Challenging Assumptions about the Achievement Gap.PDF

Challenging Assumptions about the Achievement Gap Author(s): Al Ramirez and Dick Carpenter Source: The Phi Delta Kappan, Vol 86, No (Apr., 2005), pp 599-603 Published by: Phi Delta Kappa International Stable URL: http://www.jstor.org/stable/20441858 Accessed: 10/12/2013 12:39 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive We use information technology and tools to increase productivity and facilitate new forms of scholarship For more information about JSTOR, please contact support@jstor.org Phi Delta Kappa International is collaborating with JSTOR to digitize, preserve and extend access to The Phi Delta Kappan http://www.jstor.org This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions A SpecialSectionon theAchievementGap Challenging Assumptions The national dialogue about the achievement gap can help policy makers and educators findways to better serve minority students However, school policy and practicemust be founded not on perceptions of group stereotypes, Mr Ramirez and Mr Carpenter argue, but on knowledge about each student's needs and strengths * ? BY AL RAMIREZ AND DICK CARPENTER ERE IN Colorado, snow is particularly significant.It affects our economy through winter recreation and tourism,reducesthe dangerof forestfires, 4J *.,#*? and provideswater for most residents And while to thecas ual observer the snow all looks the same,Coloradansknow differently We evaluate each snowfall not only by its quantity but also by its qual ity, that is, how wet it is Sometimes themoisture contentof the snow is low,which sets off a rush of snow boarders and skiers to themountains AL RAMIREZ is an associate professor in the Department of Educational Leadership & Policy Studies, University of Denver DICK CARPENTER is an assistantprofessor in the Department of Leadership, Research, and Foundations, University of Colorado, Colo rado Springs Illustration by Mario Noche APRIL 2005 599 This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions but supplieslittlewater to the arid landscape Other timesthemoisture content ishigh,which contributes greatlyto thestate's watersupply Thus differences with in thegeneralcategoryof snowarecriticaltoour state's healthand future Similarly,"within-group" differencesareimportant to recognizewhen we look at student achievement, particularlyas it relatesto raceor ethnicity.Since the Brown v.Board ofEducationdecisionmore than 50 yearsago,much attentionhasbeenpaid to significant differences.This focus resultedin "between-group" policiesandpracticesdesigned to reducesuchdispar ities.Yet,until quite recently,educationalresearchers, policymakers,and practitionershavepaid far lessat tention towithin-groupdifferencesthatareprobably as importantas thosebetweengroupsand could, in fact,help us figureout how tonarrowthedifferences betweengroups Take, forexample,Latinos,who now constitutethe largestminority group in theU.S andwho are cer in thepublicschools.' As agroup, well represented tainly Latinostudentssharemany similarcharacteristics that set themapartfromothergroups.On average,Latino more segregated studentstendtobepoorer,attend schools, and live in urban areas.Latino students also account for the largestnumberof studentsservedinprogramsof While thesecharacteris acquisition English-language tics typify the group of studentswe call Latinos, itwould be amistake to assume that all Latino students have white studentsintoa singlecomparison group.In short, the importance ofwithin-groupdifferenceseclipsesthe importanceof between-groupdifferences ACHIEVEMENT GAP RESEARCH Researchon theacademicachievementgapbetween majority and minority students is sometimes misap plied by policymakers and practitioners,and this in turncan leadto ineffective and evencounterproductive programsforstudents.? Media coveragefurtherexacer batesthismisunderstanding about the laggingacadem icperformanceofminority groups, for it oversimpli fiescomplexdata inorderto fit theconventionsof news reportingand tomanufacturecatchyheadlines More over,policydiscussionsanddebatesabout theachieve ment gap have missed themark by casting the problem as a "minority group"phenomenon,without consider ing thedynamic interplayof variablesthataffect the learningof any individualchild Of the achievement gap research that does consider factors in addition to race or ethnicity, much of it in volvessuchhome-basedvariablesas socioeconomicsta tus,home language,and parent involvementor such school-based variablesas schoolsegregation and teach er quality.However, the findingsarefarfromconclu sive Beginningwith home-basedvariables, much of the researchindicatesthat the income levelof a student's family ishighly correlatedwith academicsuccessin similarneeds or requirethe same typeof education.2 Yet currentpoliciesand educationalpracticesdirect school, a phenomenon that is indeed true for Latino Latinostudentsarebuilton suchassumptions students.4 ed toward Moreover, in some studies the effectof so and have had the unintended consequence of hurting cioeconomicstatusoften overwhelmsthe relationship thestudents'futures,educational andotherwise Among between race or ethnicity and academic achievement, theseovergeneralized policiesandpracticesarepresum sinceminoritygroupstendtobe overrepresented among Yet that all with the ing students SpanishsurnamesneedEng not all researchers agree about the impact poor lish-language-acquisition classes;creatinga policyof de of socioeconomicstatuson studentlearning,and some factosegregationby assigningLatino studentsonly to cite other factors as having more influence.' schoolswith EnglishasaSecondLanguage(ESL)pro home language, theresearch remains Regarding mixed Latinostudentsarepotentialdrop with cross-cuttingissues.Some research and crammed grams;andpresuming ers believe that a student's language background is cen outs ratherthancollege-boundstudents When policymakersandeducationprofessionalsre tral to success in school, particularly when it is related main obliviousto thesefalseassumptions, misinterpre to the level of parents' education.6 For example, the tationsoccur,and stereotypicalthinkingprevails.In U.S Department of Education reported that, in 1999, deed, our investigation of the achievement gap under of raceand ethnicity, scoresthe relativeinsignificance comparedtootherfactorsthatmost affectstudentlearn ing.Furthermore, we havefoundthatthe"achievement gap"betweenLatinoandwhite students may be a "phan tomgap"derivedfromthepracticeof lumpingallnon 600 the percentage of Latino parents with a high school or higher educaton was 49% for thosewho spoke mostly Spanishat home and 83% for thosewho spokemostly Englishathome.Other researchers find thatmaintain ingSpanishas thehome languageenhancesacademic achievement when combinedwith other factors?7 Still PHI DELTA KAPPAN This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions others contend that language back groundaccountsfor littleinexplain Take the Lead Become a Certified TESA Coordinator! ing student achievement.8 In contrast to themixed findings on the role of language, there isgen TESA is a dynamic, research-based training pro gram designed to promote teacher behavior that creates equity in the classroom, improves student academic performance and attendance, decreases eralagreement on amongresearchers the importance of parent involve ment, particularly for black and La studentdisciplineproblems, and improves class room climate Research reveals TESA classrooms tinostudents.9 While parentinvolve outperform non-TESA classrooms ment takesmany forms, numerous researchers haveconcludedthatthe most significant type isassisting chil dren with schoolwork at home."' Par ent involvement also plays an impor tant role in students' course-taking patterns For example, JamesValadez illustrateshow Latino parents influ ence theirchildren'senrollmentin algebra and advanced mathematics courses " Turningto school-based variables, someresearchers concludethatschool segregationsignificantlyaffectsthe academic achievement of minority students.Gary Robertsdescribesa spiraling relationship inwhich stu Aug 4-5, 2005 - San Diego, CA Aug 18-19, 2005 - Minneapolis,MN Sept 8-9, 2005- Miami, FL Sept 22-23, 2005 - Philadelphia,PA Oct 6-7, 2005 - Albuquerque,NM Oct.20-21, 2005 - Cleveland,OH Oct.27-28, 2005 - LosAngeles,CA Nov 3-4, 2005 - Dallas, TX Dec 1-2, 2005 - San Francisco,CA Jan.26-27, 2006 - Honolulu,Hi Feb.2-3, 2006 - Atlanta,GA Mar 2-3, 2006- Denver,CO Mar 16-17, 2006- Indianapolis,IN Mar.30-31, 2006 - Raleigh,NC Apr.6-7, 2006 - Salt LakeCity,UT Apr.27-28, 2006 - Omaha,NE May 4-5, 2006 - LosAngeles, CA May 11-12, 2006 - Toronto,ON June 8-9, 2006 - Nashville,TN July22-23, 2006 - LasVegas, NV * The2-dayTESACoordinator Certification Training preparesyou to teachTESAprogramsto certificated staffatschoolsites * The$300 registration fee includesthe2-daytraining, TESACoordinator Manual,Awareness Packet,instructional video, interaction wall chart,and refreshments Discount foron-siteTESA Coordinator Trainings To request a registrationformor if you would likeadditional informationregarding the TESAor PESAprograms,please call (800) 566-6651 M;A Look for the PESAtrainingschedule on page 593 of this issue O E-mail:tesa.pesa~lacoe.edu Website:http://streamer.lacoe.edu/TESA otion in dent achievement and segregation teract in a negatively correlated fashion.'2 John Ogbu writes of a "cultural ecological" model inwhich minor ity students perceive ongoing patterns of discrimina tionandprejudicewhen comparingtheirexperiences to those of theirmajority peers,which then inhibits aca demic achievement.' Other authors have identifiedwhat they call an "oppositional culture,"which ismost preva lent in schoolswith smaller percentages of minority stu dents.'4 In such situations, minority student engage ment, participation, and achievement all suffer, and any achievementgap isexacerbated Finally, although teacher quality has enjoyed atten tion in the literature of the achievement gap, research ersdifferregarding itssignificance Forexample, Harold Wenglinsky, Jonah Rockoff, and Peter Denner and his colleagues all find a strong relationship between teach er quality, defined in terms of training or experience, and student learning.'"Yet Theodore Eisenberg indi cates that advanced subject-matter knowledge on the part of teachers does not translate into higher levels of student learning."1Considering the emphasis given to this factor in the No Child Left Behind Act and the importance this law attaches to closing the achieve ment gap, teacher quality is a particularly salient vari able WHAT WE STUDIED Based on our experiences in schools and our review of the educational research on the achievement gap, we hypothesized that academic achievement for Latino students would be based on factors similar to those that we hypothesizedthat affectall students.Furthermore, within-groupdifferencesin theLatino studentpopu lation would be much larger than the differences be tween white students and Latino students In order to test these suppositions, we examined data in the Na tionalEducationalLongitudinalStudy (NELS:88) NELS:88 is a comprehensivestudyauthorizedby theU.S Congress and conducted by theNational Cen ter forEducation Statistics It is a seriesof cohort studies of American students that began in 1988 with eighth gradestudentswho were followed into high school, postsecondary education, and thework force Follow up studies were done in 1990, 1992, 1994, and 2000 NELS:88 uses both questionnaire data and test data APRIL2005 This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions 601 foreachstudent.Inaddition,NELS:88 involvesques tionnairesfor the schoolprincipal,for two teachers, and forparents The sampleforour studywas drawnfromthe12th gradefollow-upstudy.To determinebothwithin- and between-groupdifferences,we calculatedeffects for whites, blacks,andLatinos.Thus our sampleincluded dataon 15,618students:2,170 Latinos, 1,660blacks, and 11,788whites.We chose to lookatmathematics While includ as themeasureof studentachievement ingothersubjects would provideamore completepic of thedatabasere tureof achievement,idiosyncrasies measuretomath quiredthatwe limitour achievement ematicsonly The NELS:88 databasecontainsmany hundredsof possiblevariablestodrawupon andcombine inexam iningfactorsthatcouldbe importantto studentachieve ment Thus itwas necessaryforus to identifyfactors that,basedon other research,havebeen shown to in Our reviewledus fluencestudent learninggenerally to investigatethe relationshipbetweenmathematics achievementand the followingvariables: * socioeconomicstatus, * languageother thanEnglish regularlyspokenat home, *participation at any time in an ESL program, * time spenton homework, * classsize, *numberofminority studentsin theclass, *numberof unitsof algebrataken, *numberof undergraduate coursestakenby theteach most frequently, er in thesubjecthe or she teaches *numberof graduatecoursestakenby the teacherin most frequently, thesubjecthe or she teaches *familycomposition(i.e.,twoparentsin thehome), * levelof parent involvement,and *urbanicity(i.e.,urban,suburban,or rural) who spokeEnglishathome,who dents.Latinostudents had never been enrolled in an ESL class,who came from intactfamilies,andwho spentmore timeon homework than demonstrated higherlevelsof academicachievement Latinostudents who did not sharethesecharacteristics These relationships were similarforwhite students to Turning between-groupdifferences,none of the variableswe consideredrevealeda statisticallysignifi cantgap betweenwhites andLatinos.Socioeconomic background,experiencein anESL class,unitsof alge bra, and level of parent involvement had a similar im pact on the achievement of both white and Latino stu dents However, hours of homework were not a good predictorof studentachievementforLatino students, while thisvariablewas a good predictorforboth black studentsandwhite students.Finally,while thediffer enceswe foundbetweenwhite studentsandLatinostu dentswerenot significant,thedifferences betweenblack studentsandwhite studentsandbetweenblackstudents andLatino studentswere significant While these findings are important, a caveat isworth bearing inmind They not indicate a simple, straight inwhich increases or decreasesinone linerelationship, variableaffectstudentachievementin directpropor tion.Nevertheless,our findingsdo clearlyindicatethat familyincome,thenumberof parentsin thehome, the numberof algebraunits taken,the levelof parent in volvement,and the levelof English-languageskillsare significantpredictorsof academicachievementforLa on thesefactors tinostudents Moreover,thedifferences among Latinos are greater than those between Latinos and whites And many of the same factors exert a simi lareffecton achievementforwhite students Our research also indicates that the achievement gap isnotmonolithic Instead,it isa richlytextured,com Our findingsunderscore plex,andnuancedframework the need to disaggregate student data into many com binations of subsets in order to understand the dynam FINDINGSAND RECOMMENDATIONS ic relationshipsthatexistwithin andbetweengroups Our analysisdiscoveredthat the "achievement gap" gaps"thatexistbothbetween reallyconsistsof "multiple andwithin groups.Socioeconomicstatusand partici pation inESLwere themost significantfactorsforall groupsof students.For themost part,white andLatino weremirror imagesof eachother, studentachievement and eachwas affectedsimilarlyby eachof the factors we examined But thiswas not thecaseforblackstudents in thesample.For example,Latinoandwhite student 602 achievementreflectedsimilardifferencesbasedon ur banicity, but the same did not hold true for black stu The practice of lumping together data from all stu dents of colorand even data from divisions within a single group is amistake that isbound to produce poor policy choicesand poor educationalpractices Data-driven decision making is gaining popularity with educatorsand policy leaders Thismethodology holdsmuch promise tohelp us betterunderstandthe needsof students,to evaluatethe effectivenessof our educationprograms,and to informparentsand key PHI DELTA KAPPAN This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions stakeholders about our schools However, we must re main wary of the allure of numbers and conscious of the destructiveness of flawed research.We must be care fill about jumping to conclusions simply becausewe find a number that implies a difference between groups of students We must alwaysinvestigate theunderlyingfac tors that contribute to the average score for any group of students As our researchdemonstrates, taking action based on limited data and analysis isprofessionally ir responsible We have an ethical obligation to be thor ough in our understanding of the phenomena we study in our schools Finally,what is evident from our investigation is that both school-based factors and home-based factors are important to the success of every child, regardless of racialor ethnicdifferences.Schoolpolicyandpractice must be founded not on perceptions of group stereo types, but rather on knowledge about each student's needs and strengths Thus the voices of parents, teach ers, and students must be included when practitioners and policy makers seek to design betterways to serve stu dents The national dialogue about the achievement gap has the potential to help policy makers and educat ors find ways to better serve Latino and other minority students But ifwe are to create such constructive poli cies, research and practice must be based on thought ful reflection about what we know rather thanwhat we assume 1.Gill Griffin, "ColorChange: African-Americans and Latinos Reassess Their Relationships inWake of Changing Demographics," San Diego Union-Tribune, 23 February 2003, pp 1-2 Hersholt C.Waxman, Shwu-yong L.Huang, andYolandaN Padron, "Motivation and Learning Environment Differences Between Resilient and Nonresilient Latino Middle School Students," Hispanic Journal of Behavioral Sciences,vol 19, 1997, pp 137-56 FenwickW English, "On the Intractabilityof theAchievement Gap inUrban Schools and theDiscursive Practice of Continuing Racial Dis crimination," Education and Urban Society,vol 34, 2002, pp 298-311; and Alejandro Portes and Ruben G Rumbaut, ImmigrantAmerica: A Portrait (Berkeley:University of California Press, 1990) Sampson L Blair andMarilou C Legazpi, "Racial/EthnicDifference in High School Students'Academic Performance:Understanding the Inter weave of Social Class and Ethnicity in the Family Context," Journal of ComparativeFamily Studies, vol 30, 1999, pp 539-55; and Alejandro Portes and Dag McLeod, "Educational Progress of Children of Immi grants:The Roles of Class, Ethnicity, and School Context," Sociologyof Education, vol 69, 1996, pp 255-75 SharonAnne O'Conner and KathleenMiranda, "The LinkagesAmong Family Structure,Self-Concept, Effort, and Performanceon Mathematics Achievement of American High School Students by Race," American SecondaryEducation, vol 31, 2002, pp 72-95; and Sammis B.White, "Socioeconomic Status andAchievement Revisited," Urban Education, vol 28, 1993, pp 328-43 TraceyDerwing et al., "Some FactorsThat Affect the Success of ESL High School Students," CanadianModern Lan guage Review, vol 5, 1999, pp 532-47 KEEPING BLACK BOYS KengtpgA BlackDoys |OUTOF Education SPECIAL EDUCATION /5 a !byDr JOianza Kunijuifu * 180)pages $15.95 (Paper)* ISBN:09749(XX)28 David P Dolson, Add $3.95 forpostage "The Effects of Span ish Home Language Is there a relationship between Ritalin and co Use on the Scholastic caine? Between illiteracy and special education? Why are males placed in special education more Performance of His than females? panic Pupils," Jour What percent are mainstreamed back to the regu nal of Multilingual lar class? andMulticultural De To order call: 1-800-552-1991 E-mail us at:AAl@AfricanAmerican1mages.com velopment,vol 6, 1985, Visit our website at: http:Hlwww.AfricanAmericanlmages.com pp 135-55; and Ana AFRICAN AMERICAN IMAGES Celia Zentella, "La 1909 West 95th Street * Chicago, Illinois 60643 tino Youth at Home, in Their Communi ties, and in School: The Language Link," Education and Urban Society, vol 30, 1997, pp 122-30 David Adams et al., "PredictingtheAcademic Achievement of Puerto Rican andMexican-American Ninth-Grade Students," Urban Review, vol 26, 1994, pp 1-14; and Raymond Buriel et al., "The Relationship of Language Brokering toAcademic Performance, Biculturalism, and Self-EfficacyAmong Latino Adolescents," Hispanic Journal of Behavior al Sciences,vol 20, 1998, pp 283-96 9.William Jeynes, "AMeta-analysis:The Effects of Parental Involvement onMinority Children's Academic Achievement," Education and Urban Society, vol 35, 2003, pp 202-18 10 Charles V Izzo et al., "ALongitudinal Assessment of Teacher Per ceptions of Parent Involvement inChildren's Education and School Per formance,"American Journal of Community Psychology,vol 27, 1999, pp 817-39 11 JamesR Valadez, "The Influence of Social Capital onMathematics Course Selection by Latino High School Students," HispanicJournal of Behavioral Sciences,vol 24, 2002, pp 319-39 12 Gary J Roberts, "The Effect of Achievement on Student Friend ships inDesegregatedSchools,"Equityand Choice,vol 5, 1989, pp 31-36; RussellW Rumberger and J.DouglasWillms, "The Impactof Racial and Ethnic Segregationon theAchievement Gap inCalifornia High Schools," Educational Evaluation and PolicyAnalysis, vol 14, 1992, pp 377-96; and Richard R Valencia, "Inequalities and the Schooling of Minority Students in Texas:Historical and Contemporary Conditions," Hispanic Journal of Behavioral Sciences,vol 22, 2000, pp 445-59 13 John U Ogbu, Minority Education and Caste: TheAmerican System in Cross-CulturalPerspective(New York:Academic Press, 1978) 14 JeremyD Finn and Kristin E Voelkl, "School Characteristics Re lated toStudent Engagement,"Journal ofNegroEducation, vol 62, 1993, pp 249-68; andTomas D Rodriguez, "Oppositional Culture andAca demic PerformanceAmong Children of Immigrants in theU.S.," Race, Ethnicity, and Education, vol 5, 2002, pp 199-216 15.HaroldWenglinsky, "How SchoolsMatter: The Link Between Teach er Classroom Practices and Student Academic Performance,"Education PolicyAnalysisArchives, vol 10, 2002, available at http://epaa.asu.edu/ epaa/v1On 12; Jonah Rockoff, "The Impact of Individual Teachers on Student Achievement: Evidence from Panel Data," abstract available at and http://econwpa.wustl.edu/eprints/pe/papers/0304/0304002.abs; Peter R Denner et al., "Connecting Performance to Student Achieve ment: A Generalization and Validity Study of the Renaissance Teacher Work Samples Assessment," paper presented at the annual meeting of theAssociation of Teacher Educators, Jacksonville, Fla., 2003 16 Theodore A Eisenberg, "BegleRevisited: Teacher Knowledge and Student Achievement inAlgebra," Journalfor Research inMathematics K Education, vol 8, 1997, pp 216-22 APRIL2005 This content downloaded from 98.176.112.184 on Tue, 10 Dec 2013 12:39:57 PM All use subject to JSTOR Terms and Conditions 603 ...A SpecialSectionon theAchievementGap Challenging Assumptions The national dialogue about the achievement gap can help policy makers and educators findways... coursestakenby theteach most frequently, er in thesubjecthe or she teaches *numberof graduatecoursestakenby the teacherin most frequently, thesubjecthe or she teaches *familycomposition(i.e.,twoparentsin thehome),... comparedtootherfactorsthatmost affectstudentlearn ing.Furthermore, we havefoundthatthe "achievement gap"betweenLatinoandwhite students may be a "phan tomgap"derivedfromthepracticeof lumpingallnon 600 the