race and gender of superintendents. Resources are measured by taking account of the percentage of low-income students in a district, as well as several other indicators of school district wealth. [Table Four About Here] Model One of Table Four shows the analysis with the corporal punishment ratio as the dependent variable. The results indicate that social and economic resources do play a considerable role in the administration of corporal punishment, as both the percentage of low income students and the amount of revenue per pupil are significant and in the predicted direction. Interestingly, districts operated by female superintendents show a noticeable disinclination to dispense corporal punishment. The same does not hold true for Latino or African-American superintendents. Model Two considers the effect of resources on the suspensions ratio. Standing in stark contrast with corporal punishment, economic resources are unrelated to a districts willingness to issue out of school suspensions. African-American superintendents nearly show a significant unwillingness to use out of school suspensions (t-score 1.61), which is particularly notable given that African-Americans seem to suffer from out of school suspensions to a greater degree than other racial/ethnic groups. As was the case with suspensions, greater economic resources are not associated with a reluctance on the part of school districts of issue expulsions. On this occasion it is Latino superintendents who possess an aversion to using this specific mode of discipline. It could be that there exists far less discretion when it comes to issuing suspensions and expulsions than there exists with corporal punishment. In communities where resources are more abundant, concerned individuals can step in to ensure that corporal punishment is not used as a disciplinary option, or at least not used loosely. Communities with fewer resources are less able to apply pressure to districts. Synopsis and Conclusion Overall, these results seem to indicate that discipline policies have a much more potent impact among minorities than they do among Anglos. The effect is most consistent for African-Americans, who are hampered by suspensions both in terms of their dropout rates and performance on high stakes testing. Latino performance was likewise disadvantaged by high rates of suspension, however greater levels of corporal punishment were actually negatively associated with the Latino dropout rate. Evidence provided by this study appears to support those who contend that suspensions are often used to push out students. Moreover, the results indicate that this strategy will negatively influence student performance. The rate at which corporal punishment is used was also found to be dependent on the resources available to members of a school district. These findings compliment some of the qualitative studies done on the issue, as well as studies that have relied primarily on descriptive statistics. It appears that the increased emphasis placed on harsher discipline in recent years may produce negative consequences that policymakers may not intend. References: Arnez, Nancy L. 1978. Implementation of Desegregation as a Discriminatory Process Journal of Negro Education 47 (Winter), 28-45. Bireda, Martha R. 2000. Education for All Principal Leadership 4 (Dec), 8-13. Blair, Frank E. 1999. Does Zero Tolerance Work? Principal 79 (Sept), 36-37. Curwin, Richard L., and Allen N. Mendler. 1999. Zero Tolerance for Zero Tolerance Phi Delta Kappan 81 (Oct), 119-20. Cartledge, Gwendolyn, Linda C. Tillman, and Carolyn Johnson. 2001. Professional Ethics within the Context of Student Discipline and Diversity. Teacher Education and Special Education 24 (Winter), 25-37. Ekstrom, Ruth B. 1986. Who Drops Out of High School and Why? Finding from a National Study Teachers College Record (Spring), 356-73. Civil Rights Project. Opportunities Suspended: The Devastating Consequences Of Zero Tolerance And School Discipline Policies. June 2000. Claiborne, W. Disparity in School Discipline Found: Blacks Disproportionately Penalized Under Get-tough Polices. The Washington Post pA03, Dec 17th, 1999. Fasko, Daniel, Jr. 1995. An Analysis of Disciplinary Suspensions Paper presented at the Annual Meeting of the Mid-South Educational Research Association: Biloxi, MS Nov 8-10, 1995 Gordon, Rebecca, Libero Pianan, and Terry Keleher Facing the Consequences: An Examination of Racial Discrimination in U.S. Public Schools. Applied Reseach Center, March 2000. Gregory, James F. 1995. The Crime of Punishment: Racial and Gender Disparities in the use of Corporal Punishment in U.S. Public Schools. Journal of Negro Education 64 (Autumn), 454-462. Keleher, Terry. Racial Disparities Related to School Zero Tolerance Policies Testimony to the U.S. Commission on Civil Rights. Feb 18th, 2000. McAndrews, Tobin. 2001. Zero-Tolerance Policies ERIC Digest 146 (March) Meier, Kenneth J., and Laurence OToole. 2003. Public Management and Educational Performance: The Impact of Managerial Networking. Public Administration Review 63:6 689-699 Meier, Kenneth J., and Joseph Stewart, Jr. 1991. The Politics of Hispanic Education. Albany, NY: State University of New York Press. Meier, Kenneth J., and Joseph Stewart, Jr., Robert E. England. 1989. Race, Class, and Education: The Politics of Second Generation Discrimination. Madison, WI: University of Wisconsin Press. Meier, Kenneth J., Robert D. Wrinkle, and J. L. Polinard. 1999. Representative Bureaucracy and Distributional Equity: Addressing the Hard Question. The Journal of Politics 61 (4), 1025-1039. Shaw, Steven R, and Jeffery B. Braden, 1990. Race and Gender Bias in the Administration of Corporal Punishment. School Psychology Review 19 (3), 378-83. Skiba, Russ, and Reece Peterson. 1999. The Dark Side of Zero Tolerance: Can Punishment Lead to Safe Schools? Phi Delta Kappan 80 (Jan), 372-376. Slaughter-Defoe, Diana T., and Karen Glinert Carlson. 1996. Young African American and Latino Children in High Poverty Urban Schools: How They Perceive School Climate. Journal of Negro Education 65 (Winter), 60-70 Texas Education Agency. Dropout Study: A Report to the 77th Texas Legislature. December 2000. Texas Education Agency. Report on Public School Dropouts. Setember1999. < http://www.tea.state.tx.us/research/pdfs/9798drpt.pdf> Table One: Impact on Dropout Rates By Race Latinos Blacks Anglos Corporal Punishment -4.569 .801 2.213 Ratio (-1.46) (0.51) (0.57) Suspensions 5.104** 4.304** 1.646 Ratio (2.05) (3.05) (0.55) Expulsions -19.422 -3.172 5.572 Ratio (-1.54) (-0.37) (0.22) Expulsions (Zero Tolerance) 23.111 -6.358 -9.344 Ratio (0.54) (-0.50) (-0.23) Constant 1.535** .867 .914** (10.12) (1.29) (5.65) R squared .04 .06 .004 F test 1.99 2.77 0.21 N 184 174 184 (t scores in parentheses) *p<.1 **p<.05 Table Two: Impact on Dropout Rates By Race Latinos Blacks Anglos Percentage of .033** .013** .009 Low Income Students (5.24) (2.32) (1.60) Percentage of -.031** --- --- Bilingual Students (-3.03) --- --- Revenue -.000 -.000 -.000 per Pupil (-1.12) (-0.27) (-1.02) State Aid -.000* -.000 .001** (-1.77) (-1.39) (2.28) Teacher Experience -.049 -.110* -.142** (Years) (-0.86) (-1.67) (2.09) Corporal Punishment -9.321** -.115 1.791 Ratio (-3.04) (-0.07) (0.48) Suspensions 1.059 3.120** -.995 Ratio (0.44) (2.24) (-0.34) Expulsions -16.175 -3.135 -5.257 Ratio (-1.38) (-0.37) (-0.22) Expulsions (Zero Tolerance) 15.241 -3.005 1.292 Ratio (0.38) (-0.24) (0.03) Constant 2.405** -.280 -2.578* (2.28) (0.22) (-1.94) R squared .21 .12 .14 F test 5.13 2.71 3.43 N 184 174 184 (t scores in parentheses) *p<.1 **p<.05 Table Three: Impact on TAAS Pass Rates By Race Latinos Blacks Anglos Percentage of -.220** -.298** -.118** Low Income Students (-4.45) (-4.18) (-3.31) Percentage of -.148* --- --- Bilingual Students (-1.72) --- --- Revenue -.000 -.000 -.001 per Pupil (0.31) (-0.60) (-1.50) State Aid -.000 .001 -.002** (0.16) (1.14) (-3.40) Teacher Experience -.075 -.309 .119 (Years) (-0.18) (-0.54) (0.43) Percentage of .177** .247** .085** Minority Teachers (3.74) (3.97) (2.84) Corporal Punishment -2.136 -19.543 -18.681 Ratio (-0.10) (-1.33) (-1.17) Suspensions -28.802* -23.615* -14.602 Ratio (-1.67) (-1.93) (-1.23) Expulsions 135.553 -15.040 -12.049 Ratio (1.64) (-0.20) (-0.12) Expulsions (Zero Tolerance) -442.376 114.733 -17.700 Ratio (-1.56) (1.06) (-0.11) Constant 79.206 83.988** 98.500** (1.25) (7.61) (17.64) R squared .23 .21 .27 F test 6.05 4.80 7.14 N 184 174 184 (t scores in parentheses) *p<.1 **p<.05 Table Four: Social and Economic Influences on Discipline Policies (All races) Corporal Suspensions Expulsions Expulsions Punishment (Zero Tolerance) Percentage of .090** .011 -.006 -.005* Low Income Students (3.74) (0.41) (-1.13) (-1.77) Revenue -.001** -.000 -.000 -.000 Per Pupil (-2.05) (-0.51) (-0.13) (-0.26) State Aid -.000 .001 .000 .000 (-1.28) (1.42) (0.58) (1.34) Teacher Experience .224 .122 .036 .005 (Years) (1.48) (0.69) (1.11) (0.28) Percentage of -.0199 -.021 -.007 .002 Minority Faculty (-0.76) (-0.71) (-1.22) (0.59) Female Superintendent -.345** .090 -.020 .003 (-2.80) (0.63) (-0.79) (0.18) Black Superintendent .510 -3.174 -.014 -.082 (0.30) (-1.61) (-0.04) (-0.38) Latino Superintendent .961 -.056 .285* -.036 (1.19) (-0.06) (1.67) (-0.35) Percentage of -.048 .026 .008 .002 Black Students (-1.63) (0.78) (1.34) (0.48) Percentage of .000 .136** .006 .001 Latno Students (0.06) (4.48) (1.09) (0.26) Constant 2.679 1.446 .006 .107 (0.84) (0.41) (0.01) (0.28) R squared .22 .26 .04 .03 F test 4.98 6.13 0.82 0.52 N 184 184 184 184 (t scores in parentheses) *p<.1 **p<.05  The term zero tolerance is reported by Skiba and Peterson (1999) to have first been applied to the drug war in the 1980s. Its success in that initiative perhaps foreshadowed its success in eradicating delinquent behavior among students.  Of course, this could simply result from at-risk students being encouraged to dropout before they had an opportunity to fail the TAAS exam.  The expulsion ratio comes close, with a t-score of 1.64. This indicates that expelling students before they can fail the high stakes test may by a trick that works. * 44|779999::::3<5<Y<Z<@@WW[]ddJi]iii j"j#j3j4j5j6jhjqjjjjjkk+lClTlVlmmmmooppppqqXqYqZqvqqqqOrrr 6OJQJOJQJ OJQJo(6 5B* phCJH*0J>*B*ph jU>*5 j0JU>*CJ56K*UVWXYZ[\]l$a$$a$$`a$dd $d`a$d`vvvvvwwwwGwjwkwwwwwxxQxvxwxxxyxxxxxxxxyyyyyyy y!y"y#y$y>yDyLyMyNyZyzy{yyyyyyyz@zAzbzzzzzzz!{"{N{r{s{{{{{||G|h|i|||||||||||}}$}%}b ! Ed$`a$ $d`a$ $d`a$d`d!Z#6%)+1444Z7[7e7f7{77:::5=6= d^`d$a$d`d`d6=M=d=@@@yCzCCJJMNORTXVqVX[]]` bdd$da$d`$a$ d^`ddddJiKiLiMiNiOiPi\i]iiikUlVl2m3mmmnn:o;oo$a$$da$d`$a$oopppq r rrrsstuuu5v6vvvvvvvvvvv`dd[$\$rs sbsysstt)t*t+t,tttjuuuu4v5vvvvvvvw.w7w9w@wAwFwTwYw\w`wcwgw|wwwwwwwwwwwwwwxx xxx3xFxGxPx]xaxgxlxoxtxvxxxxxxxx 5OJQJOJQJ>*@CJOJQJ @OJQJ6@OJQJ5@OJQJ50J5 5B* ph66mH mH KvvvvvvvvvwwwwGwjwkwwwwwxxQxvxwxxxyxx &d P xxxxxx$yKyLyNyZyzy{yyzzzz {"{|G||||#}%}3}U}V}r}}}}}}}}}e~j~p~u~w~x~~~~~~~~~~~n%*KRV>*@CJOJQJ6@OJQJ5@OJQJCJ CJOJQJ>*@OJQJ @OJQJOJQJOxxxxxxxyyyyyyy y!y"y#y$y>yDyLyMyNyZyzy{y &d P {yyyyyyyz@zAzbzzzzzzz!{"{N{r{s{{{{{||` &d P |G|h|i|||||||||||}}$}%}3}U}V}r}s}}}}} &d P  &d P %}3}U}V}r}s}}}}}'~(~J~w~x~~~~~>mn,-_678Tp1`abc҂ӂ EFRӄCqrTW†Æ ABcd}~_}'~(~J~w~x~~~~~>mn,-_6`V]rwzـ݀(,/467`u~‚ǂʂς #,15:=BYbdnqvyº>*@OJQJ@CJOJQJ6@OJQJ5@ CJOJQJ>*@OJQJ 5OJQJ5@OJQJ @OJQJOJQJI678Tp1`abc҂ӂ EF ^` &d P  &d P yǃɃуԃك܃ ,56?AGJQinw|„Ȅ˄҄܄!&/169BKOX]afinDž̅хՅ؅݅ @CJOJQJ CJOJQJ_RӄCqrTW†Æ &d P  ^`38<@EIMRWφ~+,ӉՉj0JCJUCJ j0JU>*@CJOJQJ CJOJQJ@CJOJQJ! ABcd}~+ӉԉՉ+ӉԉՉ 1h/ =!"#$% i(@( NormalCJmH 0@0 Heading 1$@&56@6 Heading 2$d@&>*@@@ Heading 3$$@&`a$564@4 Heading 4$@&>*CJ,@, Heading 5$@&<@< Heading 6$$d@&a$5<A@< Default Paragraph Font<C@< Body Text Indent `&B@& Body Text2@2  Footnote TextCJ8&@!8 Footnote ReferenceH*(U@1( Hyperlink>*B*.P@B. Body Text 2CJ W@Q Strong5Y6QՃ,//Ճ  z zՃrxVyՉEMOUWY6=dovx{y|}6 ՉFHIJKLNPQRTVXZ%}ՉGS[334ՃX4 z  nu\a=BZ_a j ==5?:?F?K?GPOPfUnU]cbcccdddddde eeegghhhhhhi$iiijjlm mymzmm,n1nnnnnҁׁփ')5 C <KzQ#W#..;<QQa\m\]crcsccd$d6dKdhdddd fTfVfkfgg;iRiiipk|kkkkkQlUlbmmn(nnnnnqq%s&s?s@sFsGsbtetvvwwwwJxMx {{{{{{{{Ӄփ333333333333333333333333333333333333333333333rrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrochadC:\Documents and Settings\rrocha\Application Data\Microsoft\Word\AutoRecovery save of disptables.asdrrocha.C:\WINNT\Profiles\rrocha\Desktop\report024.doc Nick Theobaldhome:web:reports:report024.docrrNs}փ@LLXI**LL Ճ @GTimes New Roman5Symbol3 ArialY CG TimesTimes New Roman? Courier New"qh՚&뚁&]4wj6!20I2q Table One myname Nick Theobald Oh+'0  8 D P \hpx' Table One ablmynamenynaynaNormalnNick Theobaldo3ckMicrosoft Word 10.1@F#@`D@&q@*9j ՜.+,0 hp  ' mycompany 6I  Table One Title  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\^_`abcdefgijklmnoqrstuvwzRoot Entry F |1Table]WordDocument"SummaryInformation(hDocumentSummaryInformation8pCompObjX FMicrosoft Word DocumentNB6WWord.Document.8