The Mentor Phenomenon and the Social Organization of Teaching
Author(s): Judith Warren Little
Source: Review of Research in Education, Vol. 16 (1990), pp. 297-351
Published by: American Educational Research Association
Stable URL: http://www.jstor.org/stable/1167355 .
Accessed: 18/11/2013 14:33
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.
.
American Educational Research Association is collaborating with JSTOR to digitize, preserve and extend
access to Review of Research in Education.
http://www.jstor.org
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Chapter6
TheMentorPhenomenonand
the SocialOrganization
of Teaching
JUDITH
WARREN
LITTLE
of California,
Berkeley
University
Policymakersand educational leaders have thrust mentoringinto the
vocabularyof school reformas part of a mission to rewardand retaincapable teacherswhile obligatingthose teachers, implicitly or explicitly,to
contribute to the improvementof schools and the quality of the teacher
workforce. Formuch of the past decade,the term mentorhas been prominently associated with proposed shifts in teachers'professionalrelationships and with altered teacher roles in schools and school districts.
Mentoring is a principal component of state-initiatedteacher incentive
programs(Hart, 1989; Neufeld, 1986; Wagner,1985), university-based
teacher preparationprograms(Huling-Austin,1988), and local programs
of teacher induction and professional development (Stoddart, 1989).
Proponentsof mentoringargueits meritson the basis of a "mutualbenefits" model (Zey, 1984). By this argument,investingsome teacherswith
the specialtitles, resources,and obligationsof mentorshipwill more readily assure various individual and institutional benefits. The mentors
themselves will receive public acknowledgementof their accumulated
knowledge,skill, and judgment. Novice teacherswill receive supportthat
mediates the difficulties of the first years of teaching. Careeropportunities in the occupation will be enriched. And schools, restructuredto accommodate new teacher leadership roles, will expand their capacity to
serve students and to adapt to societal demands.
Rhetoric and action have nonethelessoutpaced both conceptualdevelopment and empiricalwarrant.Indeed, a certain "manicoptimism"prevails (Elmore, 1989). Relative to the amount of pragmatic activity,
My thanks to Ann Weaver Hart, who served as advisory editor on this chapter, to
Nathalie Gehrkeand MarkSmylie for their commentsand advice, and to SusanSatherfor
her assistance in searchingthe literature.
297
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
298
Reviewof Researchin Education,16
rehowever,the volumeof empiricalinquiryis small.Ina comprehensive
view publishedin 1983,Merriamfoundscantmaterialon mentoringin
academicsettingsand madevirtuallyno mentionof mentoringfor purposesof teacherinduction,professionaldevelopment,or careeradvancement amongpublicschoolteachers(see also Galvez-Hjornevik,
1986).
That is, as recentlyas 1983 there was no distinctline of researchon
mentoringin education-and certainlynoneon mentoringin K-12teaching. Nonetheless,the scaleof policyinterestand practicalexperimentaof considerable
tion since 1983suggestsa naturalopportunity
magnitude
of thesespecializedteacherleadto examinethenatureandconsequences
ershiproles.This reviewoffersthe beginningsof a rathersubstantialresearchagenda.
In this essayI examinementorshipas a structuralandculturalfeature
of schoolsand the teachingoccupation.The focusis the organizational
and occupationalsignificanceof mentoringamongpracticingteachers,
withemphasison issuesrelatedto schoolorganization,
occupationalsocialization,andthe structureof theteachingcareer.I beginwitha conundrum:Howto accountforthe rapidlyescalatingpopularityof mentoring
in an occupationthatprovidesfewprecedentsfor formalandlegitimate
leadershipby teacherson mattersof professionalpractice.
Curiosityaboutthe originsof mentoringamongteachersstemsin part
fromthe culturallegacyof the mentor-prot6g6
relationship.In the classical tale fromwhichthe termis derived,the departingOdysseusentrusts
Mentorwiththe careandguidanceof his son Telemachus.
The relationshiprequiredof Mentora fullmeasureof wisdom,integrity,andpersonal
investment.It requiredthat Telemachus,as prot6g6,honorthe differences in maturityand circumstancethat separatedthem. The relationshipbetweenmentorandprot6g6wasprofoundlypersonalandmutually
respectful,even thoughit was essentiallyasymmetrical.It exactedhigh
demandsand yieldedsubstantialrewards.
The contemporary
treatmentof mentor-prot6g6
relationsis substanto the riseof
tiallymorenarrow.Clawson(1980)tracesits transformation
whena "morepractical,less comprehensive
apprenticeships,
conceptof
mentors"emerged,linking mentors primarilywith career and less
broadlywith adultmaturation(p. 146;but see Kram,1983;Levinson,
Darrow,Klein,Levinson,& McKee,1978).In the mentorprogramsthat
haveswepteducation,the demandson the mentor'scompetence,character,andcommitmentareoftenmuted,reducedto formaleligibilitycriteria and specific job descriptions.Clawson argues, however,that
and mutualityremainthe essenceof the role. Like
comprehensiveness
othertheorists,Clawsonconfersmentorstatusonlyon thosepersonswho
fulfillseveralpotentialroles(seealsoAnderson&Shannon,1988;Schein,
1978;Zey, 1984).Mentorsworthyof the nameserveas teacher,sponsor,
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
299
rolemodel,confidant,andmore.First-person
accountsdetailthe wayin
whichmentorshipsunfold,touchingpersons'lives as wellas theirwork
andachievingtheireffectsout of the reciprocalregardin whichmentors
and prot6g6shold one another(Parkay,1988).
Even narrowedto occupationalsocialization,then, the concept of
mentorshippromisesa greatdeal. To dignifyteachers'responsibilities
is to signifyinescapably
thementor'sspecialcawiththetitleof "mentor"
pacitiesand to invokea specialrelationbetweenthe mentorand other
relationis best
teachers.In Gehrke's(1988b)analysis,thementor-prot6g6
economythat is funinterpretedfromthe perspectiveof a gift-exchange
damentallyincompatiblewith narrowlydefined,utilitarianinteractions
un(a marketeconomy).By this view,mentorshipis mostappropriately
derstood"withina systemof gift exchangewherecostscannotbe calculated;wherelaboris not measuredby hoursat a specifiedratebutby an
interiorclock;andwhereworthis judgedby the individualin termsof its
personaleffect,andbythegroupin termsof its supportof unity"(p. 193).
actions
Somecriticsvoiceskepticismaboutlegislativeor bureaucratic
benton convertingthe fundamentally
personal,informal,andintenserelationsof mentoringto formalarrangements
(Clawson,1980;Gehrke,
1988a, 1988b;Kram, 1986;Zey, 1984).The broaderculturallegacyof
mentoringpresentsa modelof humanrelationshipthatdoes not lend itself well to policy intervention.Commonto instancesof "significant
mentorship"
(Hardcastle,1988)area breadth,mutuality,andinformality
difficultto achievewithinthe confinesof bureaucratic
(see
arrangements
andbureaucratic
enClawson,1980;Schein,1978).Heavilystandardized
vironments,accordingto Zey (1984), do not supportmentoringwell.
of teachOthercritics,alertto the particularhistoryand circumstances
ing,arguethatmentorrolesarealsolargelyincompatiblewithprevailing
values,norms,and structuresof the occupation(Smylie,1989;see also
Griffin,1985).Formalinitiativesto developandsupportmentorrolesare
thusin somerespectsan odd enterprise.To resolvethe riddleof howformalmentoringhascometo passin educationrequiresin partthatwe understandwhattranspireswhenfundamentally
personalrelationsbecome
the objectof formalorganization.Also,it requiresthatwe attendto the
apparentdisjuncturebetweenthe egalitarianand individualistictraditions of teachingand the specialstatusimpliedby the title of mentor.
Thesepuzzlesoccupya largeproportionof the researchon mentoring.
Inthe firstandlargestsectionof thispaper,therefore,I relyon implementationstudiesthatchroniclethe emergenceof the mentorroleandthe attemptedreconciliationof presentpurposeswith inheritedtraditions.In
dilemmasto
the remainingsectionsI employthe majorimplementation
interpretthe practiceof mentoringfromthe perspectiveof its two most
commonlystatedpurposes:teacherinductionand careerenhancement.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
300
Reviewof Researchin Education,16
These sections are organized respectivelyby two aspects of the mentor
phenomenon that are prominent in literaturespanningeducation, business, and other professions:(a) conceptions orientedto helping,with emphasis on emotional support,skill development,and workperformance,
and (b) conceptionsorientedto advancement,with emphasison enhanced
career opportunityand reward.
THEEMERGENCE
OF FORMALMENTORROLES
In principle, mentor roles satisfy three related policy problems.
Mentoring responds first to problems in the occupational induction of
teachers. Experienced teachers acknowledged for their own record of
classroomaccomplishmentare invited to pass their knowledgeon to novices. Second, "the mantle of mentorship"(Lemberger,1989) purportedly
createsan incentive for teacherretentionand commitmentby conferring
public recognitionand rewardon the most accomplishedteachers. Last,
the concentrationof discretionaryresourceson mentorssignalsa shifting
strategyfor local professionaldevelopmentand programinnovation;districts employ mentors as staff developmentand curriculumspecialistsin
pursuitof broad school or district priorities. In all of these policy rationales, the implicit logic is that the concentration of resources on a relatively small proportion of teachers will yield benefits for the larger
teacher population and for the institutions that employ them.
Implementation studies have clustered around the major local and
state initiatives that exemplify these three policy interests. The various
initiatives are similar in their origins, tending to arise as policy-level responses to workforce and workplaceissues. They are alike, too, in devoting substantialinstitutional resourcesto an enterpriseby which teachers
themselves bolster the capacities and commitments of the teacher work
force. The initiatives nonetheless vary in the focus and clarityof the purposes and strategiesthey pursue,in the degreeto which the intendedroles
and functions depart from traditions of autonomy and equal status
among teachers,and thus in the burdenof changethey present.The various implementation studies reflect these differences as well.
Three examples illustratethe initiatives on which researchhas focused
and the directions researchers have pursued. In California's Mentor
Teacher Program,legislators placed rewardand recognition for experienced teachers foremost. The emphasis on careerincentives for individual teachersis reflectedin the flow of dollars:Twothirdsof the program's
resourcesgo directlyinto the hands of the mentorteachersin the form of
moderatelylarge stipends ($4,000 per year). Althoughthe legislatorsexplicitly anticipated that mentors would in turn contributeto teacher induction, professional development, and leadership in curriculum and
instructional improvement, districts were granted a wide range of latitude
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
301
to shapea programresponsiveboth to local interestsand local constraints
(Wagner,1985). Far West Laboratoryfor EducationalResearchand Development undertooknine case studies and a statewidesurveyas partof a
2-year investigationof the first stages of programimplementation.Case
studies supplied the main insights (Bird, St. Clair, Shulman, & Little,
1984;Hanson, Shulman,& Bird, 1985;Shulman,Hanson,& King, 1985),
which in turn weretested againsta broaderarrayof policies and practices
in a surveysampleof 291 districts(Bird, 1986;Bird& Alspaugh,1986).
In Connecticut,mentor roles were introducedas an essential featureof
the state's new proceduresfor teacher certification and induction. These
purposes were primary, career incentives were secondary (Allen &
Pecheone, 1989). Mentors are expected to help preparebeginningteachers to satisfy the state's criteriafor certification in 15 competency areas.
In this regard,the Connecticutprogramis similarto othersin which mentors' work is linked closely to local or state evaluation standards
(Huffman& Leak, 1986; Stoddart, 1989). The ConnecticutState Department of Educationhas supportedprogramhistories and programevaluations of the Connecticut Beginning Educator Support and Training
(BEST) program since its inception in 1985-86 (Allen, 1989; Allen &
Pecheone, 1989; Martin, 1987; Neufeld, 1986). Conceptually,the state's
initiative is premised on a "screensand magnets"strategy(Sykes, 1983).
Each of the studies, therefore,traces the developmentof the mentor role
as part of a system in which more stringentassessmentof teaching(better
screening) is backed by more rigorousand frequentsupport of teachers
(more compellingmagnets).The various studies follow district-levelpilot
programs through a steady expansion in number from 5 to 25 and
throughthe less steady,sometimes turbulent,searchfor sharedgoals and
workablestrategies.In the most recent phase, evaluations of pilot efforts
in 25 of the state's 166 districtsconcentrateon the short-termeffects associated with certain instrumentalaspects of programimplementation,especiallythe effort to achieve a subject-gradelevel match betweenmentors
and new teachers (Allen & Pecheone, 1989).
In still other sites, career incentive programs (Hart, 1989; Hart &
Murphy,1989b;Smylie & Denny, 1989) or school improvementconsortia
(Wasley,1989) havepromotedspecializedleadershippositions that beara
close resemblanceto mentor roles, thoughtheir titles differ.Rarelydo the
evolving roles entail formal authorityfor personnelor programmatters,
althoughthey do engagesome teachersin the supervision,assistance,and
instruction of others. These experiments most closely approximatethe
"school restructuring"applications of mentorship, linked not to specific
provisions for teachercertificationor induction but to broadlystated aspirations for teacher leadership and career enhancement. Local career
ladder experiments in Utah, also spurred by the screens and magnets
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
302
Reviewof Researchin Education,16
into school-leveldefinitionsof
logic,havebeenthe site of investigations
leadershiproles,withspecialemphasison the fit betweenteacherleadershipandschoolgoals(Hart& Murphy,1989b)andon the "rolepolitics"
surroundingdevelopmentof teacherleaderand teacherspecialistpositions(Hart,1989).A similarinquiryhasfolloweda localcareerenhancement venturein which teacherleadershippositions were introduced
througha side letteragreementto a locallynegotiatedteachercontract
(Smylie& Denny,1989).
Enrichedby advancesin the studyof innovations(McLaughlin,
1987)
andby theoriesof workredesign(Hackman&Oldham,1980;Nicholson,
studieshavechartedtheprogressof mentorinitia1984),implementation
tivesby closelyattendingto localcontextsand organizational
dynamics.
All the implementationstudies incorporatein-depthinterviews;each
adds,to varyingdegrees,otherelementsof a casestudyapproach.Interviews with district administratorsestablishthe place of the mentor
phenomenonin relationto otherdistrictpriorities,goals,and history.
District-levelcoordinators,principals,and mentorsthemselveslink the
to similar
programto local schooland districthistory,and particularly
leadershiprolesthat hadbeenwellor poorlyacceptedby teachersin the
fill in the details
and districtadministrators
past.Unionrepresentatives
of localnegotiationoverthe formandcontentof mentors'workandthe
conditionsof theirselection.Mentorsevaluatethe arrangements
madeto
preparethemfortheirnewrolesandto supportthemin theirwork.The
mentorsand the principalsand teacherswithwhomtheyworkdescribe
andassesstheactualworktheyhavedonein theircapacityas mentor,and
the time and otherresourcesavailableto do it. Principalsweighthe adof the arrangement
fromthe perspectiveof
vantagesand disadvantages
the individualschooland describetheirown influence(or lackof it) in
shapingthe mentors'generalrole or specificresponsibilities.
Thesestudieshavein commonthattheyconstructthe implementation
problemnot onlyas the pursuitof broadpolicygoalsandthe implementationof a discreteprogram,but also as the redefinitionof institutional
roles,professionalrelationships,and the workof teaching.In some respects,those who wouldimplementmentorrolesare confrontedwith a
two-partchallenge:to introduceclassroomteachersto a rolewithwhich
theyareunfamiliar;andto introducethe roleitselfto an institutionand
occupationin whichit has few precedents(Bird& Little, 1985;Little,
1988;Smylie & Denny, 1989).Consistentwith this perspective,Hart
(1989)castherstudyof school-levelteacherleaderandteacherspecialist
positionsin Utah as a case in role innovationand workredesign.The
"substantialdiscretion"attachedto new teacherleaderpositionsled
Smylieand Denny(1989)to frametheirown studyaroundquestionsof
role definition and role evolution. Similar issues surrounding role defini-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
303
tion guidedthe earliestinquiriesinto the Connecticutpilot programs
(Neufeld,1986),andpersistentissuesof roleambiguityandroleconflict
surfacein morerecentwork(Allen,1989).FarWestLaboratory
described
its comprehensive
studyof the CaliforniaMentorTeacherProgramthis
way:
The [mentorprogram]
maybe describedas an effortto retainskillfulteachersandto improve teachingby promotingdirect,rigorous,and consequentialactivitiesand relationships
The[studies]askedwhetherandhowdistricteffortsto
betweenmentorsandotherteachers.
implementthe mentorprogrampromotedthoseactivitiesandrelations.(Shulmanet al.,
1985,p. 2)
Bowingto ConservativePrecedent
A singledominantthemeemergesfromthe implementationstudies:
Mentorinitiativesencounterconsistentpressureto accommodatethe individualisticand egalitariantraditionsof teachingand to discountthe
statusdistinctionsimpliedbythementortitle.Attheschoollevel,wefind
few casesin whichthe mentorrolesignalsa reorganization
of authority
relationsor an increasein the school'scollectiveinfluenceon practicesof
teaching(for one exampleof sucha case, see Hart, 1989).
Certainconditionsof implementation
constrainor enablelocalactors,
movingthemtowarda moreambitiousor a less ambitiousconceptionof
the mentorrole.Amongthemarethe paceat whichimplementation
proceeds,the sheeropportunityfor workthat is describedas "mentoring,"
andtheprecedentsthatshapeexpectationsformentors'performance.
On
thewhole,theseconditionshavefavorednarrowdefinitionsof thementor
role and conservativesolutionsto implementationproblems.
Pace of Implementation
Mentorprogramshaveproliferatedrapidlyoverthe pastdecade.The
magnitudeof changeimpliedbythementortitleinvitesa pacethatis slow
enough to achieve properly "integrativeagreements"(Pruitt &
Carnevale,1982),but briskenoughto sustainmomentum.The persistence of problemsrelatedto role definitionin the Connecticutsites suggeststhat a slowpacealone(3 yearsof "pilot"activity)does not ensure
that proponentsof the innovationwill grapplesuccessfullywith established practice(Allen& Pecheone,1989;Martin,1987).A rapidpace,
however,coupledwith highpublicvisibility,almostcertainlyguarantees
thatdistrictswillsettleon conservativesolutionsto the predictableproblems that arisewhena proposedinnovationrunscounterto established
normsand structures.
California'sMentorTeacherProgramillustratesthe problemsassociatedwithrapidstartsanda fastpacein theearlystagesof a complexinno-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
304
Reviewof Researchin Education,16
vation.Pressedto movequicklyfollowingpassageof the state'somnibus
reformbill, teachersand administratorsin many Californiadistricts
achievedthe fit betweenpolicyintentandlocalcontextthatBermanand
McLaughlin(1978) describeas "mutualadaptation"by compromising
certainprincipal(butcontroversial)
tenetsof the legislativeintentunderlyingmentorships.Californialauncheda precipitousscheduleof implementationin its firststages,urgingdistrictsto adoptthe programin the
closingmonthsof the 1983-84schoolyear(Bird,1986).Althoughparticipationin the Californiaprogramwas voluntaryand althoughlocaldistricts retainedsubstantialdiscretionin decidinga conceptionof the
mentorrole,the scaleof the fundingpropelleddistrictsto accepttheprothe mentorprogramconstituted
gram.By 1986-87,thebudgetsupporting
morethanone half of the state'scategoricalstaffdevelopmentfunding
(Littleet al., 1987).Districts'accessto state-controlled
fundingforprofessionaldevelopmentresourceswasthusdirectlylinkedto participation
in the mentorprogram.Meanwhile,a scheduleof implementation
linked
to the state'sfiscalyearpresseddistrictsto decidequicklywhatformthe
localprogramwouldassume.The resultwasa pervasiveeffortto define
the mentorrole withinthe boundariesof familiarrolesand functions.
Basedon ninecasestudiesanda surveyof 291 districts,Bird(1986)concludesthat"agooddealwaslost,andlittleor nothinggained,by hastein
the mentorprogram"
implementing
(p. 7). California's
experienceis mirrored elsewherein the implementationstudies. Hart and Murphy
(1989b),too, attributeconservativeprogramdesignsin Utahto the press
of time in the earlystagesof implementation:
"Becausethe time left by
the state ... between planningand implementationwas limited, job de-
scriptionsoftenweremodeledafterpreexistingspecialprojectsandunrecognizablefromconventionalpractice"(p. 15).In Connecticut,where
the state'steachercertificationlaw introducesa relativelystandardized
conceptionof the mentorrole,a 3-yearsequenceof piloteffortshasnot
relievedthe stateof manyof the sameimplementation
problemsfacedin
sites withless well-bounded
purposesand fewerprogramspecifications.
Externallyestablishedgoals,evenwhenbroadlyacceptedin principle,do
not appearto overcomethe holdexertedby long-standing
andtaken-forgrantedwaysof working.The slowerpacein the Connecticutsites,however, may have helped to forestall the kinds of agreementsthat
compromisekeyprinciplesandtherebycontributeto a patternof "vanishingeffects"(Malen& Hart, 1987).
Rapidstartsplacea premiumon reducingthetangleof competingpreferencesand countervailingpractices.Smoothstartsachievedthrough
large-scalecompromisebodeill forlong-termsuccess.Integrative
agreements,accordingto some theorists,requireboth tolerancefor conflict
and sufficientopportunityfor conflictresolution(Pruitt& Carnevale,
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
305
1982).Case studiesof innovationin 12 schooldistrictsled Huberman
and Miles(1984)to concludethat
smoothearlyusewasa badsign.Smoothlyimplementing
sitesseemedto getthatwaybyrethegradientof actualpracticechange.
ducingtheinitialscaleof theprojectandbylowering
This "downsizing"
butalso
got rid of mostheadachesduringthe initialimplementation
threwawaymostof thepotentialrewards;
theprojectoftenturnedintoa modest,sometimes
trivial,enterprise.
(p. 273)
Broadlyconceivedpolicyinitiatives,introducedrapidlyin thespiritof reform,holdout mentorrolesas one elementin a newconceptionof teachers'professionalrelations.But mentorinitiativesconstitutea directand
substantialchallengeto someof themostpowerfully
establishednormsof
teachingandto establishedauthorityrelationsin schools.On thatbasis
alone,argueMalenand Hart(1987),such initiativesare especiallysusceptibleto the problemof vanishingeffects.The problemis exacerbated
whenthe paceof implementation
outstripsthe humanand materialresourcesavailableto managethe change.
Opportunity
Thefit betweenthe rhetoricandtherealityof mentoringis in largepart
a functionof opportunity.
The implementation
studiescastthe question
of opportunityin twoways.Theyfirstconfrontthequestion,Towhomdo
the opportunitiesor obligationsof mentorshipfall?The formaldesignathengivesriseto the secondquestion:Whenandhow
tion of "mentors"
do mentorsconducttheirwork?
The formalizationof mentorrolesbringswith it institutionalcontrol
over selection,or the systematicstructuringof teachers'opportunityto
the criteriaandproassumeprofessionalleadership.Issuessurrounding
cess for selectionhaveconsumeda largeshareof the politicalandmateand haveoccupieda central
rial resourcesdevotedto implementation,
placein research.In one recentassessmentof the prospectsfor teacher
leadership,Little(1988)observedthat
themostvolatileissuein formalteacherleadership
initiativeshasbeenteacherselection....
Theselectionof leadershasbeencastbothas a technicalproblem(whataretheacceptable
andas a politicalproblem(whowillteachersacceptas leaders,if
criteriaforperformance?)
anyone?).(pp. 100-101)
Littleconcludesthatthe selectionproblemis an artifactof isolatedwork
in schools,a problemthat achievesits presentmagnitudeonly because
manyteachershaveno sensiblegroundson whichto grantor denysomeone the rightto leadthem.BirdandAlspaugh(1986)observe,"Onwhat
basisdo personswhoworkmostlyin isolationaccepta decisionthatsome
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
306
16
Reviewof Researchin Education,
of themarebetterpreparedforleadershipthanothers?"(pp. 53-54). Acceptableselectioncriteriaand processescomprisea largeshareof teachers' overalljudgmentsaboutmentorprograms,and a largeportionof
theircomplaintswhenthingsseemto go wrong.In interviewdataassembledfromteachersin onecareerladdersite,"discontentanderodedcommitmentto the districtemerged... whenteachersquestionedthequality
of selectionand discriminationassociatedwith the new roles"(Hart&
Murphy,1989b, p. 23). Selection criteria,processes,and outcomes
formedthreeof theeightcriteriathatRuskus(1988)invitedteachersand
mentorsto use in judgingthe overalleffectivenessof California'smentor
initiativein five districts.Acrossdistricts,the perceivedvalidityof selection "wasthe most importantdeterminantof perceivedprogrameffectiveness"(p. 199).The districtwiththe highestratingon selection(and
alsothe highestprogrameffectivenessrating)employeda two-stageselection processin whichextensivepaperscreeningand principals'ratings
werefollowedby interviewsandobservationsof the highestrankingcandidates.In thatdistrict,an objectiveratingformbasedon statedcriteria
resultedin the selectionof one thirdof the applicants.Bycontrast,a districtwithconsistentlylowratingson selection(andthelowestratingoneffectiveness)employeda more cursoryreview procedure,conducted
interviewswithall candidatesbutno observations,andwascriticizedby
some teachersas relyingon "subjectivefeelings"ratherthan "realevidence"(pp. 202-203).
To whatextentdo the formalselectionprocesses-whichmayinclude
formalapplications,peerand supervisorrecommendations,
interviews,
observations, simulations, or portfolios-capture the prospective
mentor'spersonaamongcolleagues,or reflectteachers'expectationsof a
mentor'sefforts?Teachers'complaints,recordedanecdotallythrough
case studyaccounts,suggestthat a selectionprocesscenteredon a small
sampleof teacher'spresentworkmaybe inadequateto assurethe breadth
anddepthof teacherexperienceandknowledgethatmaybe an essential
prerequisiteto leadershipon mattersof professionalpractice.Available
case studiesprovidefewexamplesof selectionprocessesin whichmultiIn most
ple linesof evidence(Peterson,1984)areassembledpersuasively.
of the sites describedby the implementation
studies,teachers'eligibility
for leadershippositionswas thoughtto be satisfiedprincipallyon the
basisof short-termclassroomobservationsortestimonybypeersandadministrators.Few assessedthe particularcombinationof classroombased expertiseand collegialinvolvementsthat presagesuccessin the
mentorrole.Teachersin one case studyproposedthatselectioncriteria
balancingteachers'classroomexpertiseandtheirabilityto workwithcolleagueswould be "morein keepingwith the meaningof a 'mentor"'
(Shulmanet al., 1985,p. 14).
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
307
Selectionissuesoccupya centralplacein the implementation
panoply;
the resolutionof thoseissuesaffectsopportunitybyaffectingindividuals'
accessto the positionand others'dispositiontowardthem.Nonetheless,
the obligationsof mentorshiparesatisfiedandits benefitsassuredonlyin
the actualexerciseof the role.Teachersjudgementorsby the expertise
thattheydemonstrateandby the efforttheyexpendafterbeingselected.
Crucialto teachers'acceptanceof the roleof mentor,then,is theirability
to confirmthe worthof individualmentorsin actualperformance.
Despitethe scrutinygivento the processby whichteachersareselected
to be mentors,a still greaterburdenof proofrestson the mentorwho,
onceselected,mustnowactuallymentor.Herethe issueis thecongruence
among formalselectionmechanisms,the actualdemandsof performance,andthe informalregardof colleagues.Selectionturnsoutto be less
an eventthana continuingprocessby whichmentorsearntheirtitleson
thejob. Throughmyriaddailyencounters,andthroughsubtleandnot-sosubtlegestures,teachersaffirmor rejectthe mentors'acclaimedstatus
(Bird,1986;Hart& Murphy,1989b).On the basisof vignettesof teacher
andmentorinteractions(Allen,1989;Shulman&Colbert,1987),onecan
concludethatthe closerthe mentorcomesto exertinginfluenceon other
teachers'work,the morestringentbecomethe demandson the mentor's
competenceand character.
A teacheris selectedas a mentorprincipallyon the basisof accomplishmentswithchildren;the teacheris subsequently
acceptedas a mentoron
with fellowteachersand administrators.
the basis of accomplishments
The demandson mentors'expertisearefrequentlyfargreaterthana prospectivementormightanticipateon the basisof selectioncriteriaalone.
Admittedly,mentorsappearmore sanguineaboutthe effectivenessof
serve.Teachersareless
theirworkthanarethe teacherstheypurportedly
inclinedto judgethementors'effortsto be sufficientlyinfluential.Teachers in five Californiadistrictsconsistentlyratedmentors'effectivenessin
mentors(Ruskus,1988).In
less glowingtermsthandid the participating
each of threeareasof interaction(effectivenesswith newteachers,with
teachers'mean
experiencedteachers,andin facilitatingcommunication)
ratingsof effectivenessweresignificantlybelowthose of mentorsthemselves(p. 156).Teacherswerealsoless willingto attributeimpacton studentprogressor teacherretentionto the effortsmadeby mentors;again,
the meanratingson thecross-district
sampleof teachersandmentorsdifThisfindingparallelsresultsof a casestudyof schoolferedsignificantly.
level instructionalsupportteams in which team membersratedtheir
directcontactswithteachersas morefrequentandmorepotentthandid
the teachers(Little& Long, 1985).One explanationaccountsplausibly
forthesecomparablepatternsin thetwostudies.Fromthementors'point
of view, even a few direct consultations or classroom visits constitute a
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
308
16
Reviewof Researchin Education,
schedule.Fromtheperspectiveof
highlevelof activityin anovercrowded
teachersat large,most of whomhavenot been toucheddirectlyby the
mentors'activities,the workis lessvisibleandlessvisiblyconsequential.
Allowingfor veryrealdifferencesin effortandcapacityamongmentors,
the fact remainsthat teachers'perceptionsof theireffortand effectiveto acquiredirectevidence
nessarelargelycontingenton theiropportunity
of theirwork.Whenasked,teachersdiscriminatefinelybetweenthosein
leadershippositionswho do muchand thosewho do little,thosewhose
workmakesa solidcontribution
andthosewhoseworkis "trite"or "frivolous"(Hart& Murphy,1989b).Ruskuscollects,butdoesnot report,data
thatwouldpermitherto distinguisheffectivenessratingsgivenby teacherswhoweredirectlyinvolvedwithmentorsfromthoseratingsbasedon
less immediatecontact.Suchmeasuresof mentors'effectiveness,sensitive to variationsin directinvolvementbetweenmentorsandteachers,
willprovidea morecrediblebasethanwe havenowforexplainingteachthe natureandexers'acceptanceof the mentorroleandfordetermining
tent of mentors'influenceon teachers.
Concernfortheperformance
aspectof theselectionandsubsequentacceptanceof mentorshas led statesand districtsto supplymentorswith
skill trainingor peer supportgroups(Bird& Little, 1985;Kent,1985;
King,1988).Bysuchtraining,mentorsaresometimeshelpedto makeexinstruction,and
plicitandaccessibletheirownknowledgeof curriculum,
classroommanagement.
Often,however,theyareaskedto adoptconcepts
and terminologyderivedfromclassroomresearch(Kent,1985)or from
state and local teacherevaluationguidelines(Allen& Pecheone,1989;
Huffman& Leak,1986).Communication
skills,consultationstrategies,
and classroomobservationtechniquesalso forma largeshareof most
trainingagendas(see, e.g., Brzoska,Jones,Mahaffy,Miller,& Mychals,
1987;Little& Nelson, 1989;Stateof ConnecticutDepartmentof Education, 1988).Indeed,evena cursoryreviewof suchguidesleadsone to
concludethat the processof mentoringtakesconsiderableprecedence
over its substance;training activities are heavily weightedtoward
relationsbetweenmentorsandteachersor
ensuringsmoothinterpersonal
administrators.
common
Specializedtrainingfor mentorshasbecomean increasingly
and prominentcomponentof role development(e.g., Thies-Sprinthall,
1986).The earliestprogramventuressparkedwidespreaddisputeabout
the needfororganizedtrainingandsupport.Opponentsof suchtraining
madethe casethatthe veryselectionof teachersas mentorswasintended
to signala highlevelof professionalcapacity,whereasadvocatesof training underscoredthe unfamiliardemandsof mentoringfor which the
classroomprovidedlittle or no preparation
(Bird& Little, 1985).In the
first2 yearsof California'sstate-supported
program,nearly40%of par-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
309
ticipating districts allocated no resourcesat all for postselection support
of mentors; many others relied on occasional workshops sponsored by
county offices of educationor on otherout-of-districtopportunities(Bird
& Alspaugh, 1986).
Structuredtraining and support appearsmore likely where mentoring
is linked to a single state or district policy goal, as in Connecticut'suse of
mentors for teachercertification (Allen & Pecheone, 1989), Los Angeles
Unified School District's assignmentof mentorsto new teachers(Little &
Nelson, 1989), and Toledo's involvement of experiencedteachersin the
evaluation and tenure of new teachers (Stoddart, 1989). At the school
level, organized training and support are more likely where administrators and teachers have forged a clear link between the mentor role and
school-levelgoals. In these instances,clearlydefined policy purposessurrounding the mentor role increase the availability of training, but also
dictate its content. By contrast, teachers are more likely to assume new
leadership roles without benefit of formal training and support where
those roles remain personalized, entrepreneurial,and less clearly connected to institutional priorities (Hart, 1989; Smylie & Denny, 1989).
Trainingis typicallya post hoc accommodation,followingthe awardof
mentor status. Only rarelydoes selection to mentor roles requirethat the
teacher first acquire experience in other mentor-likecapacities, such as
serving as a supervisor of student teachers. Nor do selection processes
typically assess prospective mentors' disposition toward sharing ideas
and materials,assisting others, or taking initiative with regardto professional practice (Allen, 1989; Smylie, 1989). These characteristicomissions from mentorselection routinesmay proveconsequential,bearingin
unanticipatedways on mentors'own perceptionsof their role and on the
expectations that others hold out for it.
Despite the apparentrangein availabilityand content of mentortraining, there are virtually no studies that trace the contributions made by
postselection training to the subsequentperformanceof the mentors, or
to their success in relationshipswith teachersor administrators.No studies compare mentors who receive trainingwith those who are left to their
own resources. Nor have there been any attempts to assess the relative
leverageto be gained by investinginstitutional resourcesin postselection
training versus various forms of preselectionpreparationof individuals,
groups, or organizations.
The performanceimperativesof mentorship renderthe second aspect
of opportunitycrucial: when and how mentors do their work. This performance aspect of opportunity is fundamentallyan issue of time, the
most highly valued and closely protected of teachers' resources.Those
who control mentor programs,whetherteachersor administrators,signal
the importancethey attributeto mentor roles by the amount of time they
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
310
Reviewof Researchin Education,16
allocate for mentors' work, by policies that govern when the work of
mentoringcan be done, and by the formaland informalexpectationsthat
define what work counts as mentoring. Thus, some districts reserve
mentoringfor time outside the school day,whereasothers arguethat the
benefits of mentoring requireteacher-to-teacherconsultation embedded
in the daily work of teaching. Some districts or schools insist that
mentoringentails direct one-to-oneworkwith individualteachers,in ana
out of the classroom;other sites are broadlypermissiveabout the nature
of the mentor's activities.
The time mentors spend tends to be treated as a proxy for their effort
and effectiveness. Teachersjudge the worth of mentors in part by the
amount of time they visibly devote to the workof mentoring.Visibility is
a crucialcomponent of the time equation. When mentors do the workof
mentoring directly with teachers (or in their immediate presence),they
enable teachers to judge the quantity and quality of their contributions.
Accordingto Hart and Murphy(1989b), "visibility of the work played a
key role in teacherassessmentsof the worthof the program"(p. 27). Visibility is diminished when mentors' work is reservedto time outside the
instructionalday,to settings outside the mentors' own school, or to tasks
far removed from the classroom (Bird, 1986). For example, the most
prominent object of mentors' attention during the first 2 years of
California'sMentor TeacherProgramwas curriculum.Mentorsin many
districtsworkedon individual curriculumprojects,largelyout of sight of
their colleagues. Judgingby district coordinators'estimates of mentors'
time (Bird & Alspaugh, 1986), mentors on averagespent more than 60%
of their time doing something other than workingwith fellow teachers.
On the basis of mentors' own accounts, this estimate may be conservative. Fewerthan one in five of the districtssurveyedrequiredthat mentors
consult with or assist other teachers. The California experience is not
unique. The 13 teacher leaders whose work was detailed by Smylie and
Denny (1989) exemplify the discrepancybetween aspirationsand actual
performance."Althoughvirtuallyall the leadersreportedthat they had interacted with and assisted other teachers, none of these activities was
ranked among those consuming most of the leaders' time. The leaders
have, therefore,spent most of their [time] engagedin activities that seem
at variance with their primaryconceptualizationsof their roles. As one
commented, 'There is much involvementat the district level. However,I
need to do more at the buildinglevel, more one-to-one conversationswith
teachers'" (p. 8).
A permissive stance towardthe actual substanceof mentors'work has
enabled districts and schools to secure short-runagreementsto implement a mentor program. Except in Connecticut, where mentoring is
linked to certification (Allen & Pecheone, 1989), teachers have success-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
311
fully sought assurancesthat teachers' essential autonomy would not be
jeopardized by mentors' intrusions into their classrooms.Such a permissive stance, however, tends to produce a low rate of direct teacher-toteacher involvement of the very sort needed to convince teachers that
mentors are fulfilling their obligations (Bird, 1986; Huffman & Leak,
1986; Little, Galagaran,& O'Neal, 1984). Mentors respond by seeking
ways to showcase their work to teachers. Sometimes they succeed in
broadeningtheir base of support;in other instances,their effortsto publicize their activities only intensify teachers'opposition. Hart and Murphy
(1989a) note that "praise was profuse" when teacher leaders tied their
workclearly and productivelyto improvementsin teachingand learning,
but that complaints were equally profuse when leaders wasted teacher
time in superficial activities inappropriatelymatched to teachers' interests or sophistication, or devoted their energies to short-termprojectsof
dubious value. Teacherswith strongacademic recordsand high performance ratingshad the highest aspirationsfor what might be accomplished
throughteacher leadershippositions, and were the most critical of shortcomings produced by programcompromises. These teachers were most'
approving of long-rangeassignments that "gave teachers the power to
function by marshallingthe talents of other teachersto achieve learning
by groups of students,"while they "ridiculedshort term, limited assignments" (p. 20). One teacher was "scathingin her criticism of trite, unnecessary tasks disconnected from outcomes" (p. 21). Visibility alone, it
appears, is not sufficient to win teachers' endorsements.
Whatever the benefits that follow when mentors engage in "close-tothe-classroom"consultation or other involvementswith teachers,the attendant compromises are not lost on the mentors or others with whom
they work. In the name of school improvement or career enhancement,
mentorship programsadd to the burdensof the school-site faculty by removing capable teachers from the classroom. Time spent by a mentor in
the classroom of a beginning teacher, for example, is time lost to the
mentor's own classroom. Teachers routinely devote less time to
mentoring during the school day than they are allotted by programresources (Allen & Pecheone, 1989; see also the data on underuse of allocated release time as a component of school improvement programsin
Berman& Gjelten, 1984). Releasetime budgetsintended as a supportfor
the programmay turn out to be a burden for mentors. Release time that
draws teachers away from primary classroom responsibilities underscores, perhaps ironically, the marginal status of mentoring activity by
placing teachers'work with fellow teachers in competition with the fundamental work of the classroom. To fulfill the obligations of mentoring,
mentors risk compromisingother valued institutional goals and increasing the strain on themselves as individuals. In one programevaluation,
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
312
Reviewof Researchin Education,16
the shortage of qualified substitutes, the additional planning time required to preparefor substitutes, and the loss in instructionaltime and
quality all led teachersto assess releasetime as an "impractical"form of
support for the program (Allen & Pecheone, 1989). By contrast, they
would have welcomed regularlyscheduled contact time duringthe salaried workday.In sum, the structureof time and task constrainsor enables
mentors' work with teachers.
Precedents
Leadership by teachers is not entirely without precedent. Available
models for leadershiproles rangefrom department-or grade-levelheadto
committee chair to specialized staff development and curriculumdevelopment roles. On the whole, however,teachershave few models for an assertive conception of the mentor role, models that legitimatethe kinds of
relationships implied by the term. Despite certain long-standingprecedents for formal leadership positions (Wasley, 1989), roles specifically
dedicated to interpersonalguidance on matters of professionalpractice
continue to representa substantialdeparturefrom organizationaland occupational tradition. Furthermore, districts or states rarely consider
whethernewly proposedroles are compatible with or in conflict with existing leadership opportunities (Hart & Murphy, 1989b).
Few of the available implementation studies explicitly confront local
leadershipprecedentsand their significance for newly introduced roles.
In six of FarWest Laboratory'snine case study districts, respondentsdescribed other roles with mentor-likefeatures.In one district, demonstration teachers and resourceteachers were admired for their assistance to
new teachers and their contributions to curriculum (Shulman et al.,
1985). These models of teacher-to-teacherexchangedisposed teachersfavorablytowardthe idea of mentorship.In anotherdistrict,a troublesome
precedentwas createdin the form of remediationteacherswho had been
widely viewed by teachersas servants of administration;supportfor the
mentor role was less readily secured (Hanson et al., 1985).
The various implementation studies all highlight the ways in which
mentoringis a departurefrom business as usual. Mentorroles turn out to
be an innovation of considerable complexity. Unlike new curricula or
pedagogicalmethods, this innovation is not subject to individual adoption at the level of the classroom;rather,it is a social relationthat requires
joint action andjoint acceptance.The "smallestunit"on whichthe fate of
implementationrests (McLaughlin,1987, p. 174) is thus not the individual teacher but the mentor-prot6g6pair. To stress the image of the
mentor-prot6g6pairis not to overlookthe fact that many designatedmentors or leaders work alone (e.g., to develop curriculum)or work with
groups of teachers in workshop-typesettings. It is, rather,to underscore
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
313
the fact that the term mentor inevitably implies the existence of a proteg6
whetherthat prot6g6is a specific individual (e.g., a 1st year teacher),or a
diffuse body of teachers (e.g., potential users of the mentor's curriculum
ideas). The success of mentorshipthus rests in part on the prot6g6s'willingnessto be mentored,whetherdirectlyor indirectly.Individuals'capacities, beliefs, and incentives do not accountsatisfactorilyfor local success.
The skills and intentions of mentorsare insufficientto sustaintheir interactions with teachers, which are approved or condemned within larger
circles of peers. Mentoring is irretrievablya social and organizational
phenomenon, and as such its utility as an organizationalresourceand a
career incentive is shaped by social interaction.
Nicholson's (1984) theory of work role transitions helps to highlight
some of the disjunctures between the demands of mentoring and the
likely perspectives and experiences of prospective mentors. The ordinary stress associated with role transitions is intensified when new roles
present radically different requirements from persons' experience, and
thus exert new demands on personal knowledge, skill, judgment, confidence, and initiative. The particularadjustment that individuals make
to their new roles can be traced, accordingto Nicholson, to their assessment of the differencesor similarities in workdemands, the dispositions
they have acquired through past socialization, and the arrangements
that now govern their work in a new position. However, Nicholson's
analysis stops short of anticipating certain fundamentalconditions that
bear on the specific case of mentoring among teachers. First is the degree of clarity and normative agreementthat provide meaning to a role
and direction for its performance.Although differentiatingpositions on
the basis of the discretion they permit, Nicholson appears to assume
that persons make a transition to an established role with reasonably
well-defined (even if broad) normative boundaries. The more ambiguous the role, the fewer are the grounds to which the individual can turn
to judge his or her own capacity to succeed in the role or by which he or
she can interpret feedback on performance (Dubinsky & Yammarino,
1984). Mentor roles are markedly ambiguous. Throughout the implementation literature,observers record the uncertainties of mentors, administrators,and teachers regardingthe central purposes of mentorship
and the specific behavior in which mentors should or might engage.
Hart (1989) tells of the teacher leader who punctuatedan interview with
a poignant question: "Whatshould we be doing?"(p. 26). The 13 teacher
leaders interviewed by Smylie and Denny (1989) voiced similar concerns; although the new leaders were relatively secure in their own
knowledge and in their aspirations, "they were much less certain about
whethertheir fellow teachers understoodtheir leadershiproles and what
those teachers and their principals expected of them in those roles" (p.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
314
Reviewof Researchin Education,16
8). In the earlystagesof Connecticut'sexperimentalprograms,teachers
designatedas "assessors"receivedmore supportfrom teachersthan
those designated"mentors,"despite the apparentlydisadvantageous
connotationsattachedto the formertitle. Neufeld(1986)attributesthis
unanticipateddevelopmentto the clearpurpose,predictablebehavior,
and structuredrelationshipsassociatedwith the assessorrole, and the
comparativelyhigh level of ambiguitysurroundingthe mentorrole.
Theuncertainties
of purposeor practicethatindividualmentorsexperienceoftengo unrelievedbyorganizational
interventionorsupport.Amroledefinitionhavebeengreatestwhere
biguityandconflictsurrounding
mentorrolesremainunlinkedto anylargerpicture,wherenormsareunfavorableto professionalgrowthor careermobility,and whereteachers
havebeen left to "inventtheirrolesas they wentalong"(Hart,1989,p.
24). Rolesappearto developmost fullywhereteachersand administratorsestablishthe teacherleader'sintendedcontributionto widelyshared
in support
goals,thenexploitresourceseclecticallyandopportunistically
of the leader'sactivities.Butin the localcareerenhancementprojectexaminedby Smylieand Denny(1989),"thedistrictdecidedintentionally
to leaveopenthe specificrolesandresponsibilities
associatedwiththese
positions.It was the responsibilityof the teacherswho wouldassume
thesepositionsto developthem"(p. 4). In the absenceof organizational
purposeand sanction,mentors'individualand idiosyncraticeffortsto
fulfilltheirperceivedobligationsmayonlyheightentheirvulnerability.
In
the careerladdersitesstudiedby HartandMurphy(1989b),teacherssuccessfullydevelopedleadershippositionsonly in thoseschoolswherethe
role wasgiveninstitutionalpurposeand structure.
In the absenceof organizational
sanction,mentorsmust relyon personal resourcesto penetratelong-standingprotectionssurrounding
teacherautonomy.The actionsthatmaybe requiredto give meaningto
the termmentorare preciselythose proscribedby the dominanttraditions of noninterference
in teaching.Withinthe confinesof the classroom,teachersrecognizeand defenda wide rangeof practiceas falling
legitimatelywithintheboundsof teaching;theroleof mentorhasno such
acceptedheritageof wide,diverse,observablepracticeas its warrant.In
most schools,teachersenjoywide latitudeto constructtheir relations
withstudentsandto maketheircurricular
andinstructional
choicesin accordancewith personalpreference;
less latithey enjoycorrespondingly
tudeto commenton or attemptinfluenceoverotherteachers'classroom
work.Theteachingroleis mostproblematic,mostnarrowlydefined,and
most constrainedpreciselyin the area(collegialinvolvementand influence) wherethe mentorrole placesits greatestdemands.
and
Mentoringis, on its face,at oddswiththeprevailingorganizational
occupationaltraditionsin teaching.Birdand Alspaugh(1986)describe
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
315
"thementors'dilemma"as the tensionbetweenthe leadershipexpectations implicitin the title of mentorandthe inheritedtraditionsof autonomy and equality:"the scarcityof traditions,organizationalarrangements,andnormsof interactionthatwouldalloworenablementorsto do
enough, with enoughother teachers,to earn their extra pay and resources"(pp. 3-4). Mentorsor teacherleadersareat substantialriskof
definingtheir positionsin waysthat sparkthe resentmentof teachers.
Ironically,teachersmay move fromclassroomteacher,with substantial
discretionoverthe mannerof theirwork,to a mentorshipthatfindsthem
moreconstraints.In this
exercisingless discretionand accommodating
instance,then, role ambiguitypromisessomethingmorethan personal
or organizational
incoherence.It laysthe groundforacdisappointments
tive conflictamongteachers,or betweenteachersand otherconstituent
groups.Suchconflict,in turn,has its own consequences.Teacherswho
haveservedas mentorsor teacherleadersdeclineto do so again;the intendedcareerincentiveis dilutedboth for them and for the colleagues
who havewitnessedtheirdefeat(Hart,1989).Facedwithmountingdissent,the institutionmakesmovesto renderthe roleharmless-and thus
useless(Bird,1986).Theprospectof increasedorganizational
capacityis
weakened.The standardof mutualbenefitis compromised.
Finally,theoriesof roletransitiontendto assumethatpersonsmakea
transitionfromone roleto another-fromteacherto principal,forexample. Even theories that account for newly createdand individually
roles"assumethat one entersfullyinto the new
wrought"idiosyncratic
role (Miner,1987).The rise of formalmentoringconstitutesa radically
differentcase,one in whichthe conditionsforroleconflictandroleoverload (Biddle,1979)are likelyconsequences.In most instances,teachers
retain the identities,obligations,perspectives,and affiliationsof the
the perspectives
classroomteacherwhileaddingon, usuallytemporarily,
and perquisitesof leadership.The ambiguitiessurrounding
mentorship
are compoundedas teachersattemptto satisfytwo sets of roledemands
thatarenot alwayscompatible.Provisionsforrotatingthe opportunities
forleadershipamonga largepoolof teachersplacea premiumon preserving one's identityas a classroomteacherand one'ssocialstandingwith
peers.Forindividualteachers,mentorshipsrepresentnot permanentpositionsbut short-termopportunitiesthat one teacherlikenedto a short
stepladder:
"youstepon andyoustepoff"(Hart&Murphy,1989b,p. 27).
Smylieand Denny(1989) concludethat the teacherleadersare
norms.
in a precarious
andambiguouspositionwithrespectto violationsof professional
Theyarewellawareof thispositionbutseemto wantit bothways.Thatis,theyseemto want
associatedwiththeirleadership
andrecognition
the additionalresponsibility
positionsbut
at the sametimetheywishto retaintheirstatusin the collegium.(p. 15)
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
316
Reviewof Researchin Education,16
In sum,conservative
the
underscore
patternsof policyimplementation
conundrumthat prefacedthis review:How mightwe accountfor the
of the
emergenceof a formalrolethatbothsustainstheessentialcharacter
that
or
and
overcomes
relation
successfully
displaces
mentor-prot6g6
normsof privacyand equalstatus?In attemptingto accountervailing
count fully for the prevailingconfigurationsof mentoring,researchers
havegivenspecialattentionto two problemsassociatedwithspecialized
teacherleadershiproles:the problemof expertisein teaching,and the
problemof expertstatusamongteachers.
The Problemof Expertisein Teaching
A recurrent
paradoxcanbe expressedthisway:mentors'claimsto professionalexpertisearebothdemandedby the roleanddeniedby history
and circumstance.Implicitin the title of mentor,advisor,consulting
teacher,or masterteacheris the presumptionof wisdom-accumulated
thatcanserveasthebasisof sensitiveobservation,astutecomknowledge
mentary,sound advice, and constructiveleadership.Demonstrated
knowledgeandskillaretheessentialgroundon whichtheroleandtitleof
mentorarefounded.Inthedistrictsurveycompletedby FarWestLaboratory (Bird& Alspaugh,1986),districtcoordinatorsratedsubjectmatter
and pedagogicalknowledgeas the two most essentialqualificationsfor
of mentorinitiativesis confoundedby two
mentoring.Implementation
issues relatedto expertisein teaching:debateover the existenceof an
agreed-upon
bodyof knowledgeto guidepractice;andthe accessibilityof
in bothtechnicalandsocial-organizational
teachers'knowledge,
senses.
WhatMentorsKnow
Whatis the natureof knowledgeto whicha mentormightlayclaimknowledgethat couldserveas the basisof a relationshipwithteachers?
Criticshavearguedthatthelowlevelof agreed-upon
expertisein teaching
increasesthe stresson practitioners,
constrainshelpgivsimultaneously
ing,andleadsteachersto discountcriticismor advice(Edgar& Warren,
1969;Glidewell,Tucker,Todt,&Cox,1983).Inthepastdecade,however,
twodevelopments
havealteredtheviewof availableexpertisein waysthat
nowinformthe mentorinitiatives.First,districtshaveincorporated
into
staff developmentand teacher evaluation a body of presumably
codifiableknowledgearisingfrommorethana decadeof classroomresearch(Brophy& Good, 1986).Thisline of classroomresearchhassupplied much of the expectedlanguageand content for mentors'work.
Second,teachers'ownpracticalknowledgehasbeengrantedmoreattention and greaterdeference(Buchmann,1986; Elbaz, 1983; Pinnegar,
1987;Yinger,1987).Studiesof teacherthinking,planning,andsituated
decisionmaking(Clark& Peterson,1986)havehelpedto replacea "dim
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
317
view of teacherknowledge"(Feiman-Nemser
& Floden, 1986, p. 512)
with a morerespectfulview.This line of researchmaygo farto bolster
publicfaiththatsometeachers,at least,commandthesortof expertgrasp
of teachingthat warrantsmentorstatus.
Theclaimsthatunderliementors'legitimacyrestbothon theavailability of an externallyvalidatedknowledgebaseandon the credibilityof a
workforce.Inpractice,externallyderivedreknowledgeable
recognizably
searchknowledgeand teachers'own experientialknowledgehaveoften
been accordeddifferentweight.Wheredistrictsclosely structurethe
relationandwherethatrelationis tightlycoupledto permentor-prot6g6
sonneldecisions,externallydeterminedprioritiesand terminologyare
likelyto overridementors'individualpreferencesand practices.In formaljob descriptionsand in the contentof training,research-based
contentappearsto dominateexperience-based
wisdom.Groundedlargelyin
classroomresearchon discreteinstructionalor classroommanagement
practices,thepracticeof mentoringin suchsiteshascometo reflecta skill
orientationtowardteachingand teachersthatmirrorsits presentdominancein professionaldevelopmentand teacherevaluationmoregenerally.Bythis view,the primarypurposeof mentoringis to produceskilled
performance.The task of the prot6g6is first to elicit or recognizethe
mentor'sskill and then to emulateit. This is an orientationreinforced
wherementoringis joined with certificationand evaluation(Allen&
Pecheone,1989).
The skill-orientedconceptionsof teachingand mentoringare less
clearlyevident in cases where the mentor role is more loosely and
permissivelyconceived. In open-endedand voluntaryconsultations
amongteachersor in the completionof specialprojects,mentorsfind
morelatitudeto exploittheirownknowledgeandinclinations.Whatever
otherdifficultiessucha stancemayengender(e.g.,problemsof roledefinitionor problemsof substantivemeritin the workmentorselectto do),
it appearsto elicit more readilythe form of teacherknowledgerepresented in studies of teacherthinkingand teacherplanning(Clark&
Peterson,1986)or in conceptionsof teachingas structured
improvisation
(Yinger,1987).In thesebroaderconceptionsof teachers'knowledge,discreteskillis embeddedin metacognitive
patternsthatenableteachersto
makesenseof theirwork.Whatteachersknowis manifestin theirsituatedjudgmentsand in the interpretations
they construct.In Kennedy's
(1987)analysisof professionalexpertise,personsrequirebothtechnique
anda graspof its underlyingrationaleasthebasison whichto innovateor
to exercisejudgment.The importanceof being able to capturethe
and not merelyto labeland reintentionalityof teachers'performance,
produceobservablebehavior,arguesfor a particularkindof relationbetweenmentorand prot6g6.Returningto Gehrke's(1988b)critique:The
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
318
16
Reviewof Researchin Education,
gift of the mentoris not narrowlyconceivedtechnique,but "anewand
wholewayof seeingthings"(p. 192).
Access to Mentors'Knowledge
Proponentsof mentoringtakeforgrantedthatproperlyselectedmentors will be a sourceof expertknowledgeto others.Accessto mentors'
knowledge,however,is arguablyproblematic.Canmentorsexpresswhat
they knowin a manneraccessibleto others;will theyhavesufficientopportunityto do so;andif theycan,willtheyfeelobligatedto do so?Mattersof opportunityhavebeentreatedelsewherein thisessay.At issuehere
are mentors'abilityto articulatetheirownexpertknowledgeandthe incentivesor disincentivesthat surroundclaimsto expertise.
Accessto mentors'knowledgeis in part a functionof the technical
capacity to make explicit certain underlyingprinciplesof practice
(Kennedy,1987).Oneconsequenceof the persistentprivacyof the classroomis thatteachersrarelyhaveoccasionto talkto fellowteachersin detail abouttheirwork.Evenmorerarelyaretheycalledon to talkaboutor
displaytheir workfor purposesof helpingotherssucceedin teaching.
Teacherscometo experiencetheirwork-and to describeit-as intuitive,
done withoutmuchconsciousframingor reflection(Buchmann,1986).
To use one's own expert knowledgein the day-by-day,moment-bymomentenactmentsof teachingis a differentmatter,intellectuallyand
fromarticulating
thatknowledgeforthebenefitof anothinterpersonally,
er'sunderstanding
and practice(Yinger,1987).Althoughwe havesome
examplesof how mentorsimaginesuchtalk, basedon theirspokenresponsesto simulatedrequestsor problemsposedto them by beginning
teachers(Parker,1989),it seemsprobablethatsimulatedresponsesoverestimatementors'willingnessto proposestraightforward
diagnosesand
to offerdirectadvice.Further,
wehaveno wayof knowingfromtheseoneway simulationshow beginningteacherswouldinterpretthe responses
mentorsgive;can beginningteachersdetectthe knowledgethat informs
mentors'comments?In studiesof preserviceteachers,McAlpine,Brown,
McIntyre,and Haggar(1988)discoverwithwhatdifficultyexperienced
teachersexpresswhattheyknow,andwithwhatdifficultynoviceteachers
learnto elicitandcomprehend
thatknowledge.
Onestudyof video-taped
interactionbetweenteacheradvisorsand experiencedteachersprovides
some evidencethat genuineinterestand good intentionsdo not stop
participantsfromtalkingpastone another(Littleet al., 1984).Advisors,
employinga languagederivedfrom classroomresearch("objectives,"
"waittime"),analyzedlessons.Teachers,employinga differ"transfer,"
ent language,analyzedthe ebbandflowof a classroom.Liketheteachers
describedby Pinnegar(1987),they examinedwhenand whythey knew
studentswere"withme,"or whattheydid to "pullthemin"if theywere
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
319
not. Examplesof naturallyoccurringexchangesbetweenmentorsand
teachers,now absentfromthe publishedliterature,mightenableus to
groundglobal assessmentsof perceivedeffectivenessin concreteinstancesof communicationaboutteaching.
Accessibilityis onlypartlya matterof one'sabilityandopportunityto
articulateone's principlesand practices.It is also shapedby the incentives or disincentivesthat promptmentorsto extendtheirexpertiseto
othersor to withholdit. In their studyof professionalsupportamong
teachers,Glidewellet al. (1983) foundthat the relationbetweenstress
(need for help) and actualhelp seekingwas mediatedby the degreeof
teachers'commitmentto establishedoccupationalnormsof autonomy
and equality.Basedon Glidewellet al., we wouldexpectthat mentoring
relationswouldbe most likelyto bearfruitwherecommitmentsto individualautonomywereweakandwherecountervailing
normsof collegialthe caseliterature,
ity prevailed(seealsoRosenholtz,1989).Throughout
we find instancesin whichmentorsexpresshumilityabouttheirownexpertise,fearingcollegialcensure(Littleet al., 1984).Therewardsaretenuous indeed.Neitherin formalevaluationschemesnor in the informal
rewardstructureof schoolsareteacherscelebratedforcontributing
to the
successof otherteachers,nor penalizedfor failingto do so. In fact,the
rewardstructuremayoperateto accentuatethe normof privacyand to
promotehoardingof insights,methods,and materials.Revealingthe
"hiddencost of sharingexpertise,"Allen(1989)describesa "mentoras
miser"syndromethatprevailswhenmentors'knowledgecomprisesa privatestoreof ideasandmaterialsthatformthebaseof professionalstanding and senseof self.
The Problemof ExpertStatus AmongTeachers
Publicly acknowledgedand rewardeddifferencesin expertiserun
counterto inheritedtraditionsin teaching(Feiman-Nemser
& Floden,
1986;Smylie&Denny,1989).Withinthecultureof teachers,informalacarecommon,but formalexpertstatusis suspect.Thus,
knowledgements
mentorsdiscounttheirspecialexpertiseas a basisof theirprofessionalrelationswithteachers.Theproblemof expertiseis at the heartof the pervasiveambivalenceaboutmentoringandthe sourceof whatBird(1986)
characterizes
as the mentors'dilemma:"Theinstrumentalstatusdifferencesthatthe MentorProgramcallsforarevirtuallywithoutprecedentin
teaching'segalitarianand individualistictradition"(Bird& Alspaugh,
1986, p. 3).
Mentorrolesachievespecialsignificance(and are renderedspecially
problematic)in an occupationthatis constrainedby normsof equalstatus and autonomy,is flat in its careerprofile,and in whichan agreedupon body of professional knowledge and practice is absent. Occupa-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
320
Reviewof Researchin Education,
16
tional traditionsare mirroredin organizational
realities;schoolsrarely
structurethe workof teachingto promotethekindof mutualinterdependencefavorableto mentoring(Little,1988).Smylie(1989)hypothesizes
that mentoror masterteacherinitiativesareless likelyto win teachers'
supportthan are other formsof collegialexchange(e.g., peer support
withthe dominantsogroups)becausethe formerarelargelyincongruent
cial contextsof teachingand the psychologicaldispositionsof teachers.
Relyingon theoreticaladvancesin the studyof cooperationandhelping
on thecontributions
of Deutsch(1982),Smylie
behavior,andparticularly
thatarecooppositsthatteachingfavorscollegiallearningarrangements
erativeratherthancompetitive,thatassumeequalityof powerandinfluenceamongall membersof thegroup,thatstresssocioemotional
support
overtaskorientation,andin whichsocialinteractionis governednot by
bureaucratic
rulesbut by tacitlyheld norms.Smylieconcludes,
Masterteacherprograms
induceformalstatusdifferentiation
amongteachers.Theyplace
teachersin superordinate
and subordinate
rolesand suggestnonreciprocal
relationships
betweenmasterteachersandotherteachersmaybe govamongteachers.Therelationship
ernedbyrulesandexpectations
andmaybegearedmore
developedoutsidetherelationship
towardtaskperformance
andaccomplishment
thansocioemotional
support.In addition,
elementinthattheextrinsicrewards
associated
withattaintheymaycontaina competitive
of statusdiffering masterteacherstatusarelikelyto be dependent
upontheperpetuation
ences.(p. 11)
Thehistoryof implementation
is in partthehistoryof accommodating
the tensionssurrounding
leadershipin teachingby teachers.Presumably,
therearetworesponsesto the conflictbetweennormsof equalstatusand
the implicationsof the mentorrole.In one response,districtscouldwork
to justify legitimatedifferencesbased on demonstrateddifferencesin
knowledge,skill,andcommitment(Bird&Little,1985).Inoneof thetwo
case studysites describedby Hart(1989),this appearsto havebeenthe
strategy.The modalresponse,however,has been to diminishthe status
implicationsof the title andthe otherresourcesattachedto the role.Despite someprominentexceptionsin the caseliterature,the maintrends
showeffortsto accommodateratherthanalterthe egalitarianand individualistictraditionsthat inhibitthe developmentof mentorroles.
Themaintendencyin formalizedmentorprogramshasbeento diminish the statusdifferencesimpliedbythetitleof mentorratherthantojustify those differencesor to createthe conditionsconsistentwith their
existence.Status-reduction
strategiesmayenableschoolsanddistrictsto
secureshort-termsupportandto neutralizeresistancein the earlystages
of a program.The nine case studiesassembledby FarWestLaboratory
(Birdet al., 1984)andthe five careerladderdistrictsexaminedby Hart
andMurphy(1989b)showthe diversemeansemployedby districtsto re-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
321
duce the status significanceof formalteacherleadershippositions. In negotiation with teachers' associations, districts broadened the range of
mentors' tasks but reduced their demands on special expertise, making
the tasks more like familiar sorts of "extrawork for extra pay."They enlarged access to a wider pool of teachers by modifying selection criteria
and providingfor frequentrotationof opportunitiesto apply.In doing so,
they obscuredthe place that special expertise occupies in leadershipand
established the presumption that the earned right to lead was widely, if
not uniformly, distributed among the teaching force.
Tensions surroundingmentors' expert status are also alleviated by organizing the work of mentoring at a distance from the classroom. The
norm of noninterferenceis honored in part by the generallypermissive
orientation toward mentors' direct involvement with teachers;mentors
workwith individualteachers"byrequest."In largemeasure,mentorprograms have achieved constituent support not by pursuingthe classic dimensions of close interpersonal exchange and consultation associated
with mentoring, but by evolving a generalizedservice role in supportof
staff development and curriculumdevelopment (Little et al., 1987). Districts have absorbedthe mentor role into a general pattern of specialist
positions that provide out-of-classroom opportunities for individual
teachers and expand the district's capacity to pursue district goals.
Among the sources of influence on mentors'plans in Californiadistricts,
district priorities rankedhighest;althoughthere are wide within-district
and between-districtvariations, many mentors worked throughout the
district with greaterconsistency than in their own schools, and worked
with their own faculties as a group less often than with groupsassembled
for districtwide workshops(Bird & Alspaugh, 1986).
Finally,the implicationsof the mentortitle are softenedby focusingthe
work of mentors in domains where status differencesare genuinely more
acceptable,as in supportfor first year teachers,teachersnew to a district,
teachers confronted with new instructionalassignments,or teachers engaged in innovation. Of these, mentoring for purposes of teacher induction is the dominant case.
IN SUPPORT
THEDOMINANT
CASE:MENTORING
OF TEACHERINDUCTION
Intended improvementsin teacher induction have supplied the dominant rationalefor the proliferationof mentor roles and thus the main setting for empiricalresearch.Fullytwo thirds of the publishedreferencesto
mentoring in the 1980s concentrateon mentoringas a principalcomponent of induction programs(Gray & Gray, 1985; Huling-Austin, 1988;
Stoddart, 1989). Implicitly, the main benefits of a mentor-prot6g6 rela-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
322
Reviewof Researchin Education,16
in the
tionshipareachievedwithinthefirstyearsof teaching,culminating
certificationdecisionin somestatesor in thetenuredecisionin localdistricts.In this sense,the dominantfunctionsof mentoringin education
in busiparallelsomeof the recentformalizedmentoringarrangements
or
on
ness, industry, governmentby concentrating organizational
entry
(Zey, 1984).
The introductionof organizationally
sponsoredmentoringassumes
the
particularsignificanceagainst backdropof researchon conditionsof
teachers'entryinto teaching.Althoughsomeaccountsof teachercareers
distinguishbetweenrelativelyeasy and relativelypainfulbeginnings
(Huberman,1986),observersspanningat leasta centuryhavehighlighted
the "realityshock"thatcommonlyfollowswhennoviceteachersabruptly
andwithoutassistanceassumefull-scaleandfull-timeresponsibilities
for
teaching(Lacey,1977;Lortie,1975;Veenman,1984;Zeichner&Gore,in
press).Suchconditions,saycritics,drivecapablepeopleout of teaching.
Forthosewho remain,thesesameconditionsplacea premiumon tricks
of thetradethatenableteachersto survivebutthatalsoretardtheirdevelof teaching,theircapacityfor
opmentof moreprincipledunderstanding
criticalanalysisor "expertiseas deliberateaction"(Kennedy,1987, p.
148;see also Carter,1988;Nemser,1983).Oncein commandof a rudimentaryset of knowledgeandskill,teachers(likeotherbeginningprofesandmaydiscount
sionals)mayengagein behaviorthat is self-validating
criticism from others (Bucher& Stelling, 1977; Feiman-Nemser&
Buchmann,1986).Worse,suchconditionsmaybothproduceandperpetuatemarginalperformance
in the classroomandtenuouscommitmentto
teaching(Bridges,1986).
Under the terms of reforminitiativesin the early and mid-1980s,
teacherinductionhasbeentheobjectof effortsto expandsupportfornew
teacherswhilealsotighteningscrutinyof theirperformance.
Mentorship
occupiescenterstagein the designof suchefforts.Its proponentsanticipatethat by directassistanceand personalinvolvementwithnewteachers,mentorswillrelievesomeof thestressassociatedwiththeintellectual,
social,andemotionaldemandsof firstyearteaching.Thetest of mentor
roleslies in mentors'abilityto alleviatethe shockof entryintoteaching,
hastenthe pace of learningto teach,modelfavorableprofessionalrelations amongteachers,and reinforceteachers'loyaltyto the profession.
To whatextentdo suchformalizedarrangements
simplyextendnatubetween
rally occurringhelpingrelationships
experiencedand novice
teachers?At issueherearerelationships
of a magnitudeandintensityadequateto ensurenot only the comfortand self-confidenceof beginning
teachers,butalsotheirprofessionalcompetenceandcommitment.Little
(1987) proposes,
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
323
A distinctionis in orderbetweenthesocialsupportthatputsnewcomers
ateaseandtheproandpracticeof teaching.... Withoutdifessionalsupportthatadvancesone'sknowledge
thecentralissuehereis one
minishingtheimportof moralsupportandemotionalsolidarity,
of professional
relationsthatgo wellbeyondthe usual"buddy"
(p. 498)
arrangement.
Despitethewidelyrecognizedmaximthatteachersinviteothersto "askif
youneedhelp,"it appearsthatveteranteachersrarelyengagein relations
with beginning teachers that would warrant the designation of
mentorship.To the extent that beginningteachersreceivethe kind of
close attentionthat accordswith the imageof mentoring,it commonly
derivesfromsourcesotherthantheirpeers.Eventeacherswho claimto
have had a mentortypicallyfound their supportoutsidethe teaching
ranks;only3 of 41 teachersinterviewedbyGehrkeandKay(1984)identified otherteachersamongthe significantmentorsin theirlives. These
findings are consonantwith other portrayalsof informalinduction
(Lortie,1975).As one mightpredicton the basisof generalizedprofesconditionsof theworkandworkplace,insionalnormsandthe structural
formal mentoringof beginningteachersby experiencedteachersis a
low-incidencephenomenon.There are wide individual variations,
thoughsome schoolsmorethan othersdisplaynormssupportiveof intense and consequentialsupport(Little, 1987;Meister,1987).
Surveyscenteredspecificallyon mentoringrelationsmayoverestimate
the importanceof mentoringto teachers'careersas they are presently
constituted.In the surveyconductedby Gehrkeand Kay (1984),many
teachersclaimedto havehad a mentor,but relativelyfew of theserelathe levelof involvementthatClawson(1980)andother
tionsapproached
relation.Despitethe
theoristswouldcountas a genuinementor-prot6g6
vivid portraitsof the positiverolemodelswhomteacherslateremulated
or the negativeoneswhomtheydenigrated,rarelydo we encountertributesto a mentorin teachers'first-handaccountsof choosingto teachand
developingone'steachingovertime(Elbaz,1983;Macrorie,1984;Mead,
1989;Measor,1985;Nias, 1989).Mentorship,it appears,is not firmly
rootedin the informalconventionsby whichneophytesarebroughtinto
and academicshavepromoted
administrators,
teaching.Policymakers,
formallyassignedmentorshipson thegroundsthattheybothexpandsupportandhelpto justifymorestringentevaluation,therebyimprovingthe
prospectsfor a strongteacherworkforce.Lookingto the traditionsof
teachingand preferencesof teachers,however,formalmentorshipmay
in pursuitof institutionalpurconstitutea caseof "contrivedcollegiality"
& Dawe,
posesto whichteachersmayor maynot subscribe(Hargreaves
1989).Nonetheless,formalmentoringis on the rise,justifiedprincipally
as a remedyforinadequateinductionsupportandorganizedprimarilyin
termsof expandedreservesof help.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
324
Reviewof Researchin Education,16
Mentoringas Help-Giving
Because formal induction practicesare dominated by a conception of
mentoring as help or assistance, they are usefully interpretedin light of
advances in the study of helpingbehavior.Studies that highlightthe help
giving aspect of mentoring range from programevaluations that assess
the perceived utility of mentor assistance (Huffman & Leak, 1986) to
microinteractional studies that probe the interpersonal dynamics of
mentor-prot6g6interactions (Allen, 1989; Shulman, 1987). Virtually
none of these studies, however,has been informed explicitly by the kinds
of theoreticalconstructsthat have shaped social-psychologicalinvestigations of helping behavior during the past decade.
The antecedents,character,and consequencesof help all are rendered
problematicby recent research;neither the nature of help nor its virtue
remainstaken for granted.Gergenand Gergen(1983) underscorethe social construction of helping, maintaining that instances of help assume
meaning only in the context of an interpretivesystem. In examples that
rangefrom a bystander'sgestureof aid to a victim of troubleto large-scale
interactions between Third World countries and aid-donatingagencies,
researchersfind that both the definition of help and persons'attitudestoward it are conditioned on a complex host of individual and social circumstances(Fisher,Nadler,& Whitcher-Alagna,1983;Gergen& Gergen,
1983; Gross & McMullen, 1982).This literaturebeginsto account for the
occasions on which help is soughtor not, offeredor withheld,acceptedor
rejected.
The programimplementationliteraturehighlightsthe dilemmas of the
teacher turned mentor, and the conditions surroundingemergenceof a
mentor role. The induction literature,by contrast, focuses more closely
on the teacher as the potential beneficiary of mentorship.By turning attention to the recipient of help, recent researchquestions long-standing
assumptions about receptivityto and gratitudefor aid. The choices that
persons make to solicit aid or to accept assistance when it is offered are
determined in largepart by their assessment of its psychologicaland social costs: the costs to their sense of competenceand their status with important others, and the obligations they incur by accepting proffered
resources (Gross & McMullen, 1982).
Fisher et al. (1983) employ four theoreticalmodels to explain persons'
probableresistanceto or acceptanceof aid. Equitytheories,togetherwith
related reciprocityand indebtednessmodels, start with the premise that
personsseek parityin their interpersonalrelations.To the degreethat persons find themselvesindebtedin waysthey cannotrepay,or believe themselves to be implicitly derogated by their participation in a helping
relationship, they can be expected to resist help (see also Greenberg&
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
325
Westcott, 1983). In business and industry, the success of informal
mentorships rests largelyon the mutual benefits they demonstrate.The
assistanceand sponsorshipprovidedby the mentoris compensatedas the
prot6g6 is delegated a larger share of the work, contributing to the
mentor'sown productivityand careerprospects(Zey, 1984). Reciprocity
is achieved. By this argument, mentorships among teachers can be expected to thrive to the extent that the participantsdetect some measureof
mutual gain in the exchange of ideas, materials, methods, and labor.
A second and complementarytheoreticalperspectivesuggeststhat persons are reluctantto seek help when they believe that doing so will unduly
restrict their own freedom to act. Reactancetheory applies most clearly
where seekinghelp entails restrictionson physicalmovement (e.g., hospitalization) but a broaderinterpretationis possible. This generalizedpsychological disposition toward freedom of choice may be accentuatedin
the context of teaching and other professionalized occupations, where
professionalnorms favor autonomy.Based on reactancetheory,then, we
can expect help to be welcomed to the extent that it expands a beginning
teacher'srangeof curricularor instructionaloptions and sense of efficacy.
To the extent that mentorsare seen as agentsof control who curtailcurricular and instructionalchoice, however,help will be resisted. In one example, beginningteachers objected to teacher leaders' implications that
they were only "reallyteaching"when they employed narrowlydefined
"principlesof effective instruction"or clearlyobservable"elementsof an
effective lesson plan" (Hart, 1988, pp. 10-11).
Attributiontheories rely on persons'own interpretationsof the conditions and consequencesof help to account for the incidence of help seeking. The complexities and subtleties of attribution theory cannot be
representedadequately here, but certain main theoretical premises appear to have particularimport for the success of mentoringin teacherinduction. First, teachersare more likely to believe help is legitimatewhen
they can attributethe need for help to the complexitiesof the task and the
situation (externalattribution)ratherthan to the limitations of their own
competence (internal attribution). Formal teacher induction programs
may induce receptivityto help by declaringpubliclythat the first yearsof
teaching are especiallydemanding,regardlessof individual skills and talents. Second, there is evidence that help is more often and more favorably
accepted when it is offered than when it must be requested.The very act
of requestingassistance may prompt internal attributionsof failure (see
also Gross, Wallston, & Piliavin, 1979). The prevailing norm of
noninterferencein teachingtakesthe formof an informalrulethat one offers advice only when asked. Such an axiom may inadvertentlydepress
help seeking.
In the fourth formulation developed by Fisher et al., the relationship
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
326
Reviewof Researchin Education,16
between the situational conditions that create a need for help and persons' actual responsesto aid are ultimatelymediated by potentialthreats
to self-esteem, social identity, and relations with others. To the degree
that seekingor acceptinghelp representsa threatto self-esteem,these theorists argue, persons will persist in attempts at self-help. To the extent
that threatsto self-esteem and social standing can be avoided (or advantage gained)personswill seek help from others. Combiningthis view with
the other relevant theoretical frames, we can anticipate that threats
to teachers' self-esteem are alleviated when the helping relation with
mentors stems from legitimately difficult circumstances rather than
from personal inadequacy,when it permits or even requiresa degree of
reciprocity,when it adequately preservesthe teacher's freedom to act,
and when it demonstrably contributes to the teacher's success and
satisfaction.
Teacherinduction programsthat are founded on the utility of help confront both the general culturalambivalence about help seeking and specific occupational prohibitions surrounding interference in teaching.
Independent of their individual capacities and dispositions, teachers'
attitudes toward mentoringare affected by general occupational images
associatedwith professionalautonomy and by local norms governingaid
and assistance (Rosenholtz, 1989). Applying the broad social-psychological perspective associated with researchon helping, Glidewell et al.
(1983) examinedthe incidenceof help among teachers.The centralpremise of their work is that stress-producingconditions can be expected to
stimulate help seeking. In teaching and in other service professions,they
argue,stress is exacerbatedby lack of experience,lack of availableexpertise, ambiguitysurroundinggoal attainment, and departuresfrom an optimal client load. These are factorsthat to varyingdegreesaffect teachers
in general, and that plausibly affect beginning teachers most. All other
factorsbeing equal, one might expect beginningteachersto be avid seekers of professional support or eager recipients of the support offered by
others. But all other factors are not equal. In particular,Glidewell et al.
demonstratethat the relationshipbetweenstressand help seekingis modified by teachers' commitments to traditional norms of autonomy and
equal status. The norm of autonomy not only establishesa rightto independent practice,but also obligatespractitionersto take care of their own
problems;the norm of status equality constrains practitionersto reject
implications of status difference. Where these traditional norms have
been weakened or displaced by norms of collegial support, teachers
openly request and offer help, and the predicted relationship between
stress and help seeking is sustained. Under such circumstances,
competence-baseddifferences in status appearto be acceptable(see also
Smith & Sandler, 1974). Responsibility for the successful induction of be-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
327
ginningteachersor othernewcomersis widelydiffused.One of the five
case studyschoolsdescribedin Meister's(1987)reviewof school-based
inductionprogramsexemplifiesthis situation.Wherethe traditional
normsholdsway,however,teachersfindcovertwaysto relievestresswithout exposingtheir difficulties.Overt requestsfor assistanceare rare.
Through"experience-swapping"
they garnerinformation,advice, and
sympathyindirectly.Discussionsof teachingacquirea piecemealcharacter,of doubtfuldepthand consequence.
Therelationsbetweensocial-psychological
conditionsandpersonalreactionsare intensifiedwherethe tasks requiringhelp are crucial.The
morecentralthe focusof helpto one's professionalidentity(classroom
instruction,in the case of teachers),the moresalientbecomethe conditions that supportor threatenone's senseof self. Wherecompetencein
teachingis judgedby individualprowessin the classroom,help seeking
maybe suppressedas teachersattemptto hide errorsand publicizesuccesses (McLaughlin& Pfeifer, 1988). Teachers'participationin mentoringmaybe affecteddirectlyby the externalpressureto perform,and
the consequencesassociatedwith failure.Thisis a prospectso farunexaminedin the educationalmentoringliterature,althoughit hasits parallels in businessand industry(Zey, 1984):Genuinementoringis more
whereeachperson
widespreadunderconditionsof highinterdependence,
bearsthe consequencesof others'successor failure.
In sum, recentadvancesin researchon helpingforceus to examine
more closelythe taken-for-granted
assumptionsthat undergirdformal
mentorprograms.The characterandconsequencesof mentors'relationshipswithbeginningteachersareproductively
placedin a broadersocialperspective,andformalmentorprogramsconsideredin the
psychological
widercontextof socializationinto teaching.
Mentoringin the Contextof TeacherSocialization
Studiesof discreteinductionactivitiesformthe largestsinglebodyof
researchexplicitlycenteredon mentoringpracticesand relations.Precisely becauseof its associationwith specificinitiatives,however,this
researchtendsto displaya narrowlyprogrammatic
conceptionof induction. It is virtuallydevoidof reference,forexample,to the largeandrich
literatureon inductionintogroups,socisociologicalandanthropological
eties,andoccupations(Schlechty,1985).Studiesof teachersocialization
helpto locatementoringrolesandpracticesin the largercontextof occupationaland organizationalinduction.Althoughrecentreviewsof the
teachersocializationliteraturegive little or no explicitattentionto the
termmentoring,they do accountfor the rolecolleaguesplayin shaping
teachers'perspectivesandpractices(Zeichner& Gore,in press).Individual studiesexaminethe powerof fellowteachersas positiveor negative
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
328
Reviewof Researchin Education,16
role models (Measor, 1985;Mead, 1989) and the normsgoverningadvice
giving among teachers that are central to mentoring (Glidewell et al.,
1983; Little et al., 1984).
A socialization perspectivemakes centralthe web of professionalrelations and institutionalpurposesin which the mentor-teacherrelationship
resides. It permits us to trace the character and consequences of
mentoringnot only to the psychologicaldispositions and technicalcapacities of individuals, but also to the social context that enables or constrains such relationships(Smylie, 1989). It entertainsa largerdefinition
of the teacher's role, encompassing the teacher in the classroom, as a
member of a faculty,and as participantin a wider professionalcommunity (Little, 1987). To sort out the relativecontributionsmade by mentors
to beginning teachers' success and satisfaction in teaching will require
studies that encompass valued outcomes that range from a basic command of pedagogicaltechnique to a capacity and inclination for wellinformed innovation.
The broad socialization consequencesof mentoringare maskedby research designs that remain conceptuallyundeveloped and methodologically narrow.To date, we remain unable to assess the claims that have
been made about the influence of mentoringon teachers'classroomperformance, their long-term development, or their career commitment.
Most accounts are based on post hoc questionnaires(Huffman & Leak,
1986) or on interviewswith mentorsand teachers(Allen, 1989);there are
no published accounts of observed mentoring in action, even though
some study designs provide for observationof selected activities (Wasley,
1989). Most studies concentrateonly on direct participantsin induction
programs.In the absence of a comparisongroupof unmentoredteachers
it is difficult to determine what, if anything,mentors contributedto the
possible differencesbetween painful and easy beginnings,as Huberman
(1986) characterizesthem. Analyses fail to distinguish various dimensions of involvement and impact, relyinginstead on globaljudgments of
utility and anecdotal accounts of the content of interactions.Aggregate
analysesof beginningteachers,mentors,or even mentor-teacherpairsobscure consequential aspects of the specific school context in which
mentoring is attempted. Post hoc, global assessmentsof a mentor's usefulness (Huffman& Leak, 1986) or official recordsof mentors'activities
(Odell, 1986) would be profitablysupplementedby detailed histories of
mentors' interactions with beginning teachers. Such studies might
productively combine elements of the structuredfield experiment, ethnography,and biography.
Programevaluations and case studies have only just begun to fill out a
detailed picture of the actual work of mentors with beginningteachers.
Despite theoretical and methodological limitations, certain themes
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
329
emerge. In the early stages of mentoring,consultationon curriculumand
instruction takes second place to information or coaching about the system at large (Odell, 1986; Stoddart, 1989). Newcomers, whether new to
teaching or new only to the local system, expect mentors to help them
make sense of the institution's formal and informalrequirementsand resources (Odell, 1986). Activities oriented towardemotional supportconstitute a small percentage of officially recorded support activities, but
loom largein importanceto beginningteachers(Allen & Pecheone, 1989;
Huling-Austin, 1988). It seems likely that beginningteachers will judge
even activities intended for other purposesin accordancewith their effect
on personal confidence and security.
The emphasis on comfort and harmonious relations between mentor
and teacher may preclude productive confrontationwith importantbut
difficult matters of practice (Hollingsworth,1989). In this, mentoringin
support of beginning teachers differs from the informal mentoringthat
grooms selected individuals for leadership positions either in business
(Zey, 1984) or in educational administration(Baltzell & Dentler, 1982).
In these instances, the mentor's and prot6g6'scareerinterests are closely
linked. The prospectsthat they will rise or fall together help to drive an
emphasis on competence. It is through close attention to practice that
mentors become assured that their judgment (in recruitingthe prot6g6)
has been validated and that prot6g6sacquirepersonalconfidence in their
professional competence. To uncover the complex connections between
small practicesand largerschemathat advanceunderstandingof teaching
would seem to require both shared curiosity and joint scrutiny of practice. The relation between mentorship and "eased beginnings"ceases to
seem self-evident.One might imagine, for example,that the presenceof a
mentor makes the first year of teaching more strenuousin the short run,
even while promising substantial rewardin the longer term. However,
available accounts of mentoring suggest that mentors and beginning
teachers spend too little time in one another'scompany,and too little of
that time on actual classroom work, to achieve such understanding.
Even linked to formal induction programs,mentoringremains a relatively low-incidence phenomenon; beginning teachers typically report
sparing contact with their mentors (Allen & Pecheone, 1989). The dilemma is exacerbatedwhen mentors and beginning teachers work at a
distance, assigned to different schools, grade levels, or subjects, or committed to different beliefs about teaching and learning (Shulman &
Colbert, 1987). This has been the impetus for districtsto attempt subject
and grade-levelmatches in pairingmentorswith beginningteachers.Presumably,such matches permit the mentor to establish a persuasiveset of
credentials, both formal and experiential, and to supply a substantively
rich base for advice, assistance, and consultation. When asked, teachers
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
330
Reviewof Researchin Education,16
claimto prefera relationshipwithmentorswhosepresentassignmentor
teachinghistoryis closeto theirown(Huffman& Leak,1986).However,
interviewswith teachersand mentorswho have been matchedon the
basis of gradelevel and subjectpresenta less clear picture(Allen &
Pecheone,1989).In districtswherethe mentorrelationremainsambiguconstrainmentorsfrom posing
ous, where normsof noninterference
toughquestionsaboutpractice,andwherementoringtakesplacelargely
outsidethe classroom,thereis littleapparentreturnfroman investment
in subjectandlevelmatches.Theattemptin teacherinductionto achieve
matchmaking
throughformalassignmentappearsto fail at leastas often
as it succeeds.It founderson its inabilityto producegenuineinterdependencewhereit doesnotexistin thelargersystem,andbyits inattentionto
localprofessionalnorms.It suffers,too, fromuncertaintyabouthowto
reconcilethe instrumentaldimensionsof the match(teachingassignment)withthe inescapablesocialandemotionaldimensionsof personal
interactions.In theserespects,the experiencein educationparallelsthat
of formalmentoringprograms
in businessandgovernment(Kram,1986).
menFurther,individualassignmentshavethe effectof overemphasizing
tors'individualresponsibilities
forthe successof beginningteachers,and
maskingthe largersocializationcontextin whichthose teacherswork.
None of the availablestudies,however,has examinedsystematically
the
relationshipsamongthe amountof interaction,the characterof the
mentoringrelationship,and the consequencesfor beginningteachers'
performanceand attitude.Such an analysismight addressthe policy
problemposedby Huling-Austin
(1988)-how muchsupportis enough,
or too much.
Thelitaniesof troublesurrounding
the firstyearsof teachinghavebeen
chronicledpersuasively(Veenman,1984).Coupledwith the burgeoning
researchon the subtletiesandcomplexitiesof expertclassroomteaching
(Berliner& Carter,1986; Doyle, 1979;Jackson,1968, 1986;Yinger,
1987),theysuggestsomeof the reasonswhymentoringmightbe judged
personallyand organizationally
productive.Mentorroleshaveemerged
as the favoredstrategicoption in largerpolicy initiativessurrounding
teacherinductionat the local and state levels,takingprecedenceover
otheralternativesthatmightincludereducedworkload,peergroupsupstaffdevelopment.Theveryprominenceof
port,andformallystructured
mentorrolessignalsa characteristic
aspolicystance.Thedisadvantages
sociatedwithan abruptentryintoteachingareto be relievednotbyalteringthe studentloadthatnoviceteachersconfront,or by slowingthe pace
at whichtheyassumethe fulltasksof teaching,butby increasingtheiraccess to pedagogicalexpertise,organizational
savvy,and socioemotional
support.
In principle, mentoring seems a sensible response to the present inade-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
331
quaciesof teacherinduction.Problemsof expertstatusarein somemeasure relievedby legitimatedifferencesof perspectiveandexperiencethat
mentorsbringto a relationshipwith beginningteachers.The recurrent
problemsof the firstyearteacherarereasonablywelladdressedbythe cumulativeresearchon effectiveclassroommanagementand instruction.
Formalinductionprojectspresentnaturallyoccurringexperiments,most
of whichuse mentorrolesas oneof severalelementsin a largerconfiguration of support.Suchroleshavebeencreditedwithhavinggreatereffect
on beginningteachersthan other programelements (Huling-Austin,
1988),althoughwe haveonly weakevidenceon whichto sustainclaims
for the specialsalienceof mentoring.On the whole,programevaluations
treatmentoringas a self-containedintervention(e.g., Huffman& Leak,
1986). To date, there is no publishedresearchdesignedto examine
orto testits relativepower
mentoringas oneof severalpolicyalternatives,
The ocwhenotherfeaturesof the settingare favorableor unfavorable.
casionalassistancetypicallyavailableas partof a formalmentoringarrangement,for example, is unlikelyto compensatefor problemsof
or otherformsof workoverload.Giventhe structeachermisassignment
turaland culturalconstraintson mentoring,its salienceis likelyto dependon the degreeto whichit is congruentwithotherformsof supportin
the lives of beginningteachers.Discreteprogramevaluationsand narrowlyconceivedpolicystudieshavedone little to informthe largerpicture,addressingfundamentalquestionsaboutthe placeof mentoringin
the improvementof teachingor the strengthening
of the teacherwork
force.Mentoringin educationhasderivedits mainjustificationfrominadequaciesin the inductionof teachers.The rationalesremainto be
to do so areplentiful.Suchtestswillbe most
tested,butthe opportunities
persuasiveif theyareinformedby recentadvancesin the studyof helping
and by a broadperspectiveon socializationinto teaching.
AND
MENTORING
OPPORTUNITIES
TEACHERS'CAREERENHANCEMENT
A majorimpetusforthedevelopmentof mentorrolesandotherteacher
leadershipopportunitiesrestswith the publicinterestin a teacherwork
force that is competent,committedto teaching,and reasonablystable
(Sykes,1983).Mentorshipsare promotedon the groundsthat suchprofessionalopportunities
outsidetheclassroomwillhelpsustaintheengagement and commitmentof experiencedteachersinside the classroom
(Wagner,1985).Lessdirectly,the availabilityof mentorshipsand other
specializedteacherleadershiprolesis expectedto holdout an imageof a
more attractive career to those entering teaching.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
332
Reviewof Researchin Education,16
AdvancementVersusRetention
Case studies in business and in education provide an instructivecontrast between conventional patterns of informal mentoring associated
with careeradvancementand the emergingpatternsof formalmentoring
associated primarilywith organizationalentry or with institutional concerns for adequatelevels of performance.In business and industry,informal mentoringhas been the centralelement in a "systemof professional
patronage and sponsorship"(Shapiro, Haseltine, & Rowe, 1978, p. 55)
devoted to long-termcareerdevelopmentfor a relativelysmall proportion
of organizationalnewcomers (see also Kanter, 1977; Roche, 1979; Zey,
1984). In effect, mentors "overseea career"for personstargetedfor management positions (Zey, p. 7).
Mentorshipsurvives, accordingto Zey (1984), on the basis of mutual
benefits derivedby the prot6g6,the mentor,and the organizationat large.
When the relationshipis successful,the mentorboosts his or her own productivity throughassociation with a capableprot6g6.The mentor'swork
load is eased, or time is freedto take on more ambitiousprojects,as some
shareof the workis graduallyassumedby the prot6g6.By intensivecollaboration with the prot6g6,and by havingto makehis or her own knowledge
clear,the mentor spawnsnew ideas and new methods at a higherrate. In
the words of one corporatementor: "Twocannot only work better than
one; they can often workbetterthan two" (p. 8 1). The mentor-prot6g6relation "expandsareasof permissibleinquiry"by makingit acceptableto
ask naive questions (p. 18). The mentor both develops and demonstrates
the prot6g6'sknowledgeand skill not by instructionor help but by orchestratingopportunityand by joint involvementin work.As the prot6g6wins
the favorableattention of others in the organization,the mentor'sreputation as a "promoterof good people"grows, and the mentor's own career
prospects are enhanced. The prot6g6's path through the promotional
ranksis cleared.Opportunitiesto demonstratecompetenceand initiative
are made more readily available, and are more shrewdlyconstructedby
the mentor to highlight the prot6g6'sspecial talents. The organization,
too, reapscertainbenefits. Mentorshipacceleratesthe pace at which newcomers acquire the technical, social, and political knowledgeneeded to
succeed. Widespreadmentoringhelps to retain entrepreneurialindividuals who might otherwise leave by assuringthem adequate advancement
opportunityand recognition,and by building personalas well as organizational loyalties within the corporation.A system of mentoringassures
managementsuccession and continuity. Together,these mutual benefits
to mentor, prot6g6,and organizationsustain the practice of mentoring;
the supportsfor mentoringare weakenedwhere the benefit to any of the
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
333
three is uncertain. Both for individuals and for the largerorganization,
however,these are benefits that take time to mature.
The purposesand practicesof informalmentoringin business have no
readilyapparentcounterpartin mentoringamongteachers,althoughthey
do haveclearparallelsin the practicesby which classroomteachersare informally groomed for positions as administratorsor specialists (Baltzell
& Dentler, 1982). Retention, not advancement,is the stated institutional
aim of formal mentoring among teachers. This concern with retention
shapes the conception of mutualbenefit that underliesmentorprograms.
By granting experienced teachers the status and responsibilities of
mentorship, districts expect those teachers to experience a renewal of
their enthusiasm for teaching. Prospectsfor careeradvancementare not
consideredcentral,though some mentorsdo in fact move on to administrative positions (Ruskus, 1988). By asking mentors to devote their talents and energy to the support of beginning teachers, the district
anticipates a lowerturnoverrate and more appropriatetenure decisions.
For beginningteachers, the benefit is in relief of the stress of first year
teaching and in enhanced prospectsfor job security (tenure).For the experiencedteacherswith whom mentorswork,the benefit residesin an expanded pool of ideas, methods, and materials,or sometimes in relief from
negativeevaluations.Forthe mentor,the expectedbenefitsbegin with the
status associated with the title, compensation, and other resources it
brings.Mentorsin education,as in business(Zey, 1984),celebratethe way
in which their own performanceand learningexpandsas they attempt to
review and revealwhat they know to others. In education, mentorshipincreases the mentors' own access to still other professional development
opportunities. Mentors in California districts, for example, were more
likely than other experiencedteachers to have observed in others' classrooms, and to believe that their own teaching had improved as a result
(Little et al., 1987). Ultimately,the benefits must extend to the psychicrewards that accompany a close and productive relationship with other
adults, parallelingthe intrinsic satisfactionsof the classroom. Forthe organization,successfulmentoringincreasesthe returnon investmentin selecting and hiring, and permits public assurance regarding pretenure
screening and the overall quality of the teacher work force.
Mentoring in K-12 teaching thus neither promises nor is premised
upon an advancementincentive, but ratheron other dimensions of work
that contributeto careersatisfaction.In teachers'conception of career,an
emphasis on the quality of professionalexperienceoutweighsopportunities for promotion (Bennet, 1985; McLaughlin& Yee, 1988; Yee, 1986).
Recent developments in the organizationaltheory literatureoffer an alternative orientation towardcareer,one that takes its point of departure
from the work itself and the social identities of the personswho do it. By
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
334
Reviewof Researchin Education,16
this conception, teachers are considered members of an "occupational
community" who
weavetheir perspectiveson workfrom the existingsocial, moral,physical, and intellectual
of workandcareerarecastin termsof
of theworkitself.Individualassessments
character
one's gettingbetter(or worse)at what one does, gettingsupport(or interference)from oth-
ers,exertingmore(orless)influenceoverthenatureof one'swork,andso on.(VanMaanen
& Barley, 1984, p. 289)
This perspectiveon careersuggestsa broadenedtreatmentof retention,
one which encompasses not merely teachers'decisions to leave teaching
or remainin the classroombut rathera wider view of sustained(or diminished) engagementin teaching.Teachersmay stay in teachingor leave it;
they may leave temporarilyand then return;they may reinforcetheir enthusiasmsfor the classroomor steadilywithdrawtheir laborover time, effectively retiring on the job. Hart and Murphy (1989a), following
Bluedorn's(1982) model ofjob turnover,proposea view of retentionthat
encompasses not only decisions to leave teaching or remain, but a wider
spectrum of attitudes, decisions, and choices regardingcommitment. A
similarly broad perspectiveunderlies Ruskus's(1988) analysis of teachers' orientationstowardtheir work.Ruskusdistinguishesamong teachers
who have actuallyleft education or who voice intent to leave ("attriters"),
those who leave the classroom for other positions in education ("climbers"), those who are simply putting in time ("lifers"),and those whose
enthusiasmand commitment remain high ("stars").By applyinga broadened conception of retention, researchersare able to explore the career
consequencesof teacher leadershipin ways that are tapped inadequately
by concrete decision points alone. In principle, the models that reflect
such expanded perspectives retain greater sensitivity to the actual patterns of occupationalparticipationthat teachersexhibit. At present,however, there are no studies that compare retention of mentors or other
programparticipantsto system ratesdescribingthe retentionor attrition
of classroom teachers. These include rates for denial of tenure, involuntary layoff, voluntaryattrition, promotion, dismissal, or retirement.Nor
are there studies that locate mentorshipin the ebb and flow of a teacher's
subjective career (Huberman, 1989).
Althoughactual data on retention of mentors are meager,the evolving
conceptual frameworkholds considerablepromise. It entails a shift from
a linear,sequentialconception of careerto one shaped aroundthe experience of teachers in teaching. It holds out a view of retentionthat extends
beyond decisions to leaveor stay,encompassingthe rangeof attitudesand
actions that make up commitment to teaching. Finally,it places the individual in an institutional and social context in which other factors (lay-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
335
in theteacher
offs,familyobligations)maydeterminedirectparticipation
workforce.
Powerof the CareerIncentivein Mentoring
The developmentand supportof formalmentorrolesis a substantial
policyinvestment.In California,for example,the budgetdevotedto the
mentorprogramrepresentsthe largestsingleshareof the state'scategorical staffdevelopmentfunding(Littleet al., 1987).Itsjustificationrestsin
largeparton whetherteachersfindthe roleattractive.Whatdo we know
aboutthe actualappealof the mentorrole,or aboutits powerto secureor
increaseteachers'enthusiasmfor teaching?
The rhetoricalliteratureis repletewithproposalsto policymakers
and
administratorsfor the timely developmentof teacherleadershiproles,
and with relativelysanguineassurancesof their appealto experienced
teachers(CarnegieForumon Educationandthe Economy,1986).Theresearchliteraturesuggestsa morelivelyinterplayof supportand opposition amongteachersthemselves.It hasnot escapedthenoticeof teachers'
organizationsthat the impetusbehindteacherleadershippositionsand
programsrestsoutsidethe teachingranks(see,e.g.,Cooper,1988).Some
criticsarguethatthe expansionof teachers'opportunitiesforcollaborative workhasbeenmatchedby a commensurate
increasein externalcontrol overthe substanceof teachers'work.Hargreaves
(1989)assertsthat
"teachersare beingurgedand sometimesrequiredto collaboratemore,
just at the pointwhenthereis less forthemto collaborateabout"(p. 29).
But such commentariesfail to accountfor the fact that some teachers
havein fact been activelyinvolvedin shapingsuchrolesand havebeen
eagerto applyforthem.So the questionremains,forwhomis the rolean
incentive?
Likemostincentives,the opportunityto becomea mentoris attractive
to someand not others.Unlikeotherincentives,however,the successof
mentoring(andthusthe fulfilledpromiseof theincentive)dependson the
directparticipationor tacitacceptanceof mentorsby otherteachers.The
titleof mentor,as BirdandLittle(1985)haveobserved,"name[s]halfof a
relation"(p. 3). Thus,the natureandextentof its appealto teachersis of
specialimport.As part of their broaderinvestigationof careerladder
plans in Utah schooldistricts,Hartand Murphy(1989a)assessednew
teachers'supportfor leadershippositions held by more experienced
teachers.The newteachers,all with 5 or feweryearsof teachingexperience,wererankedon the basisof gradepointaverage(academicrecord)
andprincipalratings(currentperformance)
to formthreegroupsof varying promisein teaching.In all but 2 of the 20 casesselectedforintensive
interview,the ratingsof academicpreparationandcurrentteachingperformancecoincided.The highestrankingteachersofferedthe greatest
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
336
Reviewof Researchin Education,16
support for differentiated roles, but also held out the most demanding
standards for them: clear links between the tasks of leadershipand the
core functions of teaching and learning,stable and clearly differentiated
opportunities (not short-term,marginal projects), and mechanisms for
monitoring effort and impact. By contrast, low-rankingteachers were
most likely to oppose or be indifferent towardthe new roles, or to base
their interest primarily on salary or on increases in teachers' power in
governanceand personnelmatters. To constructthe role in waysthat catered to the concerns of the low-performingteachers might broadenthe
overall base of supportin the short run, but almost certainlywould result
in losing the longer term participationand endorsementof preciselythe
target group of greatest interest to policymakers-the capable young
teacherswith high initial levels of enthusiasmand commitment.This evidence provides some grounds for advocatinga more assertivedefinition
of mentor roles. Missing from the equation so far,however,are the comparableorientations of high-performingand low-performingteachers at
more advanced career stages.
Individuals'careerhistories and careeraspirationsmay influence their
decisions to pursue or avoid mentor positions. Researchon teachers'careers has graduallyabandoned its nearly exclusive attention to the first
yearsof teachingand has begunto sketchthe outlines of certainmodal career cycles. The resultshave special significancefor the study of formally
devised careerincentives, and particularlyfor incentives that carrywith
them certain professionalobligations.The 160 Swiss secondaryteachers
in Huberman's(1989) 4-yearstudy of teacher careersexperienceperiods
of engagement and disengagement, confidence and self-doubt, experimentation and retrenchment.These are patterns that change with time
and circumstance.What teachers consider an incentive seems likely to
vary with these fluctuations in experienceand enthusiasm.Of particular
interest here are those teachers in mid-career(7-18 years experience),
whom Hubermancharacterizesas entering a period of experimentation
and activism;such teachersmay constitutethe most receptiveand appropriate pool of applicants for mentor positions. On the other hand,
Hubermanhighlightsthat same period as one in which as many as 40%of
teachers are specially prone to the kind of mid-careercrisis that may
prompt them to abandon the classroom. Would assuminga mentor role
relieve such a crisis, or only exacerbateit? Ruskus (1988) distinguishes
among mentors on the basis of their present orientation toward career.
Among 12 mentors, only 4 professed to be firmly committed teachers.
For 3, the mentor role served as one step along an intended path to administration.Of the remaining5, 1 left teaching duringthe period of the
study, and the remaining4 expressedeither intent to leave or a tenuous
commitment to teaching. This profile of a small mentor population belies
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
337
the assumptionthat mentors are drawnexclusivelyfrom a pool of professionally ambitious and entrepreneurialteachers.Ruskusdoes not explicitly analyze the effect of mentorship on the enthusiasm, indifference,or
alienation that mentors expresstowardteaching.Her implicit conclusion
is that assuminga mentor role reinforcescommitmentswherethey exist,
but does not moderatecareerdisappointmentsor dissuadeteachersfrom
leaving. Other case examples lend credence to this interpretation.
When career histories are joined by other aspects of teachers'life history and local context, accountingfor the differentialappeal of the mentor role becomes yet more complex. Because much of the work of
mentoringis added on to the school day,teachers'interest in becoming a
mentor may be contingent on the number and intensity of other obligations, both in and out of school. Family obligations or community involvementscompete with school demands.Evenwithin the school, active
participationin student activities, curriculumcommittees, or other outof-classroomactivity limits the availabilityof some of the most energetic
teachers. For teachers who find themselves already stretched thin, preferredincentives are likelyto be those that ease the burden.Embeddedin
the anecdotes collected in the implementation literatureare alternative
scenarios-preferences for increases in base compensation, more generous allotments of in-school preparationtime, fewercourse preparations.
Asked if the mentor opportunitywould hold him in teaching,one teacher
responded,"I hope so. I can't guaranteeit, though. It doesn't solve most
of my personalconcernsabout being a teachersuch as low public esteem,
low salary.... It doesn't do anythingto solve these problems"(Hanson et
al., 1985, p. 28).
Just as teachers' present circumstancesand future aspirations shape
their responseto the mentorrole, so past disappointmentsmaycolortheir
view. Teacherswho have been thwartedin more conventionalcareerpursuits may avoid any situations that requireteachersto compete for career
rewardson the basis of performance;it seems unlikely that "the embittered Mr. Pickwick,"having failed to secure the administrativepost he
desired, would find mentorship an attractive prospect (Beynon, 1985).
Whatever the factors that enter into teachers' initial response to
mentoringas a careerincentive, the eventualpower of the role lies in the
experienceit offersto the mentorsthemselves.Wherementorsfail to reap
personal and professionalbenefits from their work(Zey, 1984), or where
the risks associated with mentoringoutweighthe rewards(Yoder,Adams,
Grove, & Priest, 1985), mentoringis inhibited. On the basis of presentresearch,the benefits seem far from certainand the risks substantial.Some
anecdotalaccounts,to be sure, supportproponents'claims. The vignettes
constructed by Hanson et al. (1985) from interviews of mentor-teacher
pairs are one example. In these vignettes, mentors detail the intrinsic re-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
338
Reviewof Researchin Education,16
wardsassociatedwiththeirnewrole.Of 11mentors,8 stressedthe satisfactionsassociatedwithhelpingothers.Twoof the 8 arguedthatthe formal mentorrole createda legitimatemechanismfor sharingideas and
materials,and2 otherstookpleasurein receivingcompensationforwork
theyhadpreviouslyvolunteered.Othersderivedintellectualstimulation
fromtheirnewassociationwithothermentorsorteachersandfromtheir
expandedparticipationin conferencesandworkshops.Only2 presented
the mentorpositionas partof a clearlydefinedagendafor careeradOne acknowledges,
vancementinto the ranksof administration.
I needthe experiencebecauseI wantto be an administrator.
BasicallyI'mbuildinga re-
sume..... The money is nice but it's not my real incentive. I am not just a teacherwho is
going to stay a teacher.I have aspirations.(p. 32)
Butthe benefitsareby no meansself-evidentor uniformlyaccessible.
Teacherswho reporthavingbeenmentoredinformallyby peersfoundit
hardto imaginewhatbenefitthe mentorsderivedfromthe experience
(Gehrke&Kay,1984).Althoughinitiallyhonoredbytheirselection,mentors in formalprogramssubsequentlyexperienceconsiderableambivalence,uncertainwhetherthe mentordesignationis a blessingor a burden
(Allen,1989;Hart,1989;Shulman&Colbert,1987).Thereis widespread
evidencethatteachersmayexperienceanunwelcometransitionfromsuccessfulclassroomteacherto failedmentor.Someof the basicconditions
the rolealsoundermineit. Thetimedemandsaloneenergize
surrounding
some, but exhaustmanyothers(Hart,1988).Problemsof role stressambiguity,conflict,and overload-take theirtoll on teachers'commitment and performance.Mentorsmay be subjectedto the disdainand
censureof colleagues,andfindthemselvesforthe firsttimehavingto accountpubliclyfor theirperformance
(Bird,1986;Hart,1989).The relation betweenformal mentorsand individualteacherstends to be a
short-termaffair,offeringscantopportunityforthe essentialfeaturesof
the relationto mature(Gray& Gray,1985).Andteachers'owntenurein
positionsof mentorshipitselfis typicallylimitedto a periodrangingfrom
a fewmonthsto 3 years.The shorterthe period,the moredeleteriousthe
effects on mentors'own classroomperformanceand the fewer the
achievementsin whichthey mighttakepride.
The powerof the mentorroleto serveas a careerretentionincentiveis
furtherenhancedor dilutedby the immediatecontextsin whichmentors
attempttheirwork.Thepowerof contextis glimpsedin smallvignettesof
mentors'work(Allen 1989;Shulman& Colbert,1987),but is analyzed
morethoroughlyin full-scalecase studiesinformedby theoreticalperspectiveson work redesignand role innovation(Hart, 1989; Hart &
Murphy,1989b).The casestudiesgeneratedin the wakeof the Utah ca-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
339
reer ladder experiment, although not focused specifically on mentor
roles, examine a set of circumstancesthat closely parallelthe implementation of mentor initiatives in other locales. Cross-siteanalysesilluminate
some of the structuraland culturalfeaturesthat can be expectedto distinguish a professionally rewardingmentor experience from one that produces only anxiety and frustration.Two schools studied in depth by Hart
(1989) were in many respects comparable environments for classroom
teaching, with similar student populations, facultycomposition, and material resources.They were, however,radicallydifferentenvironmentsfor
the introduction of new teacher roles and altered professional relationships. In one school, where norms favored mutual support and problem
solving, teacherleadersjoined other teachersand the principalto fashion
their new roles in the service of widely shared school goals. Leadership
tasks were linked demonstrablyto improvements in teaching and learning, and communication about both efforts and progress was frequent
and public. A second school left its newly assigned teacher leadersto invent their own roles in relation to a faculty whose members jealously
guarded their professional prerogatives.Teacherleaders in both schools
suffered a certain degree of personal role conflict and overload, but only
in the latter school was personalstruggleto learn a new role compounded
by stresses generated by faculty opposition, faculty pressureto account
for their actions, and persistent ambiguity regardingthe main purposes
guiding their work.The Utah findings are echoed in other studies. Three
teacher leaders who were interviewedand observed in their capacities as
leaders (Wasley,1989) "all mentioned that their greatestchallengewas to
break into the school culture"(p. 7). Althoughonly one of the three leadership cases approximatesin purpose and practice the role of a mentor,
Wasleyconcludes, "Thesecases suggestthat each teacherleadershipposition is firmly rooted in its own context and that context is critically important to the success of the role" (p. 27).
Finally,the incentive powerof a new role is compromisedto the extent
that its present legitimacy and future stability are in doubt. Teachersexperience the relative stability or instability of the incentive in two ways:
continuity in an individual's access to the role, and the continued existence of the formal role within the system at large(Hart, 1988). From an
individual perspective, mentor roles constitute a small opportunity
base-a scarceresource.California'sMentorTeacherProgram,for example, funds a maximum of 5%of a district'steachers as mentors (Wagner,
1985). Only a relativelysmall percentageof teacherscan occupy the roles
at any one time, thus putting pressureon the system for short-termrotation of opportunities to expand the direct benefit to the largestpossible
pool of teachers. For the individual teacher,however,the powerof the incentive is plausiblylinked to one's prospectsfor getting it, and one's abil-
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
340
Reviewof Researchin Education,16
ity to remain in the position long enough to derive both its intrinsic and
extrinsic rewards(and to offset the strainsassociatedwith learninga new
role). Hart(1988) reportsthe shiftingof views when one teacherwho had
argued previously for frequent rotation of opportunitybegan to see the
merit of longer assignments:"all the training and the work the teacher
leaders have been throughwould be lost if we turn it over so quickly....
We need more stability in the position" (p. 24). Anotherprotests, "Acareer ladder's not a ladder if you fall off it automaticallyevery year"(p.
24). Most of the mentors interviewedby Hanson et al. (1985) asserted
their intent to remain in teaching regardlessof their future success in
competing for mentorships,but therewereexceptions.One teacherspecifies clearlythat he "will stay in the professionas long as he is mentor"(p.
46). In both studies, for at least some teachers, the association between
leadership opportunities and long-term career commitments was tied
closely to issues of individual access. Individualinterestsare thus in tension with system imperatives.
As a careerincentive, then, mentorroles appearto have differentialappeal among individuals, differentialpowerto affect retention compared
to other incentives,and differentialsignificanceundervaryingcontextual
conditions. Furthermore,the power of the incentive can be expected to
wax and wane as individual circumstancesevolve, as the configurationof
other incentives and disincentives shifts, and as elements of context yield
greateror lesser support for mentors' work.
CONCLUSION
Mentoring among teachers in American schools has been spurredby
public and professional debate over the quality of the work force, the
vigor of the teaching occupation, and the conditions of improvementin
schools. The proliferation of mentor programs results not from a
groundswellof teacherinterest,but is largelya productof policy interests
and institutional concerns. Increased public attention to certification,
tenure decisions, and teacher evaluation has driven the development of
formal mentor roles. Much of the research,in turn, has taken the form of
policy studies or program evaluations conducted in sites and settings
shapedby formalintervention.In local schools, mentorsfulfill threebasic
functions:They are guides to beginningteachersduringa period of induction; they form a local cadre of staff developers or teacher consultants;
and they lead or supportprogramand curriculumdevelopmentventures.
Of these, teacher induction programsprovide the main setting in which
the promise of mentoring has been tested. In all of them, however,the
logic of help giving dominates. It is in this regard,primarily,that mentoring among teachers departs from traditions of informal mentoring in
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
341
business and industry,wherecareeradvancementis the drivingforce and
the main source of rewardsfor both mentor and proteg6.
On the whole, researchhas been slow to pursuesome of the largerquestions implicit in the choice of mentoring as a favored policy option for
supplyingcareerretention incentives to experiencedteachersand for expanding professional support in schools. There are few comprehensive
studies, well informed by theory and designed to examine in depth the
context, content, and consequencesof mentoring.But the themes that run
throughsmallerstudies prove remarkablyconsistent. Froma rangeof discrete investigations,we can piece together a picture of the emergenceof
formal mentor roles through the implementation of local and statesponsored programs. We can begin to test the instrumental power of
mentoring to relieve the reality shock associated with teacher induction
or to stimulate innovations in curriculumand instruction. We also can
begin to assess the incentive powerof the role by finding whom it attracts,
and why, and what rewardsthey find in the role over time.
Attemptsto introducementoringrelationsinto the formalstructuresof
schools and districtsdisplaya markedconservatism.Formalprogramsreflect persistentpressuresto narrowthe definitions of the mentor role, accommodating(and thus helpingto preserve)traditionalnorms of privacy
and equal status. In the face of uncertainty,districts and schools have
sought bureaucraticsolutions to problems of professional relationship,
employingjob descriptions, selection criteria, and the regulationof opportunity to diminish problematic implications of the mentors' greater
expertise, maturity,and status. From many of the case study scenarios,
one is left with the sense that the problemssurroundingthe emergenceof
mentor roles are conceived as problems of a programto be marketed
ratherthan as problems of a culture to be built.
The conservativetenor of implementationis reinforcedwherethe purposes of mentoring remain ambiguous, where compromises are made
with regardto selection, and where mentors'opportunitiesto earn teachers' respectare diminished by constraintson time and visibility. Mentors
are inhibited further in their claims to special expertise and special
status-claims that are inescapablyimplied by their title-by the relative
scarcityof favorableprecedentsfor leadershipon mattersof professional
practice. In their efforts to fulfill their obligations, mentors encounter
both generalculturalambivalenceabout help giving and specific occupational prohibitions regardinginterferencein others' work. In effect, they
must engagein a precariousform of improvisation,writingthe scriptand
performingthe play at one and the same time for an audiencewhose sympathy is far from certain. There is a certain poignancy in the portraitsof
mentors' work.
Formallystructuredmentoringamong teachers, by comparisonto the
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
342
Reviewof Researchin Education,16
images evoked in the tale of Mentor and Telemachus(or other famous
mentors and prot6g6s),tends to be a narrowlyconceived affair with narrowlyutilitarianpurposes.The featuresof mutualityand comprehensiveness, distinguishing marks of genuine mentor relations, are hard to
detect. Pieced together,mentors' and teachers'accounts add up to a picture of a formal role much diminished in substanceand staturefrom the
one reflectedin our broadculturalimages.The relationsbetweenmentors
and teachers,on the whole, stress matters of comfort over issues of competence. They provide socioemotional support but appearto exert little
influence on teachers'thinking or performance.Teachersare more likely
to credit mentorswith providingmoral supportor enlarginga pool of material resourcesthan with exerting direct influence on their curriculum
prioritiesor instructionalmethods. In the end, these relationsappearless
mutually respectfulthan simply mutuallyreticent. The blatant disparity
betweenthe promise of the title and the patternsof practiceled a teacher
in one study to lament (even while crediting a mentor's assistance),
"Whereis the real mentoring?!"(Hanson et al., 1985, p. 27).
Though there are some significant exceptions in the case literature,
mentors more often are constrainedthan enabled by the organizational
circumstances in which they work. Some of these circumstances lend
themselves to policymakingand bureaucraticcontrol;others do not, and
are moreproperlythe object of leadershipthan rulemaking.To the extent
that ambitious, assertiveconceptions of the mentor role have been legitimated and defended, however,mentors appear more likely to engage in
the kinds of relationsand activities that one might, by common sense, associate with mentoring.Where more limited conceptions prevail, the activities and relations approximatefamiliar constructions of extra work
for extra pay.
The promise of the mentor role rests in its ability to attractthose teachers whose professionalrecordis highly regardedand who thus are able to
secure the admirationand acceptanceof other teachers.The powerof the
mentor role to serve as an incentive to career retention and enhanced
commitment has receivedfar less attention in the researchliteraturethan
its more instrumentalaspects, despite the prominentattention to career
incentives in the policy rhetoric.The major gains have been conceptual
ratherthan empirical.Theorists have recastretentionto include not only
concretedecision points or events (to leave or to stay), but a long-termset
of attitudes and actions by which commitment is enhanced,sustained,or
eroded. Studentsof teachers'lives and careersshow howthe concernwith
retention might be located in a still broaderconception of teachers'careers. And theories of work redesignand role innovation place questions
of careerincentive in the context of the relation between individual and
institution.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
343
The empirical gains are fewer. Anecdotal evidence, threaded through
the case studies, suggestssome preliminary,highlytentative, conclusions.
The attractivenessof the role and thus the incentive to compete for it in
the first place appearsto be a function of both individual careerorientation and organizationalcontext. The effect of the incentive is bolsteredto
the extent that teachersare able to match their images of the role with the
opportunities they actually encounter and the responsesthey meet from
teachers and administrators. It is diminished, predictably, when the
stresses of the new role outweighits rewards-a not uncommon development, it appears. Although cast as a career incentive for experienced
teachers and a resource for schools, mentorshipsturn out to place individuals in a personallyand organizationallyprecariousposition. The mutual benefits standard is met only with considerable difficulty.
The researchon mentoring reflects its pragmaticorigins. Policy interests and programmaticconsiderations have dominated; simple restatements of policy rationales have generally substituted for more clearly
articulatedand robusttheoreticalperspectives.A more rigoroustheoretical base is clearly available. Implementation studies have employed
theories of work redesign and role innovation to account for the emergence of mentor roles and for the particularformthey have assumed.Similarly, research on the contributions of mentoring to teacher induction
will be enriched by advances in researchon help giving or by theoretical
perspectives on socialization into occupations, organizations, and
groups. To graspthe significanceof mentor roles as careerincentives will
requirethat we locate mentoringopportunitieswithin a broaderperspective on teachers' lives and careers.
Added theoretical rigor brings certain methodological demands. The
characteristic limitations that Speizer (1981) associates with studies of
mentoring in business and the professions apply equally in education.
Among the characteristic limitations are small sample sizes, an
overrelianceon retrospectiveaccounts, the absenceof control or comparison groups, and the scarcity of longitudinal designs. Although many of
the availablestudies of mentoringin education employ multiple sites, the
numberof sites generallyremainssmall, and there is little evidence of systematic variation in those contextual features most likely to affect outcomes. Relatively few have been fully conceived and analyzed as
comparativecases adequateto the underlyingquestions of theory,policy,
and practice. There are virtually no structuredstudies that compare formal mentor arrangementswith the conditions, contexts, dynamics, and
consequencesof naturallyoccurringmentor relations.Nor are there studies that compare mentoringto other policy alternativesin teacher induction or in the domain of career incentives. Most studies are crosssectional, concentrated on the early stages of program implementation
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
344
Reviewof Researchin Education,16
and role developments. Many of the crucial questions surroundingthe
emergenceof the mentorrole, its nature,and consequences,cannotbe addressed without longitudinaldesignsthat distinguishbetween short-term
and long-termeffects on individuals and institutions.
The characteristiclimitations of small samples (an inevitability in the
study of teacher leadershippositions) might be compensated more persuasively by other aspects of research design. Sampling and selection
strategies,for example, only rarelyaccount for the web of social and professional relations in which mentors attempt their work. Designs that
sample mentor-teacherpairs offer greaterpower,though sometimes they
are weakenedby a selection bias introducedwhen mentorscontrolthe selection of teachersto be interviewed.Whenthis occurs,the sample is disproportionately composed of successful pairs. Other anecdotal and
surveyevidence suggeststhat the experiencereportedby such pairsis not
typical. The problemsof a small sample are compoundedfurtherby limitations on sources and types of data. Althoughthe pool of case materials
has grown steadily,permittinga more systematicexamination of the actual circumstances and practices of mentoring, the available evidence
often lacks credibility.Most studies rely heavily on in-depth interviews
that reveal mentors' perceptions, but also are constrained by mentors'
perspectives and experiences. The perspectives of teachers at large,
teacher prot6g6s in particular,or administratorsare representedmore
sparingly.Observationsof mentors' work are rare in study designs, and
rarerstill in published reports. Nonetheless, the sheer scale of practical
experimentationwith mentor roles suggests that methodological remedies, like theoretical sophistication, are well within reach.
This review has been constructednot only to assess and organize the
available research,but also to shape an agenda for subsequent research
and professionaldebate. Debates over the meaningof mentorshipin education derive in part from a Westernculturallegacyin which the name of
Mentor signifies wisdom, maturity,and a personal investmentin the capacities and fortunesof the prot6g6.And, on a more contemporaryfront,
they derive from an implicit comparison to perceived parallelsin business and industry, where mentorship is first and foremost a form of
sponsorship,a mechanism by which promising candidates are groomed
for the ranksof management.The specific meaning of mentoringamong
American elementaryand secondaryteachers has only begun to emerge
from a handful of comprehensive implementation studies and from a
largerarrayof small-scaleprogramdescriptionsand programevaluations.
This reviewbeganwith a conundrum:how to accountfor the rapidlyescalatingpopularityof mentoringin an occupationthat provides few precedents for formal and legitimate leadership by teachers on matters of
professionalpractice. In many respects,the puzzle remains to be solved.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
345
Woventhroughoutquite diverse inquiriesis a persistentambiguityabout
the meaning of the very term of mentor, and a certain skepticism that
mentor relationshipsat their richestcould be achievedby formalarrangement. Yet the twin aims of formal mentor programs-to rewardand inspire experienced teachers, while tapping their accumulatedwisdom in
the service of teachers and schools-contain the elements necessary to
satisfy the criterionof mutual benefit that sustain practicesof mentoring
elsewhere.That standardof mutual benefit seems a worthypoint of departure for researchand for practice.
REFERENCES
Allen, J. R. (1989). Mentorsas misers:The hiddencost ofsharing expertise.Paper
ResearchAssocipresentedat theannualmeetingof theAmericanEducational
ation,SanFrancisco.
Allen, J. R., & Pecheone, R. (1989). Making mentoringmatter:An evaluationof
Connecticut'ssupportprogramfor beginning teachers.Paper presentedat the
annual meetingof the AmericanEducationalResearchAssociation,San
Francisco.
Anderson,E. M., & Shannon,A. L. (1988). Towarda conceptualizationof
mentoring.Journal of TeacherEducation,34(1), 38-42.
Baltzell, D. C., & Dentler,R. A. (1982). Local variationsin the selectionof school
Paperpresentedat theannualmeetingof theAmericanEducational
principals.
ResearchAssociation,New YorkCity.
Artteachers'attitudestowardtheir
Bennet,C.(1985).Paints,pots,orpromotion?
andcareers(pp.120careers.In S. J. Ball&I. F.Goodson(Eds.),Teachers'lives
137).London:FalmerPress.
Berliner,D. C., & Carter,K. J. (1986). Differencesin processingclassroominformation by expertand noviceteachers.Paperpresentedat the meetingof the In-
ternationalStudyAssociationon TeacherThinking,Leuven,Belgium.
Berman,P., & Gjelten,T. (1984). Improvingschool improvement:A policy evaluation of the California School ImprovementProgram (Vol. 2). Berkeley,CA:
Berman,WeilerAssociates.
Berman, P., & McLaughlin,M. W. (1978). Federalprogramssupportingeducational change, Volume 8: Implementing and sustaining innovations. Santa
Monica,CA:RandCorporation.
Beynon,J. (1985).Institutionalchangeandcareerhistoriesin a comprehensive
livesandcareers(pp.158school.In S. J. Ball&I. E Goodson(Eds.),Teachers'
179).London:FalmerPress.
Biddle, B. (1979). Role theory:Expectations,identities,and behaviors.New York:
AcademicPress.
for
Bird,T. (1986).Thementors'dilemma.SanFrancisco:FarWestLaboratory
EducationalResearchand Development.
Bird,T., & Alspaugh,D. (1986). 1985surveyof districtcoordinatorsfor the Califor-
for Educania MentorTeacher
Program.SanFrancisco:FarWestLaboratory
tionalResearchand Development.
Bird,T.,& Little,J. W.(1985).Fromteacherto leader.SanFrancisco:FarWest
for EducationalResearchand Development.
Laboratory
Bird,T.,St.Clair,G., Shulman,J., & Little,J. W.(1984).Expandedteacherroles:
Mentorsand masters.(Interimreportto the NationalInstituteof Education,
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
346
Reviewof Researchin Education,16
ContractNo. 400-83-0003). San Francisco:Far West Laboratoryfor Educational Researchand Development.
Bluedorn,A. C. (1982). The theoriesof turnover:Causes,effects,and meaning.In
S. Bacharach(Ed.), Researchin the sociology of organizations(Vol. 1, pp. 75128). Greenwich,CT: JAI Press.
Bridges, E. (1986). The incompetentteacher.London: FalmerPress.
Brophy,J. E., & Good, T. L. (1986). Teacherbehaviorand student achievement.
In M. C. Wittrock(Ed.), Handbookof researchon teaching(3rd ed., pp. 328375). New York:Macmillan.
Brzoska, T., Jones, J., Mahaffy,J., Miller, J. K., & Mychals, J. (1987). Mentor
teacherhandbook.Portland,OR:NorthwestRegionalEducationalLaboratory.
Bucher, R., & Stelling, J. G. (1977). Becomingprofessional.Beverly Hills, CA:
Sage Publications.
Buchmann,M. (1986). Teachingknowledge:The lights that teacherslive by. In J.
Lowyck(Ed.), Teacherthinkingandprofessionalaction:Proceedingsof the third
InternationalStudy Association on TeacherThinkingconference(pp. 2-16).
Leuven, Belgium:University of Leuven.
Carnegie Forum on Education and the Economy. (1986). A nation prepared:
Teachersfor the 21st century.New York:CarnegieCorporation.
Carter,K. (1988). Using cases to framementor-noviceconversationsaboutteaching. TheoryInto Practice,27(3), 214-222.
Clark, C., & Peterson, P. (1986). Teachers'thought processes. In M. Wittrock
(Ed.), Handbook of researchon teaching (3rd ed., pp. 255-296). New York:
Macmillan.
Clawson, J. G. (1980). Mentoring in managerial careers. In C. B. Derr (Ed.),
Work,family and the career(pp. 144-165). New York:Praeger.
Cooper,M. (1988). Whose cultureis it, anyway?In A. Lieberman(Ed.), Building
the professionalculture of schools (pp. 45-54). New York: Teachers College
Press.
Darling-Hammond,L. (1988). Policyand professionalism.In A. Lieberman(Ed.),
Building the professionalcultureof schools (pp. 55-77). New York:Teachers
College Press.
Deutsch, M. (1982). Interdependenceand psychological orientation. In V. J.
Derlega& J. Grzelak(Eds.), Cooperationand helpingbehavior:Theoriesand research.New York:Academic Press.
Doyle, W. (1979). Classroomeffects. TheoryInto Practice, 18, 138-144.
Dubinsky,A. J., & Yammarino,E J. (1984). Differential impact of role conflict
and ambiguityon selected correlates.PsychologicalReports,55, 699-707.
Edgar,D. E., & Warren,R. L. (1969). Powerand autonomy in teacher socialization. Sociology of Education,42(1), 386-399.
Elbaz,F (1983). Teacherthinking:A studyofpracticalknowledge.London:Croom
Helm.
Elmore, R. (1989). Issues of policy and practice in mentorprograms.Paper presented at the annual meeting of the AmericanEducationalResearchAssociation, San Francisco.
Feiman-Nemser,S., & Buchmann,M. (1986). Knowing,thinking, and doing in
learningto teach:A researchframeworkand some initial results.In J. Lowyck
(Ed.), Teacherthinkingandprofessionalaction:Proceedingsof the thirdInternational Study Association on Teacher Thinking conference (pp. 602-627).
Leuven, Belgium:University of Leuven.
Feiman-Nemser,S., & Floden, R. (1986). The culturesof teaching.In M. Wittrock
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
347
(Ed.), Handbook of researchon teaching (3rd ed., pp. 505-526). New York:
Macmillan.
Fisher,J. D., Nadler,A., & Whitcher-Alagna,S. (1983). Four conceptualizations
of reactionsto aid. In J. D. Fisher,A. Nadler,& B. DePaulo (Eds.), New directions in helping (vol.1, pp. 51-84). New York:Academic Press.
Galvez-Hjornevik,C. (1986). Mentoringamong teachers:A review of the litera-
ture.Journalof TeacherEducation,37(1),6-11.
Gehrke, N. J. (1988a). On preservingthe essence of mentoring as one form of
teacherleadership.Journalof TeacherEducation,39(1),43-45.
Gehrke, N. J. (1988b). Towarda definition of mentoring. TheoryInto Practice,
27(3), 190-194.
Gehrke, N. J., & Kay, R. S. (1984). The socialization of beginning teachers
through mentor-prot6g6relationships. Journal of TeacherEducation, 35(3),
21-24.
Gergen, K. J., & Gergen, M. M. (1983). The social constructionof helping relationships. In J. D. Fisher, A. Nadler, & B. DePaulo (Eds.), New directionsin
helping (Vol. 1, pp. 143-163). New York:Academic Press.
Glidewell,J. C., Tucker,S., Todt, M., & Cox, S. (1983). Professionalsupportsystems:The teachingprofession.In A. Nadler,J. Fisher,& B. DePaulo(Eds.),New
directionsin helping (Vol. 3, pp. 189-212). New York:Academic Press.
Gray,W. A., & Gray,M. M. (1985). Synthesisof researchon mentoringbeginning
teachers.EducationalLeadership,
43(3),37-43.
Greenberg,M. S., & Westcott,D. R. (1983). Indebtednessas a mediatorof reactions to aid. In J. D. Fisher,A. Nadler,& B. DePaulo (Eds.), New directionsin
helping (Vol. 1, pp. 86-112). New York:Academic Press.
Griffin, G. A. (1985). The school as a workplaceand the masterteacherconcept.
ElementarySchoolJournal,86(1), 1-16.
Gross, A. E., & McMullen,P. (1982). The help-seekingprocess.In V. J. Derlega&
J. Grzelak(Eds.),Cooperation
and helpingbehavior:Theoriesand research.
New York:Academic Press.
Gross, A. E., Wallston,B. S., & Piliavin, I. (1979). Reactance,attribution,equity,
and the help recipient. Journal of AppliedSocial Psychology,9, 297-313.
Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, M A:
Addison-Wesley.
Hanson, S., Shulman, J., & Bird, T. (1985). California Mentor TeacherPro-
in theOrchardUnifiedSchoolDistrict,1984gramcasestudy:Implementation
1985. San Francisco: Far West Laboratory for Educational Research and
Development.
Hardcastle,B. (1988). Spiritualconnections:Prot6g6s'reflections on significant
mentorships. TheoryInto Practice,27(3), 201-208.
A. (1989).Teachers'
and teachers'work:Issuesof time
development
Hargreaves,
and control. Paper presented at the annual meeting of the American Educational ResearchAssociation, San Francisco.
Hargreaves,A., & Dawe, R. (1989). Coachingas unreflectivepractice:Contrived
collegiality or collaborativeculture?Paper presented at the annual meeting of
the AmericanEducationalResearchAssociation, San Francisco.
Hart,A. W.(1988). Workrestructuring
effectson teachercareerplans.Paperpresented at the annualmeetingof the NorthernRockyMountainEducationalResearch Association, Jackson Hole, WY.
Hart, A. W.(1989). Rolepoliticsand the redesignofteachers'work.Salt LakeCity:
University of Utah, Departmentof EducationalAdministration.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
348
Reviewof Researchin Education,16
Hart, A. W., & Murphy, M. J. (1989a). New teachers'responsesto redesigned
teacherwork.Paper presentedat the annual meeting of the AmericanEducational ResearchAssociation, San Francisco.
Hart, A. W., & Murphy,M. J. (1989b). Workdesign whereit happens:Five comparativecases ofschools.Paperpresentedat the annualmeetingof the American
EducationalResearchAssociation, San Francisco.
Hollingsworth, S. (1989, forthcoming). Prior beliefs and cognitive change in
learningto teach.AmericanEducationalResearchJournal.
A. M. (1986).Somerelationships
betweenteachers'
careertrajectories
Huberman,
and school improvement.Paperpresentedat the annualmeeting of the American EducationalResearchAssociation, San Francisco.
Huberman,A. M. (1989). The professionallife cycle of teachers.TeachersCollege
Record,91(1),31-57.
Huberman,A. M., & Miles, M. B. (1984). Innovationup close.New York:Plenum
Press.
Huffman,J., & Leak,S. (1986). Beginningteachers'perceptionsof mentors.Jour-
nal of TeacherEducation,37(1),22-25.
L. (1988).A synthesisof researchon teacherinductionprograms
Huling-Austin,
and practices.Paperpresentedat the annual meeting of the AmericanEducational ResearchAssociation, New Orleans.
Jackson,P.(1968). Life in classrooms.New York:Holt, Rinehartand Winston.
Jackson,P.(1986). Thepracticeofteaching.New York:TeachersCollegePress.
Kanter,R. M. (1977). Men and womenofthe corporation.New York:BasicBooks.
Kennedy, M. (1987). Inexact sciences: Professional education and the development of expertise. In E. Z. Rothkopf(Ed.), Reviewof researchin education
(Vol. 14, pp. 133-167.) Washington, DC: American Educational Research
Association.
and implementation
Kent,K. M. (1985).Development
of a program
for training
mentor teachers.Unpublished doctoral dissertation,Nova University.
King,R. M. (1988).A studyof sharedinstructional
leadership
bymentorteachers
in SouthernCalifornia.Unpublished doctoral dissertation,University of San
Diego.
Kram,K. E. (1983). Phasesof the mentor relationship.AcademyofManagement
Journal,26(4),608-625.
Kram, K. E. (1986). Mentoring in the workplace. In D. T. Hall & Associates
(Eds.), Career developmentin organizations (pp. 160-201). San Francisco:
Jossey-Bass.
Lacey,C. (1977). The socialization of teachers.London: Methuen.
Lemberger,D. (1989). The mantle of the mentor.Unpublishedpaper,University
of California,Berkeley.
Levinson, D., Darrow,C., Klein, E., Levinson, M. H., & McKee, B. (1978). The
seasons of a man's life. New York:Ballentine.
Little, J. W. (1987). Teachersas colleagues.In V. Richardson-Koehler(Ed.),Edu-
cator'shandbook:
A research
perspective
(pp.491-518).NewYork:Longman.
Little, J. W. (1988). Assessing the prospects for teacher leadership. In A.
Lieberman(Ed.), Buildinga professionalculturein schools(pp. 78-106). New
York:TeachersCollege Press.
Little, J. W., Galagaran,P., & O'Neal, R. (1984). Professionaldevelopmentroles
andrelationships:
andskillsofadvising(Reportto theNationalInstiPrinciples
tute of Education).San Francisco:Far West Laboratoryfor EducationalResearch and Development.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
349
Little, J. W., Gerritz,W. H., Stern, D. S., Guthrie,J. W., Kirst, M. W., & Marsh,
D. D. (1987).Staffdevelopment
in California:
Publicandpersonalinvestment,
programpatterns,and policy choices (Report to the California Postsecondary
EducationCommission). San Francisco:Far West Laboratoryfor Educational
Researchand Development.
Little,J. W.,& Long,C. (1985).Casesin emergingleadershipby teachers:The
school-levelinstructionalsupportteam (Reportto the National Instituteof Education). San Francisco:Far West Laboratory.
Little, J. W., & Nelson L. (Eds.). (1989). Preparingmentorsfor workwith begin-
FarWestLabguidefor mentortraining.SanFrancisco:
ningteachers.
Program
oratory for EducationalResearchand Development and Los Angeles Unified
School District. Denver: ERIC Clearinghouse.
Lortie, D. (1975). Schoolteacher.Chicago:University of Chicago Press.
Macrorie,K. (1984). Twentyteachers.New York:Oxford University Press.
Malen, B., & Hart, A. W. (1987). Careerladder reform:A multi-levelanalysis of
EvaluationandPolicyAnalysis,9(1),9-23.
initialefforts.Educational
Martin,P. (1987).Beginningeducatorsupportandtrainingprogram:
Studyyear
evaluation.Hartford,CT: ConnecticutState Departmentof Education,Office
of Researchand Evaluation.
McAlpine, A., Brown, S., McIntyre,D., & Haggar,H. (1988). Student teachers
Scottish
teachers,SCREprojectreport.Edinburgh:
learningfrom experienced
Council for Researchin Education(SCRE).
McLaughlin,M. W. (1987). Learningfrom experience:Lessons from policy im-
EvaluationandPolicyAnalysis,9(2), 171-178.
plementation.Educational
McLaughlin,M. W., & Pfeifer,R. S. (1988). Teacherevaluation:Improvement,accountability,and effectivelearning.New York:TeachersCollege Press.
McLaughlin,M. W., & Yee, S. M. (1988). School as a place to have a career.In A.
Lieberman(Ed.), Building a professionalculturein schools (pp. 23-44). New
York:TeachersCollege Press.
Whodo teacherschooseas models
Mead,J. (1989).Lookingat oldphotographs:
and why?Paperpresentedat the annualmeeting of the AmericanEducational
ResearchAssociation, San Francisco.
Measor,L. (1985). Criticalincidentsin the classroom:Identities,choices, and careers. In S. Ball & I. Goodson (Eds.), Teachers'lives and careers(pp. 61-77).
London: Falmer Press.
in Maryland.
Meister,G. (1987). Currentpracticesin newteacherdevelopment
Philadelphia,PA: Researchfor Better Schools.
Merriam, S. (1983). Mentors and prot6g6s:A critical review of the literature.
AdultEducationQuarterly,
33(3), 161-173.
Miner, A. S. (1987). Idiosyncraticjobs in formalizedorganizations.Administra-
tiveScienceQuarterly,
32, 327-351.
Nemser, S. (1983). Learningto teach. In L. S. Shulman& G. Sykes(Eds.),Handbook of teachingand policy (pp. 150-170). White Plains, NY: Longman.
andAssessment
Support
Neufeld,B. (1986).TheBeginningTeacher
Programevaluation report.Cambridge,MA: Education Matters.
Nias, J. (1989). Primary teacherstalking. London: Routledge.
Nicholson, N. (1984). A theory of work role transitions.AdministrativeScience
29, 172-191.
Quarterly,
Odell, S. J. (1986). Induction support of new teachers:A functional approach.
Journalof TeacherEducation,37(1)26-29.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
350
Reviewof Researchin Education,16
Parkay, F. W. (1988). Reflections of a prot6g6. Theory Into Practice, 27(3),
195-200.
Parker,M. (1989). Teachingteachersaboutteaching:Mentorstalk abouttheir
work.Paperpresentedat the annualmeeting of the AmericanEducationalResearch Association, San Francisco.
Peterson, K. (1984). Methodologicalproblemsin teacher evaluation.Journal of
in Education,17(4),62-70.
ResearchandDevelopment
S. (1987).Learningthelanguageofpractice
teachers:
An
Pinnegar,
frompracticing
explorationof the term "withme."Paperpresentedat the annualmeetingof the
AmericanEducationalResearchAssociation, New Orleans.
Pruitt, D. G., & Carnevale,P. J. D. (1982). The developmentof integrativeagreements. In V. J. Derlega & J. Grzelak(Eds.), Cooperationand helpingbehavior:
Theoriesand research.New York:Academic Press.
Roche, G. R. (1979). Much ado about mentors. HarvardBusinessReview,57(1),
14-28.
Rosenholtz, S. J. (1989). Teachers'workplace.New York:Longman.
MentorTeacher
ProRuskus,J. A. (1988).A multi-siteevaluation
oftheCalifornia
gram. Unpublished doctoral dissertation, University of California, Los
Angeles.
Schein, E. (1978). Careerdynamics.Reading, MA: Addison-Wesley.
Schlechty,P. C. (1985). A frameworkfor evaluatinginductioninto teaching.Jour-
nal of TeacherEducation,36(1),37-41.
Shapiro, E. C., Haseltine, F. P., & Rowe, M. P. (1978). Moving up: Role models,
mentors, and the "patronsystem."Sloan ManagementReview,19(3), 51-58.
Shulman,J. H. (1987).Expertwitnesses:Mentorteachersand theircolleagues.
Paperpresentedat the annualmeeting of the AmericanEducationalResearch
Association, Washington,DC.
Shulman,J. H., & Colbert,J. A. (Eds.). (1987). The mentorteachercasebook.San
Francisco:Far West Laboratoryfor EducationalResearchand Development.
Shulman, J., Hanson, S., & King, R. (1985). California Mentor TeacherPro-
in the Waverly
gramcasestudy:Implementation
UnifiedSchoolDistrict,1984-
1985. San Francisco: Far West Laboratory for Educational Research and
Development.
Smith, K. E., & Sandler,H. M. (1974). Bases of status in four elementaryschool
faculties.AmericanEducationalResearchJournal,2, 317-331.
dimenSmylie,M. (1989). Teachers'
collegiallearning:Socialandpsychological
sions of helping relationships.Paper presented at the annual meeting of the
AmericanEducationalResearchAssociation, San Francisco.
Smylie, M., & Denny, J. (1989). Teacherleadership:Tensionsand ambiguitiesin
organizationalperspective.Paperpresentedat the annualmeetingof the American EducationalResearchAssociation, San Francisco.
Speizer,J. J. (1981). Role models, mentors, and sponsors:The elusive concepts.
Signs:Journalof Womenin CultureandSociety,6(4), 692-712.
Stateof ConnecticutDepartmentof Education.(1988). A coretrainingmanualfor
thepreparation
teachers
andmentorteachers.Hartford:
Author.
ofcooperating
the mentoringrelationship:
A questionof goals.
Stoddart,T. (1989).Structuring
Paperpresentedat the annual meeting of the AmericanEducationalResearch
Association, San Francisco.
Sykes, G. (1983). Public policy and the problemof teacherquality:The need for
screensand magnets.In L. S. Shulman& G. Sykes(Eds.),Handbookofteaching
and policy (pp. 97-125). New York:Longman.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions
Little:TheMentorPhenomenon
351
Thies-Sprinthall, L. (1986). A collaborative approach for mentor training: A
workingmodel.Journalof TeacherEducation,37(6), 13-20.
Van Maanen, J., & Barley,S. R. (1984). Occupationalcommunities:Cultureand
control in organizations.Researchin OrganizationalBehavior,6, 287-365.
Veenman,S. (1984). Perceivedproblemsof beginningteachers.ReviewofEducational Research,54(2), 143-178.
Wagner, L.A. (1985). Ambiguities and possibilities in California's Mentor
Teacher Program.EducationalLeadership,43(3), 23-29.
Wasley,P.A. (1989).Leadteachersand teacherswholead:Reformrhetoricand
real practice. Paper presented at the annual meeting of the American Educational ResearchAssociation, San Francisco.
as a career:Promotionversusdevelopment.
Yee,S. M. (1986).Teaching
Stanford,
CA: Stanford University School of Education.
and
Yinger,R. (1987).By theseat of yourpants:An inquiryinto improvisation
teaching. Paperpresentedat the annual meeting of the AmericanEducational
ResearchAssociation, Washington,DC.
Yoder, J. D., Adams, J., Grove, S., & Priest, R. F (1985). To teach is to learn:
Overcoming tokenism with mentors. Psychologyof Women Quarterly,9(1),
119-131.
Zeichner,K., & Gore, J. (in press). Teachersocialization. In W. R. Houston, M.
Haberman,& J. Sikula(Eds.),Handbookofresearchon teachereducation(chap.
19). New York:Macmillan.
Zey, Michael G. (1984). The mentor connection. Homewood, IL: Dow
Jones-Irwin.
This content downloaded from 128.32.250.200 on Mon, 18 Nov 2013 14:33:31 PM
All use subject to JSTOR Terms and Conditions