ED 112.983

DOCUMENT RESUME

JC 750 5344

Carman, Robert' A.TITLE 'A Long-Term Study of the 'Effort Tutoring :in

Developmental athematicsi, -

INSTITUTION Santa Barbara C'ty.Coll.,NOTE 49p.

EDRS PRICE MF-$0.76 HC-$1:95 Plus PostageDESCRIPTORS Comparative Analysis; *Dropout Prevention;

Educationally,Disadvantaged; Grades. (Scholastic);Individual. Instruction; *Junior Colleges;Longitudinal Studies; *Low Ability Students;*Remedial Mhthematics; School Holding Dower; StudentAttitudes; *Tutoring

IDENTIFIERS , Santa Barbara City College

ABSTRACTLow ability freshmen enrolled in a 972 fall semester

remedial mathematics, course at Santa Barbara City College(California) were randomly divided into three groups: (1) a controlgroup comprised ,of 62 sudents meeting for one hour per week in small,

subgroups to study programmed material's; (2) a classroom tutoringgroup comprised 0 67 students meeting weekly in small subgroups,with class time divided betieen programmed Materials and tutoring bytrained, experienced tutors; and (3) a total tutoring grOU'ip comprisedof 61 students enrolled in weekly small group sessions like those forgroup two, but with all students receiving from,ong, to' five.additional weekly hours of tutoring.at Other times. At the ':end Of thesemester, each student was interviewed,using a structured interviewtechnique. Neither course grades nor GPAs differed signiii4ntly,.among the three groups. However, signifiCantly fewer students in thetutored 'groups withdrew from the course. In addition, the ittuctuxedinterviews showed that student attitudes in the tutored ItoOps .were

esignificantly more positive toward oth mathematics and all; othercourses. The pattern of increase rsistence and potitive attitudeof the tutored group continued, Ad ng the two years followingtutoring. Data are organized.inio charts and tables, and a.bibliography is appended. (NHM) 1

., .

******************************************************************MtS* Documents acquired by ERIC Include many informal,unpublis4ed* materials not available from other sources. ERIC mWkes every effort *to obtain the, best copy available. Nevertheless, items of marginal *

* reproducibility are often encountered and this affedts the quality *

* of the microfiche and hardcopy reproductions ERIC makes available *

* via the ERIC Document Reproduction Service (EDRS). EDRSNis.not* responsible for the qual0.ty of.the original document. ReproAuttions ** supplied by EDRS are the' best that, can be made from the ,orAginal. *********************************************************ssoopoom

4

c

41A LONG -TERN STUDY OF THE EFFORTSO. TUTORING IN DEVtLOPMtNTAL

b ,..MATHEMATICSc

ROBERT' A. CARMANSANTA BARBARA CITY.COLLEGE

US DEPARTMENCOP I.SEALTH,EOLICATIOTI&WELFARE,NATIONAL INSTITUTE OF

r" EDUCATIONTHIS DOCUMENT HAS BEEN REPRODUCED EXACTLY AS' RECEIVED.FROMTHE PERSON OR ORGANIZATION ORIGINAT ING IT POINTS OF VIEW OR OPINIONSSTATED 00 kfir NECESSARILY ,REPRESENT OFFICIAL NATIONBL InTrarreoF-EDUCATION POS$TiON OR POLICY

For more than twenty c ries,'possitily before thetime'of'Plato, tUtorinehas en hypothesized to be superior

other instructional methods. YetdespiVe this venerablet-history, tutoring as an educational variable` as been thesubject of remarkably littler formal study, due perhaps to the

ent impracticality of using tutors on any large scalebaSis. This scarcity of research,on tutoring is especially

. marked in highereducaton..

The 1rge majority of,the reseaVch.studies publphed on-'the effects of tutoring have been. concerned with thefelementaryand secondary school level. 'In An attempt to discern patterns

, that would be helpful in the development of tutor training'materials anh guidelines for matching tutors to students,,McClellan" reviewed the literature-omstUdenttlitoring andfound that most df\ the studies available Involved the tutoringof stuents at pre-college levels (19).- This finding wassubstantiated by Thelen (84) and by Riessman (68) in similar

. revieizs of tutoring activities and research. -

.''4/'Tutoring studie's at the elementary-secondary school .

level have focused primarily on tutoring. in reading andwriting skills. Tutoring is Mainly "cross-age" ore"inter-.

4^1

NQ

.

grade" in nature, employing older children to tutor younger, `1pw-ability children. There is subs

ude.s gpificant imp.rovement in achie. ent;,especially tf it is used to supplement conventional classroom,work. The evidence that ti4toring produces more positiveattitudes and S'elfoncepts.iin tutored children is contra-dictory and indomplete: The .changes in achieveMent andattitude of tutors participating in tutoring programs aresignificantly positive. Very few long-range or longitudinal.studies of the effects-of tutoring at the elementary-secondaryschool level -are' reported in the research literature.

Tutoring in 'Colleges

There are. two /major varieties of compensatory. .edUcationprograms in higher education: (1) those that assist educa-tionally disadvantaged or culturalay different students to t

enter institutions of higher education and (2) those thathelp&these students succeed in,academic and occupationalobjectives once/they have enrolled. Compensatory educationentry practices include modified recruitment apdmodifid ".tinancialgaid programs. As Knoe7,1 pointea out, in both kindsdf programs the purpose is to reach students who would not plan

I

'1

1

e .

s s

n6rmaIly to go on with their'edudation (42:6). The secondvariety of compensatory education program includes activitiesdesigned-to enable the educationally disadvantaged studentto succeed in conventional instructional situations.

Morrison and Ferrante pointed out that +

CollipensatOry-'programs are often developedwith'the\aSsumption that the'culturallysidifferent have academia'.abilities which are inadequate for the traditionalcollege classroom. In order to assist thebe students

-in realizing their potential, some'institutions havedeveloped special tutorial programs to support class-room activities. These programs have provided extensiveindividual tutoring in academic areas, with some schoolsdeveloping corps of tutors -which include teachers, gradu-ate students, regular'undergraduate students, and advancedstudents from culturally different Backgrounds. (61:8),

ThePrograms

In a study of fifty-three public and private two-yearcolleges conducted by the American Council on Education, it

was found that 91 percent reported some 'kind of specialtutoring prograin for academically disadvantaged students.More thin three-fourths of these colleges used regular studentsas tutors in their tutorial programs (6.1:30).' The use f

. .tutors is-increasing rapidly, especially in the two-year

7

s

Despite;the widespread occurrence of these programs,+studies of tutoring at the college leVel are very few.Tutoring Studies Mve been reported in the are of readingand writing skills (1, 8, 81, 86, 90, 92), study skills(25, 81), basic mathematics (90),, engineering. (10),_ 82, 83),

Psychology (35, and .in advanced courses (58). Tutorsare used both as supplements to conventionally taught, courses'and as the exclusive means of instruction.

The Tutors

The tutors employed-An college-level tutoring programsake usually either;faculty members or regularly enrolled' .

/ students._ Williams reported that tutors drawn from the same'so6io--econoinic,'ethnic, or racial background as the studentsbeing tutored are often employed in order"to provide rolemodels (88:196'). Fierce and Norell found that:black,tutorSwere especially effective with black students in terms ofthe perceived accessibility of the tutors (67:171) although--1no differences were fbund in either.achievement or attrition

rate

Reports regarding most college-level tut ing programs'

istate that heir major objectives are to provide (1) an ! .

increase. i achievement in the- tutored areal.(2 improxedstudy skil s, .0) enhanced self - esteem, (4) more positilleattitudes toward college, and (5) increased persi t nce incollege. The p.vailale evidenbe on achievement ns has notbeen sistently positive. Sykes (81) studied t e effects oftutoring, reading instruction, and financial ass stance uponstudent achievement, attitudes, and attrition .o communityco.11ege students enrolled ine special program for "disadvan-taged! students. Student achievement, as measured by grade-point average.; was not significantly improved .y the program.Attitudes and study habits as measured-by a s lf-image conceptsurvey (Brown-Holtzmon survey of Study .Habit and Attitudes)showed no consistent pattern Of Dhange. Red' ing skil b, asJpeasured by the Diagnottic Readirrg Tests, improved si nilfitantlybut appeared to have no effect on student aqhievement asmeasured by grade-point average. The study suggested thatthis program was somewhat effective in retaining college Istudents who might otherwise not remain in school. Unfokunately,the study was not designed to isolate a tontral confound-ing variable's so that the direct effec tutorin nAttrition 'might be determined. ....

,The Results..of Tutoring

Lucas, Gaither,.and Montgomery (48 evaluat d a tuto ial0 program .involving volunteer college fr shmeh su Jetts and a,

control group of non-attenders. Resul s indica ed that tutoringwas ineffective in raising grades.. T is study as confoundedby the difficulty of deciding whether an appro riate controlgroup would- be,non-volunteers aro/al teers w dropped outof the program. .Ewing and Gilbert ( 7) indic ted that theact Of volunteering for tutoring he can be more importantfactor in academic improvement than the tuto ing itself.

point av rages ,for tutoredy Progra s and Services

dents ear ed a grade-pointximum of 4, while non-de-point average of 2.37.tutorin alone could be

Evafis (26) cpMpare :the gradeand non-tutored Extende Opportuni(EOPS) students. Tutor d EOPS staverage of 2.54 on a sca e with mtutored EOPS students ea ned a grNo signifi t positive fleet' odetermined fr.= the study (26:54

In a similar;evalilatofi f Extended Opportunity 'ervicesMacMillan (51) failed to' Show a y,,signifii.cant difference irkachievement as measured inc eased gr de-point averages.

,Wright (90) examined the perfo mance, of althost 800 collegefreshmen who were predicted to perform nsuccessfully on thebasis of entrance testing and who rece ved tutoring by upper-/division students* in EngliSh, mathemat cs,/so ial science andscience. Results indicated thdt..tutor ng was an,effective

4 t

0

. .

means for raising achievement levels for many students.'Students predicted lowest in achievetent appeared to profit,least. from tutoring or did not take advantage' of it. Those ppredicted In the average.range appeared to profit most fromtutoring and participated most.

Taylor .(82) examined the effects of:a tutoring Iprogram.on engineering freshmen,. Thirty-onestudents participatedin the program and were matched by grade-point .average witha control group of non-tutored students. He found that studentswith a gradepo±nt average' below 2.00 were able to makeeffective'use of the tutorial program to obtain significantlyhigher grades. Furthermore, the term by term academic trendof both groups suggested that the ttitored students, progressesIsteadily upward over the.year-long-study while the controlgroup followed a downward trend. -

r.1

Ettersistudied.the effectiveness of a college tutorialprogram employing full -time bachelor's degree level tutors.It was found that tutored low achievers carrying leSs than af1.0.1 course' load were more successful in raising their grade-

.

pcant average than any other category of students (25:406).

While the evidence for the effectiveness,Of tutoring in-improving the academic achievement of college StudentS isoften ambiguous or negative, the evidence for its effectivenessin reducing attrition is very positive. Adams (1)1in aninformal,s,tudy of tutoring in'a community college rea"ding:pragram,-'.found,very pbs'itive evidence that tutoring led to.increased Persistende'in the program,,

In contrast to an annual withdraWal rate of more thanforty perbent in conventionally taught sections of the course,tutored groups had no withdrawals during a year-long study.

This finding ,ofddecreased withdrawal rates under tutor-was consistent with evajuations,dif other, personalized

ystems of instruction. Born and Hebert .(11)) found decreasedwithdrawal rates and increasediattendance under a scheme of::ihstiruatl.bw. in psychology in which students, progressed at -

.1their'own,rate through the coUrse with the assistance of tutors'i or proctors. ;

4 In MacMillan's study of Educational_Oppo*unity students,`85 percent of those tutored *persisted in college, compared-with a college-wide average o,f 74 percent (51). 'Sykes (81)in his study of similar 16w-ability students found that tutor-

. ing"had a significant positive affect on the persistence of 9.

students in the program.

rIn Taylor's study of tutored.engi ing freshmen, a

9,

:

NA

. .

significantly lower percettve,of tutored .students thanTeif-__

control group students wii:.hdrey during the first.year (83:90:---.Students in both the experimental and control grcwsjfereexceptionall(able students, ln sharp contrast to those studiedby Sykes (4) and MacMillan (51)f. In another program for.more able stvdents, Porter, and Joiner (86) employedpeer tutoring, in biology,' physici, psycholbe and social 4

science classes. Withdrawal rates in the tutored classeswere significantly lower than in non - tutored 6lagses heldthe prev ous year while the fraction of "A" and "B" gradesdid no- change significantly.

In a study of open enrollment students, Yuthas (92)matched groups on high school grade-lovint average, sex, andAmerican CollegeTesting entrance,examination,scores. Thegroups receiving tutoring in reading shbwed 'Stgnificant.improvement on the Nelson-Denny Reading Test. Colliparison

of persistence rates for.thefexperimentaland control groupsindicated a highly significatt_decrease in the withdrawal rateof tutored students.' Enrollment'in the tutoring nogram wassignificantly related to persistence' in college and o resis.-tance,to extraneous influences which normally lead to with7

, drawdl (92:234). As in most other atudies, however, it wasir6o9sible to separate 'the influence, of tutoring per, se fromhe. other instructional activities_involved in the tutorillgprogram.

rAlthough longitudinal studies the,.eSsfed,tszortutoring

;college students have been recommended O' Benz (8:12) and ,

Sykes (81:55), extensive review of the p:ublished.literatUrereveals, no long-term stuilies. The .lack of suchl studies ispaptictildrly.anfor'qunate in the light of 4despread fears atthe college level and evidence at'the secbnadiv level thattutcms tend to "teach to the test", and trade short-term

(gains in dOurse gains or test scores for deeper understanding(65:414):

Significant sand suggestive Long -rang c.sta0 0*involving tutors, Caither found that peirsitence And peinrm-ance.over a four-semester period were pot.Ogniffcantly ,

related .to enrollment in remedial rogiIgnis4: :Using a controlgroup of Students not enrolled in remedial program but ,

matched with the experimental group on'the preliminaracademic aptitude scores, Garther-T n thati

othe,aecte-mic (control) group outperfo'med the remedial; group in esiery semester. xcept the first semesterwhen the.remediai group ad the 'shelter' of beingin the remedial* program and receiving grades commen-surate with that special level (29,;10),

. Gaither found no difference in the,W.S groups with respect to.

r

4

,

$ .

attrlition rate over the two-year span-, .

'Gaither's study is especially significant for tbe,study,reported in this dissertation because it is one of the fewlongitudinal studie of .developmental progrAs at the comm-unity college level and because'it does not include tutoring'as -a variable.

Ari ,important. aasumption underlying the design of devel- .

opmental programs at 'the 011ege level id.s that they mustprovide the necessary ingredient to assure reduced withdrawalrates, improved academic performance;.and more positiveattitudes toward college. is the ,bias oi' many instructionaldesigners that peer tutoring is a yit41 ingredient in thesedevelopmental programs and that the ndgative findings ofCaither's study are kdirect.result of the lack of tutoring. ,

in the programs.he examined. All of the studies of tutoring.desOrtbed test, the effectiveness of tutoring over the'time span of a single semester or college course. k'hepresentstudy is designed to examine the ;e = ecXivehess of tutoringover a two-year period, a'substant'i l'porti6n of the college

,

7

career for. the students studied...

Low-Abilit Studen s in tip Communit

A number of studies provide a otfle of,the educational,1ydisadvantaged student in the commun y college, 'and thepicture indicates the importance of he'present study. Roueche(73) estimated that more than sevent percept of all communityOol.lege students need remedial assis ance, reading scones atthe fourth-grade level are common (2 ), withdrawal rates are

A. typically over'sixty percent (72). A

Cd e

of"the number of students in this categbry andthe severity of their problems, it is not,sUrprlsing that the

I most offered,courses in most community cOlieges are remedial-or_developmental in nature. Roueche estialated that more than,,twenty percentAbf'the average total intr'uctional budget isallocated io.remedial programs (74). According toMorrisonand Ferrante, specialized course and programs,for the educa-,tionally disadvantaged have been developed and are rapidlyincreasing in number, scope, and size (61:23), and many .

educators actively encourage' their development. Remedialreading and remedial mathematits courses are the most frequently.found remedial courses in the community college (6,55).

In a 1963 study, Schenz (75:.63)'estimated that more than.two-thirds af all community' colleges automatically requirelow-ability'Students to enroll in remedial classes, yet mostof these programp are based on faith rather than 'research,

' The evidence for the evidence forsthetffectiveness

ONO

. lir '1-..1/4-, , I .,...1. N I i . .7

.

_/...." ,..

developmental plograms in the community college'is contradic-tory and incomplete. Gaither found no net positive effectof remedial program$ in improving reading and wrl.ting skills ,

Or in reducing withdrawal rates (29). ,Sharon (77) studied teffectiveness of remedial programs and found trig the Englisremedial, course produced, only a small improvement in subsequent-performance in the r. gulag English course. Conve sely, theremedial course An At.hematics had a significant positive effecton subsequent course work in- mathematics, improvi g the averagecourse grade by more than half of a letter grade. Logak 7)found that*mmunity college remedial" programs in Ehglish andmathematicsgPoduee no significane difference in number of.vithdrawals and no significant increase in grade point averages.

'1 \ FlOue6. (73) found this lack oe effectiveness of remedial\. .courses to be widespreac. In many community college8 up to .

.,49 percent of entering freshmen enroll in remedial Courses inmathematics and English, and only 20 percent of phese'continueinto regular college level mathematics and Engliih courses.

.e

In a follow,;-up-suryey ofqUdents enrolled in severalChicago-area comM*ity colleges, Baehr (4) found that person=allzed remedial wcrk'significantly(inCO,ased the student's

.

expressed desire to persist .in college.- Heinkel.(36) compared122 students in a developmental program with a control grbupof 128 students eriolled in other programs. Males and ethnic4tinority students enrolled in the program withdrew from fewer-courses in the first' semester.than,did minority students, inthe control group. .

._

!

Addinistrators of most community colleges assume thatthe r remedial programs will _result in long-term improvementsin cademic skill's, attitudes, and performance of,the,educa,tionally disadvantaged student8 for whom they are designed,lioweV,er, the' few research studiSs designed to test these .

assumptions were based on a sher't-term, 'one semester follow-up of students enrolled. Clearly, success in college isrelated .to the ability of the student fo'persisi in collegeover a period of time long enqugh for the desired changesto take place.. Therefore, studies of the effectiveness ofthese programs in producing "Ong-range changes in persistenceand perforpance.are needed.

'Although most educators/agree on the importance of . is

improving:the self- concept ten- the educationally disadvAntagedstudent, very few studies of remedial pKograms have examinedthis variable, McDavidl% work (50) indicated that academicsuccess may result'An a mbre positive self-image and lead toincreased future academic success. Wessman (87) round a signi-ficant increase in self-ace ptance.aa measured by the CottellPersonality Inventory in udents completingHan eightg-weeksummer remedial program. Zirkel found .that -the self-concept

8 't";

'is significantly related' to academic aChievement'for bothdisadvantaged and normal students (93:211). The enhancement'of self-concepts is a goal of virtually all programs foreducationally disadvantaged students.

_?c_

The effects of negative self-image are particularlyimportant in'mathematics courses. Aiken folind that mathematicsinstruction is more commonly the target of negative attitkideethan other academic subjects such as reading and writing (2:558)These hegdtive attitudes\cap influence. the students expec-tation of future performance and affect his actual performance.Negative attitudes usually generate anxiety and hostilitytoward the subject. Attitudes toward mathematics and towardthe self are strongly related to ,performance in,mathematicsclasses (3:23).

A study of the, available research literature suggeststhat the contradictory nature of the few research studies ofremedial programs in the community college may be resolved if,we analyze_the\programs iri terms of'the extent to which theywere Personalized, that is'the,exterit. to which,they were .

designed to accommodate to 'individual differences. It is areasonable conjectitre that the unsuOcesSful.remedial programsmay have _failed bocapsa of their lack of emphasis on.factorswhich tend to.persvhb.lizelthe prpgram. It is these factOrsthat may prpduce 'long -term self-image ehhancehlent. On6 of themost effective personaliziig elements used in remedial programsis tutoring.

.

Existing studies of educationally, disadantaged studentsin the community, ,college and the remedial programs_ designedfor them lead to :Lie following assuMptions:

1. 'Self-concept is significantly related to long-termachievement and per'sistence- for disadvaptaged.studentst

'1*2. Self-concept enhandement tends to occur in person-

alit 'remedial programs,

3. :..Tutoring is%an effective means of personalizingTremedial programq, .

i

.

.1 /

1...

4. Remedial programs in which tutoring plays an integralpart should e effective iri producing long-termimprovement in academic.actiievement and persIstenceof educationally disadvantdged students.--

This assumption underlies the hypotheses being tested in this-study. '.

and therefore,'

. 9

/".

1

-Developmental MatheMatics inhthe COmMnity College'

1111

fThe phrase "developmental mathematice.is mos t often us ed

to indicati courses in arithmetic and elementary'algebra that',are usually encountered by most students-in the elementary andsecondary schools. In a 1966 survey of 73 California community

Kipps (40) found that 59 colleges offered thesedevelopmental courses. In a laterisurvey'Beal (6) examineddevelopmental mathematics courses in 185 community collegesand found that only 42 respondents spgcifically indicated thatthey offered no such programs and s4 questioned the approppi-bateness of such mathematics courses. 'In 51 of the collegesoffering, developmental mathematics, courses at least twentypercent of, all mathematics students and more than one-half 0of. the faculty were involved in these prog amgo .Fifty percentof the colleges surveye'd,required.students to enroll in adevelopmental course if their sCores on st ndardizedstests,previous grades in mathematibs, or counsel r recommendationsindicated that they wou.l.d profit 'from it. The most frequentlyindicated reasons for- the 'existence Of developmental mathematicscourses were to enable Students to- continue in regular mathe7'aatics courses (85%).or to satisfy prerequisites fdr othernon-mathematics courses.

In summary, tutorial prograMsfor low-ability students .are increasing in number and importance in the coin tunity

Colleges. Existing published studies of the effectiveness of-tutorial programs are restricted to examination of theirshort '-term results. Evidence. of the .effectiveness of theseprograms in producing significant positive changes in achieve_ -

1. went, grades, or subsequent performance in ' college,

coursesourses is Contradictory. MosttesearZh studies agree.,:thatttutorial programs are effective in reducing: attrition ofstudents enrolled, in the program. The contradictdry nature ofthe available research may be.resolved by analyZing theprograms in.terms of the extent to which they emphasize:.factors that tend to personalize instruction.

.

The basic intent'oi the pr'eseht stUdY was to `examine theeffect of tutoring on lo14 ability...Communit college students.All subjects in this experiment_were students at'Santa BarbaraCity College'enrolledjn':the 1572 fail semester in a,reMediamathematics course in basic arithmetic skills. Students.entered this course'bY referral. of counselors or teacArs onthe basis of low scores on thp'ScholasN Aptitude Test'ofinadquate performance in mathematics in' reziou6 courses.All students' scoring below the30th percentile on: the ScholasticAptitude' Test were, required to enroll in this course. Manystudents enrolled in this courte.vere also enrolled inreme courses in English and"study skills, and in highsc ool equivalent courses in the spCial sciences... A majority

1

10;..'

4

- ,

t

. . .

/of the students attending e College under p ogramsfor theeducati

attendingdisadvantaged were among the nea ly 1500 students

.enrolled in this course.

-.

11.Sampling

. ,

The-targetipopulation,for the study-was restricted 'to.)

those enrolled in the course desCribed above. Sampling froma this population resulted in the fOrmation of-tlii,ee experi-

mental, grOUps. 7 . I I

V

Group 1. Students in this control group met for one hourper week in small subgroups to study programmed materials.The meetings were supervised by a trained student tutor.Students took the four course examinations when they wished'and were. alloqed to re.Xake the examinations, using alternateforms until they' redeived a score.of 84 percent or higher.No. tutoring took4piace in the'weekly meetings attended by.these students. Students who asked questions in the smallgroup meetings were referred to explanations available in'the programmed textbook. A total of 62 students:were ingroupll. 4

I*, Group 2. Students in this classroom tutoring group were

enrolled' in. weekly 'small .group sessions exactly the same asthose for group 1 except that class time Was devoted to acombination of work in the programMed textbook and tutoring. i-Trained, experienced tutors, in a ratio of approximately bane, .--

. ',. tutor for every fine students, worked faith .the students pro -viding additional explanation, ewouragement, and generalassistance. A total of 67 students were in.cgrd#2.r,

Group 3. Students in this total tutoring group wer, enrolled in weeklyosmall,gro4p sessions exactly th sathose for group 2 except that all students receive di-tional tutoring at times other than the required class meetings.Each student received, a 'minimum. of one hour per Ileek of ,

additional tutoring and'a possible -maximum" of five tiours ofadditional tutoring..-A total of 61 students were in group.8.

, The three experlmental groups-were forpedby randomlyAssigning.si mathematics laboratory sections to-each-exper-imental category. The resulting'three experimental groups -*each consisted of six mathematics laboratory sections contain-ing a total of approximately Wfull-time fir'st-year stuaents.No outward identification of zrO.ups was made and studentswere unawaie,:that ,they had been assigned to mathematics

.

laboratory sections In any,systematie way.0

' 'In order to assure.identical treatment of all mathematicslaboratory sections included in group l,'the non-tutoredcontrol group- all sections were under th-e direc ion of the

l'. 0

1

I. ,'I.': J.. ,. J

,

, ;" y , JJ 4 ,

\ , i Id}, d . , .

N ,same' tutor: This' tutor was awar Of the natul'e:ana design ofthe .gtudy and had been carefully stru,cteil as to the_proper' .

nelpmdittalyesponsds to be made in response to studentrequests for informatidn and assistance:. ..,

. :.s ..

,'Ethical"considerations,required that tutoring,assistance.

e-

could not be withheld from students requesting it. 'The 4, -

policy followed was that any student In. the control group whorequested tutoring would be allowed to transfer to anothergroup notin the study and would be dropped frbm Considerationin'othe study. No students in this, grouR"requested either

,

tutoring or,transfer from the group. '

4J

Several'Pteps were taken to assure that there was npsystematic difference between experimental groups at the

..

outset that might influence the 4.1nal result. First, random ,assignment of students to mathematics laboratory sections.ando) laboratory sections to, experimental grodps assured thate ery firpt-year full-tide student, had an essentially equal

...

chance of being assigned to each of the experimental treat-ments. Random assigntent to experimental groupP,allowed;,the ..'

use of simple statistical procedures for.determiping the'.

significance ;)sr anyobserved differences in acAdemic perform-anceance or subj0tive responbep. ,

, ., ,,

, ..,. .-

Second, -only ruilLtime, student's, were included in .the *study.,This restrietipn was dired.t' attempt t(o, equate ptudentS on ethe to donunitmeiU to their co0:/etge prograt and .the time )

Nta a.va- lable-16r 4.r.

cademid, wbrk. .

., . c, Q c,

Thiid, only first-year student, were incauded'in'thestudy in order to eliminate any possibility of past'eXposure'to. tutoring and to assure that members ,of the' control' group,were Unaware of the.tutoring taking place in other sectionsof the course. Restriction.of the experimental groups tofirst-year students also eliminated differences dUe to pkevibusexposure to college work:

Both, experience and a urvey of the research literature,indicated that differences n response to tutbring,might becorrelated with confOundin variables such as sex, age, ethnicbackgrounds scholastic4aptitude, socioeconomic background, ,

previous training in Mathematics, amodftt. of time spent onoUtPide employment, financial aid Teceived,iattitudes, andself- concept. It was assumed that randomization equalized'these factOrs and elimiriated any sIgnificnt contribution ?"

they dIghtmake to differences in performance' between groups.,T4is assumption as tested for the variables of age, sex,

(/ scholastic aptitude, ethnic bac 'd, hours of outsideemployment, And amount of finan ial aid, in ans.ttempt todetermine if the experimental g odps differed ,significantlyan these iablet.

1

. i ..; 4. j. .--).

Differences in the.ability of tutors to _effect significant'hanges.in student penfOrmanC0 was controlled by trainingtutors prior to the study .and, Vmeeting regularly withtutors in an attempt to'maintaiii-constant standards of tutor.performance. All tutoring Wg8.performed Within, the framework -

.

.set by the progra ed instruction materials and the organi-zation. of the. Ca rse. This tended to insure( that all studentsreceived the sa e,kind of tutoring assistance.

<t e i.

, The pripary experimental variables' examined in thislongitudinal stiftIllere grades'in the developMental mathematicscourse, grades iri_Subsequent mathematics courses, overall 4p vwgrade-point averages in subsequent semesters, Withdrawal from C.individualcoUrses,and from college, and re'enrollment in college.'stter withdra;WaTheie objectivq,.quantifiable'mariables

rkere recorded'd4ping earth semeSter'for the two years followingthe initial e*D0..1Ment. Students were followed thY.ough two

4, years of their college career by meang of records maintaned."'in ttie-Department:of Admissions and Records of the college,

. .Because looSitive attitude change was expected by manyinstructors to'be a major result of tutoring, an attempt was'made to measure these changes directly over.the course of the,

.i.

first semester 'At.the end of the 1972 fall semester /eachstudent was interViewedindividually using a structured inter-view.technique in which the questions asked were carefully.planned beforehand. The was used irfpreference'tb'a Written questionn re because many 11 the -,students enrolled in the developmental mathematics course

,

were.poor,readers. Although'interviewing was.more time- g

Consumin&, it assured a meaningful response from every.student._

The qUestions used in thelinterview were,developed withthe cooperation of staff members and tutors participating inthe study. All 190 students in the three experimental groupswere interviewed using.the revised'questioalisi. Studentswho completed the course were interviewed immediately aftercompleting the final required texgmination. Students who did',not complete the colirse.but who persisted throughout thesemester,were interviewed at the last mathematics labbratorysection meeting.` Students withdrawing frpm the course were.interviewed when they requested permission; to withdraw.. In

'a few cases students withdrew Without nbtifying the instructor,and these students were interviewed by telephone if they couldnotbereached for a personal interview.

To assure* that'equivalent instructionalsituationsexisted for each grOup,,each mathematics laboratory section

, was visited brtpfly by the experimenter each week 'during the_Semester. Tutors met with the, experimenter each'week in,agroup training sessip in whibh any prOblems.relating*totutoring in,the expe imental groups were discussed.. Extensive-efforts were made to assure that all,mathematics laboratory

13 1

.1

k

I

(

A

. .

sections in a 'given expgrimental group were treated exactlyalike! Tutored groups recpived nO .special tutoring beyondthat-mormall'given,in non-experimental groups except that ,'each subject kn Foup.3. received additional- tutoring time.'[/-

. Students we e assigned to the three experimental gro pson an essentially' random basis, and'it was assumed that,thgroups did not differ significantl'y:at the start of the inves-tigation.. Study of the research lieteature and experience' '

in teaching developmental arithmetic at the'community collegelevel suggested that the initial variables most likely toiridluence student performance were sex, age, ethnic and radialbackground, Scholastic Aptitude (Quantitative)

'

Test scores,of financial aid; and number of hours of outside

employment per week. It was assumed that the random assign-.- hent of subjects to experimental groups, would eliminate-any

relative differences in performance due to these confoundingVariables.,

Comparison of the experimental,groups.Indicated thatinitially they were not significantly different in distri7bution of subjects by sex, age, ethnic and racial background,Quantitative Scholastic Aptitude test score,, fraction receiv-ing finanoialtaid, or amount of, qutside employment. The purposeof the diffeAntial treatment applied to the three experimentalgroups was to deny tutoring in developmenta.aMathematic's tosubjects in the control group and.to assure significantlylarger amount of tutoring to the total tutor ng group th-tothe classrocm tutoring. group.. .

The amount of -tutoring received wavd4termined by twoindependent methods: estimates bk. the tutors involved andestimates by the,students being tUtored: Both groups indicatedthat students in experimentalgrouP 1,-the ,control group,'received no tutoring. Comparison of tutor estimates of the''amount of,t;utoring received by groups 2.and 3 is shown inFigUre i. Ther+e'.wa' m.unanimous agreement that students ingroup 2. received one hour per week of tutoring. ;The amountof tutoring received by students in.grOup 3,as estimated byt(he tutors, varied as shown in Figure 1' with a mean value of3.39 hourS per week.

/ On the personal interview, students who had been tutoredwere asked to estimate to the nearest hour the average amountof tutoring they received each week. The distribution pfrespones for groups 2 and 3 is shown in Figure 2. The meancalculated for grup 3 was 3.89 hours per week. These esti-Aates are in close agreement with theestimates bytutors.

The slight variation in estiglates of tutoring time isA reflection of thg fact that man students regard at least

14

FIGURE 1

AMOUNT OF TUTORING

/ AS ESTIMATED BY TUTORS

Number ,of St-tidetits'

r

60-

40-

0-

1

4

Group

,Mean 10 hours/week

-4

Hours of Tutoring -per Week

Number bf'Students

20

10

2 3 4 5

. Hours, of Tutoring Per Week

Group 3

Mean 3.39 hours/week

ti

I

5

1

6

-

I

a't

F1GUR 2

AMOUNT 1DFAINGRING

AS ESTIMATEDBY STUDENTS

'N'umber of .Student's

6o/

.

/ 4o,

' . 20.

0-

41.

Numbei,of Studentp

74o-

3o,

20-

10

cr.=2 3 1 4 s5 6

Hours of Tutoring per Week

1

r

Group 2

Mean 0.99 hours /week.

Group 3

Mean 3.89 hours /week

r

==1.2 3 3/4 5 , 6.

of Tutoring per Week

.

iI

1

0 ..

a,portion of their personal contact with tutors as infqrmalconversation rather than tutoring. . Tutors regard the sane //:activity as an attempt to build the rapport needed to do an 4'effective job of tutoring.

. . ... ..

Shbrt.Tterm Performance of Experimental Groups'....

V

The two most 'imIDOrtant measures of short-term PeforM-ance for students in the experimental and control groups are-academic grades, both in the developmental mathematict 'course 4

and in othercouraes in, which they.were enrolled in the first :esemester, and persistence, in their courses and in the college.".

-,Academic Grades Figure 3 shows the distribution of gradeswithin the three experimental groups ,for those students'whocompleted the develo imentai-mathematics course-ddring,the

'first semester. Grade averages were calculated on the usual.'basis of assigning 4 points for an W grade, 3 points for a Bgrade, and 2 points for a C grade. As.dictated:by course /.

policy, only A, B, and C grades were awarded to studentscompleting the course successfully.` .11 grade of C representeda score of at least 80% on-each of the fouriCourse examinations.Students not completing all Tour examinations t the80%.level or iigher receiveeno letter grade.' Statistiai analysis.indicates that the grade distribUtions ar not significantlydifferent., ,

iep.

A second measure of short-term academic 41+formapc0,1sthe first semester grade-point avei-age Vat'courses AV-ler:thandevelopMental mathematics. ,,A sununary of these gradset appearsin Figure 4: Statistical analysis'indifates that the tpreegroups did not differltignificantlypWith respeciihe'gradesreceived in courses other ,than deveppmental.Mathtmatids.ti

. . ,

.. ...

Persistence he fact that grades in both the devbloPOipalmatheinatics course and in other c2y3ses taken during fhe firstsemester are not significantly different fdt the-ttetdred and.noon- tutored groups may seem surprising and may.appearto be adenial of the efficacy oil tutoring as an instructionalprocess.However, closes exam'ination of Figuree'3 and '4 reveals -thatadditiOnal analysis' is necessary. ,

.,

,. . ,

4

Figure 5 is a summary of the deelopmental mathematicscourse results all students enrolled. 'The. number ofdevelopmental mathematics students withdrawing from the collegeor not completing the developmental mathematics course during .

this semester is less for the tutored grOups than in t;h nontutored control group. If the mean ,grade -point average iscalculated for all.groups by assigning a valUe Of zero for d,withdrawalograde, then the average grade for the tutored groupsis larger than the averge grade fOl''the controls.

L

:1

17

..

I

%

! I

FI URt 3

.

GRADES II: DEVELOPIIEI:TAL MATHEEATICS

Number of Students...

03

V.

\

a

B C

Mean Grade 3.57

N =' 23.

Group 1

No Tutoring

A B\ C A B = C 4

Mean grade 3.68. Mean Grade.3.66

N = 37 N =,41

Group 2 Group3

One hour per week 3.5 hours per weekof tutoring of. ",tutoring;

1.

I4 ' 18

,

. 4

N,

Z

1:

4

FIGURE 4

i

GRADE POINT AVERAGES

, FOR DEVELOPMENTAL MATHEMATICS STUDENTS

vMean Grade Point Average '

for Students .Who Completed. the Developmental 'MathematicsCourse

A ,,..*...,...,-,..*......*

. Grbup 1

Non- tutored - Classroom Total

Control Group Tutoring Tutoring/

(N = 46) (N = 64) (N = 59),

- . ,..4

2.16 ,

r 4

2.36/

l

'',...,

Group 2 . Group 3

.

4 ;\

4

19

,

1 4

I

-8-

6 .

I a

1.8

,.)

40

,

a

FIGURE 5

Graaes for Studentsiltni'0110.1;

Developmental Mathematics

in the FirtSt Semes

=0.11

A

Numberinitially .

enrolled

'Nod-tt4Ored /

Control'Group-

%.

Numberreceiving apaising gradein jhe firstSemester

,23

Number 'withdrawingfrom c011ege ' 16

in. the, firstsemester

.Number nocompleting,the douree inthe first .

semester

Average grade-:_,received ,3'.57.

(A=4, B=3,0=2)

ModifiedGrade Average 1.32(A=4, B-3,C=2, W=0)

23

-G0up

ClaserOm

Tutoring,

67

3

7

3.68

2.03

)

-ZO

:

4

6

Total

Tutoring

61

)41

f

2

tEr

, 3.66

2,46

0

ST_ UDEMiS WI DRAWING FROM

DEVELOPME ETA L MATHEMATICS

FIGUR 6

.Percent.WIthdrawing frbmDevelopmental Mathematicsin the First Semester

60

40-

20-

0-

62.97.

Group

NonAtutored

Control Group

(N = 62)

/ a

44.8'/.

Group 2

eassroom

Tutoring

(N - 6,7),

C

32.8%

O

Group 3.

Total

Ilutoring

(N -'61),

21.

I a

$

yn

.._-

The resultt,Of a_n analysis of variance' performed. on . '

these data indit tes that when the irades of students vitharawingfrom the develo mental methematios course are included in the .

.analysis, the t tored.groups'earned'iignificantly highpC4grades,than dicl the,non-tntored controls.

.-- 0

From this analysis-ft is clear that the developmentalmathematics course policy of allowing all students to withdrawfrpm the course if they do not complete the required examinations,and of assigning only A, B,.and C grades, modifies the gradedistribution.pbtained. More students withdrew from the control,group than from the tutored groups, resulting in a spuriouslymean; grade-point average for the control group rIelatIve to thettitored groups. According to Figure 5 the relative persistency

,t of tutored versusnon-tutored students is an important variable,

.'figure 6s a comparison. of the relative numbers of studentswithdrawing frtb each of the three experlimental groups. SubStan-tiallfewer students Withdrew from groups 2 and 3, the tutoredgroups,-than.from group 1, the non-tutordd control group. The62.9% withdraWal rate from the developmental mathematics, coursefor group 1 is consistent with information on the withdrawalrate of 50% to 70% in, this course before tutors were employed.The 44.8% withdrawal rate for group 2 is-consistent with awithdrawal rate of 40-to 45%..in previous semesters when tutorsdere employed in a classroom tutoring situation, according toinformal records maiintained by the college.

tatistital analysis of. this informa4ion.indicates thatthese 'differences in the experimental. groups are-significantat beyonde.001 level. There is a significant difference inthe withdrawal rates for the three groups, with more students inthe non-tutored,controlgroup withdrawing than in either of thetutored groups.

This pattern of withdrawal is even more pronounced if thefrequency of-withdrawal of remedial mathematics students fromthe college is examined. .Figure 7 shows the percent of studentsenrolled in remedial mathematics who withdrew from collegeentirely during the first semester. The percent of studentswithdrawing from the non-tutored group (.25.8 %) is more thansix times the pei'cent of students withdrawing from the combinedtutored grups. (3.9%).

Adddtdonal evidence supporting the effect of tutoring onstudent persistence outside ofsthe,specific clpes,being tutoredis found in Figure 8. Figure 8 indicates that the percentof developmental mathematics students withdrawing from anycourse in the first semester is higher for the.non-tutoredcontrol grouP.(75.8%) than for either theClassroom tutoringgroup (56.7%) or the total tutoring group (49.2%).

. These differences are statistically significant.

T10fleJings agree with the research reported 'by Evans,

1. ,r ,

g,"4:2

.

lk

,

r

,,(26), Lucas, Faither, and Montgome

(59), and Riessman (68) indicatingeffect on course' glades.

(48), !Macmillan (51) Mohanat tutoring has, little direct'

I ,

These findings also support the obserVatiorts of Adams (1),Borfr`a d Hebert (10), Etters (25), Sykes (81), Taylor (83),Tille Porter, and Joiner (86 , and Yuthas (92) that tutoring

ry effective' in reducing trition. A major ihort-terM.lel Act of tutoring appears tobe the development of psychological

at

changes in students thatjqfluerice their behavior outside the ,

specific area of tutoring./1hpar neular, for this study, althoughthe tutored and non- tutored groups did not differ significantly.inthe grades earned in either the tutored course or in other courses,they did differ significantly in the rate* which they withdrewfrom the ,tutored course; from other courses, and from the college-

..during the first: lealeater.

.4. .

(

.4

2:3

7.22-

0

1

a

(t

e of

4

, .

FIGURE 7.

STUDENTWITITDRAWINd FROM

COLLEGE 'NITRE FIRST SEMESTER;

Percent. of DevelopmentalMathematiCsStudents Withdrawing from Collegein the First Semester

t

25-

20-

15-

10-

5-

0,

3

25.8Z

Ito 5 703.3Z

GrOup 1 Group 2 Group 3

Non-tutored Classroom Total'

Control Group Tutoring Tutoring

(N = 62) (N = 67) (N = 61)

-23.;

t

I

it

FIGURE 8

SillENTS WITHDRAWING FROM

COURSES IN THE FIRST SEMESTER.

.Pelicent of Developmental 4thematicsStudents Withdrailing from.AnyCourse in the ii,st Semester

75-

50-

0-

75.8%

Group 1.

Non-tutored

Control Group

(N = 62)

56.77.

Group 2

Classroom

Tutoring

(N = 67)

4

49. 2°

GPoup 3

Total

Tutoring

:, (N = 61)

AZ3

s

a

-24-,

1

o

LonA-term Performance of Experimental Groups

If the short-term psychological changes apparently *

produced by tutoring.persW'beyond,the,semester in whiih the. tutoring takesplace, and.if these changes are sufficiently.-general as is 'affect student performance in areas other than ,the tutored area, then their differential effects can btexpected to appear in the long-term academic performance andpersistence'of the experimental groups, , '

4?

Academic Grades The primary,reason for using tutoring inthe developmental methematias course was to increase the%probability that:a student would complete the course success-fully., Figure,9is a summary of the performance of allstudents enrolled in develQpmntal mathematics course duringpthe first semester of the study.

. -

. . Fon group f. the'39.studentS who,-did npt.aomplete the.course in t e rst 6ellester,f9 withdrew' from the college anddid not return. Of the remaining, 30,22 completed the- courseduring-the four semesters of the:study. A total of 45 or.72.6% of the non-tutored control group (group 1) completedthe .course successfully. -

°For,group 2, ofthe 30 students-who did not complete the'course in the first semester, two :withdrew from the college anddid not return. Of the remaining-2B,20 completed the courseduring the four semesters of the tudy. A ,total of 57 or85.1%of the classrooffi tutored group (group'2) aoMPleted the course.successfully..

For group 3, of the 20 students who, did not complete..thecourse in the First semester,, one withdrew from thecollege anddid not return. Of the remaining 19, 13 completed -the courseduring the four lemesters of the study'. A .total of 54 or 88.5%o'f the total tutoring group (group 3) completed the course:'successfully. Figure 10 is a graphic presentation of thisinformation.

'

.

Stitistdcal analysis'-indicates that the different :'in therelative numbers of students. completing the developmentalmathematics course for the three experimental groups issignificant at the:0.05 level, Significantly more of thestudents'in the tutored groups complted the developmental.mathematics course-in,the four semesters of the- study than.dia st'udehtkin the nop-tutored control.group.

One function of the developmental mathematics course isto prepare students to dontinue Pleir studie's in'a subsequent .

algebra-course. Of tne37 stud' in grotip 1 who successfully,,completed.the developmental mathematics course ..n then firstsemester and who reenroLled in- the- college in the second

0

0

te

r

FIGURE. 9

/'LONG-TERM PERFORMANCE OF DEVELOPMENTAL MATHEMATICS STUDENTS

1

Group 1 Group 2 --Group 3

y

..,i) N' Percent N Percent N ( percent.. .

Initially enrolled 62 .100 67 100 61 100.

.

Completed first semester 23 37.1 '37 55.2 41. 67.2

. Withdrew from college 16 25.8 3 4.5 . 2 .3.3. r

Never returned 9 14.5 '2 3.0 1 1.6.

Returned to course

,

in four semesters '4 6.5 1 1.5 0 0

S.

Completed, coursein foul' semesters 2 3.2 0 - 0 0 0

Returned to course,did not complete it 2- 3.2 0 1 1.6

. ,

Returned to college,did not reenroll Inthe course 3 4.8 0 0 0 0

. .

Withdrew from the course Ao 37.1 27 40.3 29.5

#Completed course *

in second semester 5 8.1 11 16.4' 8r13.1

Completed course ,

' in third semester 9 14.5 , 5 7.5 5 8.2

Completed course-in fourth semester 6 '9.7 3 4.5 '0 0

.

.

..

Did not. .complete_

, -

the course 3 4.8 8 11.9 5 '8.2

Completed-DevelopmentalMathematics .course.

,

within four emesters 45. 72.6 57 85.1 54 88.5

.

O.

. Z7

FIGURE 10

STUDENTS COMPLETING DEVELOPMENTAL MATHEMATICS

Percent Completing Developmental MathematicsWithin Four,Semesters

80-

70-

60-

50-

30-

20-

10-1

0

C

8

Group 1 / . Group 2

Non - tutoredControl Group(N = 62)

,

'A Percent completing the course in the first semester

ClassroomTutoring(N = 67) ,

0

Group 3

TotalTutoring(N = 61)

-27-

. v

Withdrew from the college and returned tocompletethe course within the four semesters of the study-

C, Continued in college and completed the coursewithin the Cow, semesters OT thestudy .

4

t

1.-

semester, only.10, or 27.0%, over enrolled in a suhgequent,mathematics course. Of the 53 students in group 2 whosuccessfully completed the developmental mathematies coursein the first semester and who reenrolled in the college in ,

the second semester, only 14, or 26.4%, ever enrolled in asubsequent mathematics course. Of the 56 students in group3 who successfully completed the developmental mathematicscourse in the first semester and who reenrolled in the college.in the'second semester, only 11, or 19.6%, lever enrolled ina subsequent mathematics course. Statistically,'there,is nosignificant relation b4ween the number of students enrollingin a subsequent mathematics course and the amount of tutoringreceived by them in developmental mathematics.

Average' grades foi; all students enrolling in algebraaffer.Successful completion of developmental mathematics areshown in Figure These grades are not significantlydifferent.

When grade averages for etudents completing the secondand third semesters are calculated by including withdrawalgrades, there is a significant increase in grade average for

.th4 tutored groups with respect to the non-tutored controlgroup. There is no significant difference in grade averagesbetween the classroom tutoring and total tutoring groups. .

I

Persistence If tutoring tends to promote significantpsychblogical changes in the student leading to decreasedatrition during.the first semester, then it is reasonableto expect, that these chang may lead to significantlyaltered patterns.of attrit on in subsequent semesters.

iree basic patterns or persistence are-found in the.experime tal-groups:

1. Students'may withdraw from he developmentalmathematiCs course and r -au..._IT a subsequentsemester. --......._

. 2. Students may withdraw froM the college and reenr611-in a subsequent semester.

3. Students may complete the elevelopmentalsmathematicscours in the first ,semester and reenroll in a.'subse ent semester.

With respect to, the, first of these patterns., It was found thatthere is no significant difference among experimental groupsin the number of students who reenrollafter withdraiging fromthe developmental mathematics course during the first.semester. This finding is consistent with the observationsof tutors and instructors associated with the developmental

01

. r;./

FIGURE 11.

'Average Grades fQr Students4-

Enrolling in an Algebra course

After Successful Completion of

Developmental Mathematics in the

First Semester

. Group 1 Group 2 Group 3

Non-tutmed Classroom TOtal

Control G up Tutoring Tutoring

N 10 114 11

Grade1

1.40 1.7 , .1 73-Average

1

1Thd grade av4rage was calculated on the be.sis of,g = 14,

C = 2, D = 0.

-29-

1

<4,

0

-4

j.

\

p.

-30-r;

,mathematics course that 70% to.80% of all students Who withdraw', from developmental mathematics will reenroll. Most of these

students-have completed some portion of the course and havea strong incentive to reenroll in order to earn a passing \

' grade.

With respect to the second pattern, there is no signifi-cant difference among experimental groups in the number ofstudents' who reenroll after withdraWing 'from the collegeduring the first semester. Although the relative frequency ofwithdrawal of stufints from college is sd2gnificantly greaterfor the non-tutored students,in the conftol group, the fractionof students reenrolling is independent'of their previousexperience with tutoring.

Witt respect to the third.

phttern, there-is no significantdifference amcIng experiment4 groups in the number of students ..

who reenroll in a subsequent semester' after successful completionof the developmental mathematics course. This finding supportsobservations by tutors and instructors in the developmentalNmathematic 'course with respect to the very positive effe.ct on. .

student mo a e of successfully completing the developmentalcoUrse . . .

. ..

There is a significant relation between the number ofstudents enrolled in college in.the fourth semester and the e

amount.of tutoring received by them. While 38'.7% of the.students in the non-tutored control group' (group 1) areenrolled in the college in the fourt4Lsemester, 86.7% of

-1 the tutored students (groups 2 and 37-are enrolled in thet college in the fourt4 semeste.. This' difference is significant

at well beyond the 0.01 level:* The number of tutored studentienrolled in the college in the fourth,semester is significantlygreater than the number of non-tutored students enrolled inthe fourth semester. This finding is consistent with theIncrease in short-term persistence. Tutored students aresignificantly more likely to persist in college than are their'non-tutored'couhterparts. ,

If tutoring is associated with increased persistende incollege,,it might be expedted that tutored students who enrollin a, subsequent algebra qourse would be less'likely to withdraw'from that course than nort-tutbred students. However, thefraction of students withdraying from algebra is essentiallythe samefor all three experimental groups. Taken togetherthe findings support the conclusion that the' amount of tutoringreceived by students in developmental mathematics is not a-significant factor in determi ng their subsequent enrollment,

''w.ades, or persistence in an gebra course. This result wasnot unexpec ed. T e developm ntal mathematics course coneeptsare qualitat velz( different f om those encountered in algebra.The developmental.mathematic course emphasizes arithmetic

e

at

computation and. practical problems rather then theoreticalconcepts and abstract problems. The skills required.for ,

success in.the algebra, course are not the same as thosein the developtental-course; they are necessary for

success in algebra but not sufficient.

If tutoring, produces psychological changes associatedwith increased long -term persistence in college courses, it :

might be expected that students in the tutored groups wouldwithdraw from subsequent courses less frequently than studentsin the non-tutored control group. In Figure 12 the percent of

1' students withdrawing from one or more courses in the secondsemester is given' for each of the three experimental groups:.More than touble the number4of.students from the non-tutore:eYcontrol group withdraw from one or more bourses in the tecondsemester than from either of the tutored groups. Again,tutoring is associated with long-term persistence in collegecourses. '

Summary and Inter ation of Interview Results

On the basis of the personal interviews held with eachstudent it was possible to assess their expressed attitudestoward developmental mathematics, toward other courses, andtoward the college.

Figure 13 is a summary of thelattitude toward mathematics k

ratings arranged abbording to experimental groups. As, expected,the expressed .attitudes of students in the tutored grkao weremore .positive than the expressedlcattit4ies of ,students In thenon-tutored control group. The Mean rating for tutored studentsfell in the neutral eategail. For most students enrolledin a developmental mathematics course the normal response toany question about, attitudes toward mathematics or mathematicscourses is strongly negative. Modifying that response fromnegative to neutral, as has apparently accomplished here,is a substantial achievement.

One question on the interview involved the change' inthe student's attitude toward other college course's over thesemester. The responses are summarized in Figure 14. 'Studentsin the tutored groups indicated a significantly more positivechange in attitude toward-all courses than did sl.Idents inthe non-tutored control group. This finding is consistent'with the significant decrease in withdrawal from 'courses inthe first semester and the more positive attitude toilard,developmental mathematics found for students in the tutoredgroup as compared to students in the non-tutored controlgroup.

Responses to the interview .questions by tutored studentsrevealed that many of them were receiving assistance from the

FIGURE 12

STUDENTS WITHDRAWING IN THE SECOND SEMESTER

N

Percent of studentswithdrawing from oneor more courses 1nthe second semester

75 73.o%

50-

25

0

a

Group l'

Noh-tUbored

Control Group

'(N = 37)

34.0%

Group 2

Classroom

,Tutoring.k,

(N 53):

a.

25.07.

Grqup3

Total

Tutoring./,

(N'= 56)

5.

FIGURE 13

Composite Attitude Toward Mathematics

Rating Based on Responses to Interview

Questions .

N Mean Rating Range .

62 8.08 5 to 12

67 12.61 8 to 17

613 12.66 9 to 17

0

A rating'of 10 or less, represents a negative attitude toward

mathematics.

kratingof.11 to 13 represents a fteu.tral attitude toward

mathematics.

Arating Of 14 or more represents a positive .attitude toward

mathematics.,

V

4t

V

1 C44

O

-s3-

41.

!a

.

4: 1

FIGURE 14

..' 7t4

Responbez-Ato Question %. in the Interview,

Question 9: Has your attitude toward other college courses

.

.

changed from what itmas when you started the

semester? If so,how?

4.

Group 1 Group 2. Group 3 .

Noi- tutored Classroom Total

Control Group Tutoring Tutoring

(N = 62). (N = 67) (N = 61)

RestrOnse(a) I have a much`:

. more negativeattitude

(b) I have a morenegativeattitude

(c) My attitudehas notchanged

(d) My attitudeis morepositive

(e) My attitudeis much morepositive

Mean Response.using(a)

(c)(e)

=

==

1,

5.

(b)(4)

= 2,= 4,

3 04

21 0

38 15

.

0

0

,

37 .33

0 15' 28.

2.56 4.00 4.46

135

. ...tutors with courses, other than developmental mathematics. Manyindicated that their attitude, towardother courses was morepositive because they were receiving thisvpersonal help andbecause they believed theycould succeed in the other courseswith this help. Responses to the interview questions bytutored'students reflected open appreciation for the tutorial. .

help they were receiving and a feeling that they could succeedin their college courses because of the tutorial help theywere receiving. For many, the tutorial assistance was viewedas a direct effort'lay the college to help them succeed.

SUMMARY

The following...are the findings related to the performapceof the experimental groups during (the first.semester:

o

1. .The.tutored and non-tutored groups did not differsignificantly with respect to the grades received in

udevelopmentallftathematics during the first semester.Grade-point averages did not differ signifitantlyamong the three groups,.

If the grades ofstudents withdrawing from thedevelopmental mathematics course are included in the,grade average, students in.the tutored groups earned:significantly higher grades thandid students in thenon-tutored control group. If grade-point averages

_are calculated by assigning a value of zero to awithdrawal gra e, students the classroom tutoringgPoup, who rec ived one hour of tutoring per week,had an average f.2.03. Students in the total ,

tutoring group, who receivedapproximately.3.5 hours:4of tutoring per eek, hack an average grade of 2.46.',Students in the non-tutored control gro4 had anaverage grade of'1.32.

2. Significantly fewer students withdrew from, the tutoredgroups, than from the non-tutored control group. Ofthe non- tutored students 62:9 percent withdrew fromdevelopmental mathematiossin the first semester while39.1 percent of the students in the tutored groups

\,

withdrew. (Th. / .

3. On'the basis of the structured-interviews, students .

in the lored'groups were rated as significantly morepo6itive n attitude toward developmental mathematicsthan were \students in the non-tutored control:group.The average rating of attitude toward mathematicsfor students ih the tutored &cups fell in theneutral category yhile, the averaMe eating fbr studentsin the non - tutored group; .fell in the negative category.

-36-

4. The tutored and nol-tutored groups did not differ *4'significantly with respect to the average gradesreceived by students .in the first semester in courses.other than developmental mathematics:

Significantly more -of the students in thelnon-latoredcontrol group withdrew from, at leastone Classthan in either of the tutored groups. Of\the tutoredstudents 53.1 percent withdrew from at lest onecourse during the first semester, while 75.8percent of the studentt in the non-tutored controlgroup withdrew from at least one course.

6. On the basis,of the structured interview re pohses,.students in the tutored groups indicated a ignificantlymore positive change in attitude toward all othercourses than did students in the,non-tutored control-group.

.The following are the findings related to the perfprmanceof the experimental groups during the three semesters s bsequentto the initial enrollment-in the developmental mathemat cs course.

For those students who withdrew from the'devel pmehtalmathematics course in the Tirst semester there wasno significant difference betweeh the' tutored a anon-tutored groups.with respect to the number ostudents who reenrolled in a subsequent semeste,.More than 70 percent of all students who withdrewfrom developmental mathematics reenrolled in S \

subsequent semester.

8. FOr those students who withdrew from the college inthe first semester there was no significant differencebetween the tutored and non-tutored.groups withrespect to the number who re:enroll in ,a subsequentsemester. More than 40 percent of all students whowithdrew from the college ".eenrolled in.a subsequentsemester. .

9. Foli thoseistudents who completed'the developmental. mathematics cours., there was no significantdifference between the tutored and'non-tutored groups

-.!in the number of students whow,;reenrolled in a-subsequent semester, Approximately 97 percent ofthe students who colilpleted the developmental mathe-matics course with1a passing grade reenrolled in somesubsequent, se star'.

The.nuMber, Cored students enrolled in the collegein the fourth s mester of the study is,-significantly

greater than the number,of non-tutored studentsso enrolled. While 38.7 percent of the studentsin the nom-tutored group were enrolled in thecollege in the fourth semester; 86:7 percent of thetutored students were enrolled.

37

*11. For those students who completed the developmentalmathematics .course in the first semester, therewas no significant difference in'the numbers fromthe tutored and non-tutored groups *ho'enrolled in

if a subsequent mathematics. course.. Only 21.7 percentof those students who successfully completed develop-mental mathematics in the first semester. later'enrolled in another mathematics course.

12. For those. students who completed devellpmentalmathematics and enrolled in a subsequent mathematicscourse there was no significant difference betweenthe tutored aria non-tutored groups in the gradesreceived in the, later mathematics' courses

4

13.' For those students who completed developmental mathe=matics and enrolled in a subsequent mathematicscourse there was no significant difference betweenthe tutored and non-tutored groups in the number whowithdrew. from the later mathematics course.

14. Thee was no significant difference in secondsemester trade-point averages earned.by students fromthe tutored and non-tutored%groups when'grade-,pointaverages are calculated using A = 4; B = 3, C = 2,D = 1, and F If withdrawal grades-are included-in the grade average by assigning them a value ofzero, students from the tutored groups earnedsignificantly higher-grades in the second semesterthan did students from theugh-tutored group.

15. If withdrawal grades 'are included in the grade 0average by assigning them a valueof zero, studentsfrom the tutored groups earned significantly higher'grades in the third semester than did students fromthe non-tutored group. A A.

16.. The number of non-tutored students withdrawingfrom's:m:1e or mere courses in the second semesterwas signifiCantly greater than, the number ofstudents from the tutored groups similarly withdrawing.-Of the. non-tutored students 73.0 percent withdrew.from at,least'one course in the third semester, whileonly.31.6 percent of the students from the tutored,'group similarly'iwithdrew.

.2.%r"63

-r-1

17. The number of non-tutored stud nts withdrawingfrom ne or More courses in th third semester wassignif cantly greater than the number of studentsfrom t e tutored group similar w hdrawing. Of-the n-tutored students 70.0 percent withdrew fromat east one course in the third semester, whileonly 31.6 percent of the stdents from the tutoredgroup similarly' withdrew.

In general, students tutored in a developmental mathematicscourse are more likely ,than their nob- tutored peers to remainenrolled in the developmental mathematics course, to remainenrolled in college, to withdraw from fewer courses, and expressmore positive attitudes toward their courses during the semesterin which tutoring takes place. Thit pattern of increased persis-Ohce continues during the two years following tutoring.,

Recommendationt

On the basis of the findings of this study the followingrecommendations seem reasdnable:

1. Decisions concerning the USO of tutors in developmentalprograms in community colleges should be based on theprobable influenee of tutoring on the p'ersistence ofstudents in college and on the, development of improvedattitudes-toward coures rather than on the expectationof higher grade-point averages. The considerableexpenses and time expended 9,p the developmentand main-tenance of tutorial programs in the'community collegecan best be justified on the basis of the effectiveness'of such programs in, uilding PositiVe attitudes

_

and enhancing self-concepts as expressed in improvedhort-term and long -term patterns of persi ence.

,

2. n programs or courses in which tut s are employed;effectiveness shogld be measured by the combinationof academic grades, persistence,.and change inattitudes. toward the program rather than by grade-pointaverage alone. This is especially true for programsdesigned for academically disadvantaged studZnts.

In basic skills or develOpmentak.programs in thecommunity college where persistence in-college orin the program is a major determinant of theprobability that a student will'succeed,-tutors andstructured, tutoring should play a major:role. Inprograms in developmental mathematics, reading skills,basic writing skills, English as a second language,or other basic academic skills, long-term persistenceis usually necessary for success. 'It'is'suggestedi, .

therefore, that tutoring should be a central componentof such prbgrams.

A

4. Tutoring should not be seen as a strategy forenhancing the4ffectiveneSs of \a, i

arithmetic course as a mean6 of preparing low--ability students for success in a subsequent. ,

algebra course. This study supplies no assurance'that tutoring per se is 'significantly 'related toincreased probability ofenrollment in a subsequentalgebra course, or to increased retention or

./ . improved grades for students who do enroll in aN

.

subsequent algebra course.. ' ep.

Tutoring \should be considered as a potentially. .

valuable/element in the design of instruction in all1 academic and occupatipnal are a.s of the comkunity

college. This is parb.i.adlarly'true in individualized,instruction prog -rams. and in progr4ms-designgd for

r

the non-traditional learner who may -be More likelythan traditional students to withdraw from collegeor who may have serious unresolved problems relatingto negative attitudes, poor self-image, anddiminished self - confidence,

6. Further research should be conducted on th relativeeffectiveness of structured tutoring in is

skills areas other mathematics: read ng, writing,language skills, or geb4ral study skills. It is.not,unreasonable to expect that studies parallel to thepresent pne will yield simU.ar results and extedapresent Understanding of the differential effebts oftutoring. Past studies in these areas have sufferedfrom the lack of structure supplied for tutors, andtherefore consistent TesUltshave not been obtaine,d.

7. Further research should be conducted on the aspects of.the tutoring process and the characteristics of tutorsthat are related to more effective tutoring. In ,

particular, the following factors should. be examinedexperimehtally:

a. the actions or strategies employed.by tutors inthe tutoring process;

b. the effects of tutor training;c. the effect of matching.tutor and student on the

basis of cognitive styles personality variables,. ethnic, racial, or socio-economic babkground,

or other characteristics;d.: the role Of processes and devices other than

programmed-instruCtion,as a structure or frame-bl./ork for the tutoringzactivities.

Further research should be performed to determine theextent to which the positive effects of the tutor°-student interaction are due to subject-matter oriented

4.0

( or group of persons.. ,

4 39, This research should be extended to student .

populations other than the educationally disadvantaged.'Tutoring is being used increasingly' in all areasof community college instruction, includingacademically talented students and'adult andcontinuing education ptograms. Most of,theseapplications of tutoring have not been the subject- .

of systematic, controlled investigation; even thoughthey represent important areas of co munity collegeinstructional activity.

activities or to the. establishment of interpersonalrelationships of a more general nature. It wouldhave been enlightening to have followed the piwressof a fourth experimental'group who were not tutoredin mathematics but were encouraged to form. an rintense persona4 relationshtp,with another perspn

10. Furtrier research should be conducted on e natureof the attitude changes that are'apparently

. ,

produced.by the tutoring process. Standardizedinstruments such as personality-inveritdries or .

other psychiatric probes should be used to describeand delimit more carefully the attitude and self-image changes otcuring during tutoring.

11. Further research should be-conducted on the differentialeffects of tutoring as related to the sex', age)4ethnic or racial background, socio-economic background,personality variables, and initial attitudes ortutored studepts. These factors haVe not. been thesubject of study in-this investigation.- -Reseal'on these raotors provide valuable inforthat n fortutors and for e'aucational planners involved indeveloping and managing tutoring prograth6, trainingtutors, or developing instructional materials to beused in the tutoring process.

*

BIBLIOgRAPHY

Adams, W. Royce. "Tutors and the Reading Lab,"Western College Reading Association Newsletter,Spring, 1971, 3

2. Aiken, LewisR., Jr. "Attitudes ToWard Mathematics,"Review of Educational Research, 40, X4),October, 1970, 551-596.

3. Aiken, L.. R. and R. M. Dreger. "The Effect of Attitudes. on Performance in Mathematics," Journal of

Educational Psychology, 52) 1961, 19-24.

4. Baehr, R. F. Project Success, Chicago City College andKennedy-Kng Colltgei. Washington, D. C.: UnitedStates Office of Education, 1969.

.

5. Bausell, R. Barker; Moody, William B.; and F. Neil Watzl."A Factorial Study of Tutoring Versus ClassrdomInstruction," American Educational Research Journal,,

. 9, No. 4, Fall, 1972, 591-597t--

6. Beal, Jack. "An Analysis of Remedial Mathematics Programs.in Junior and Community Colleges," UnpublishedResearch Report, University of Nebraska, October, 1970(ERIC publication ED 043 335)

Bell, Stanley E.; Garlock, Norene L.; and Sam L. Colella."Students as Tutors," The Clearinghouse, 44,December, 1969; 244-

-

Benz, Donald A. Observations of Academic Performance byLow Achieving College Freshmen Following Instructionby Academically Successful Students Trained toTeach Reading and Study Skill Techniques, January, 1970,Stevens Po4t, Wisconsin: Wisconsin State University

7.,

IOW

8.

0

iM

9..

10.

11.

Consortium for Researchand Developmpnt. (ERICpublication ED 054 075)

Bernstein, A. L. "kotivations in Mathematics," SchoolScience and Mathematics, 1974,, 64)0749-754.

Born, David G., and emily W. Herbert. "Further Study ofPersonalized Instruction for Students in LargeUniversity Classes,", Journal of ExperimentalEducation, 401 Fall, 1971, 6-/1. .

Brim, Orville G., Jr. "College Grades and Self Estimate sof Intelligence," Journal of Educational Psychology,45 ;.1954, 477-484.

At 11

42

12. Carman, RAert A. A Systems Approach to the Developmentof a Junior College Course in Remedial Mathematics,'Novertber, 1970 (ERIC publication ED 034 536)

-,

13. Clarke, Johnnie R. and Rose Mary Ammons. "Identificationand DiagnosiS of Disadvantaged Students," Junior.College Journal, 41, February,'1970.

14. ClOward, Robert D. "Studies in Tutoring," Journal ofExperimental Education 36, No. 1, Fall, 1967 -, 14-.25.

15. Cross, K. Patricia. "Higher Education's Newest Student,".Junior College Journal, 39, (1), September, 1968.

- 16% Cross, K. Patricia. "The Junior College 'Student: AResearch Description," Center for Research andDevelopment in Higher Education, University ofCalifornia, Berkeley, California, 1968.

17. Crown, Sidney. "Tutoring. Tutors: An Experiment ifnPsychiatric Communication," University Quarterly,24, Autumn, 1970, 422-432.

18. Deterline, William A. Training and Management ofStudent Tutors, April, 1971 (ERIC publication ED 049 133)

19. Diggory, J. C. Self Etation: Concepts and Studies,New York: John Wi y and Sons, Inc., 1966.

20. Dinner, Martha. "Tutoring by Students: Who Benefits?"Research Bulletin of the Florida Educational Researchand Development Council; 7, No.'1, Spring, 1971.

21. Dreger, R. M., and L. R. Aiken. "The Identification ofNumber Anxiety in a College 'Population," Journal

Psychology,.1957, 48, 344-351.

22. Ellson, D. G.; Harris, Phillip; and Larry Parker. /"A

Field Test of Programmed and Directed Tutoring,"The Reading Research Quarterly, 3, Spring, 1968,341-367.

23. Ellson, D. G:; Larry Barber, T. L. Engle,' and LeonardKampwerth. !!Programmed- Tutoring: A Teaching Aid

.

f.,f1.

1 6zs

; 7'.

121

7

a:o

71," Readin Research Quarterly,

_1 . Ar

%..,

.

i. .

24''. Endler, Norman and Danny Steinberg.' ."Prediction of Academic.Achievement at ti University Level," Personnel andGuidance Journal, 41, (6); FebrUary,.1963, 694-699.

.

1

r.

C.

-.43-

25. Etters, E. Martin. "The artorial Assistance tn CollegeCore Courses," Journal of Educational Research,60, No. 9, May -June, 1967,',406-7.

26. Evans, Jerome. "Free Access'': What it Takes,"1. Change, 7)Marcli, 1974, 53-54.

27. Ewing, T. N. and W. M. Gilbert. "Controlled Study of theEffects of Counseling on Scholastic Achievement6' ofStudents of Superior Ability," Journal of CounselingPsychology, 14-,, 1967; 235-239.

28. Frager, Stanley and Carolyn Stern.' "Learni'ng by Tutoring,"The.Reading Teacher, 23; Febriaary, 1970, 403-405.

29. _Gaither, Loren. A 'Study of Remedial Students, March, 1968- (ERIC publication ED 025 253).

30. Hamburg, Morris. Statistical Analy s for Decision Making,New York: Harcourt, Brace and WoRld, Inc., 1970.

31.. Harris, T. L. "Summary of Investigations Relating td Reading,'Journal of Educational Research, 60, 1967, 237-245.

32. Harrison, Grant Von. "Training Students to Tutor,"Santa Monica, California, Systems DevelopmentCorporation, 1967, .

33. Harrison, Grant Von. "Structured T i g," May, 1971 /(ERIC' publication ED 053 080).

464. Harrison, Grant Von and.Arthu'rM. Cohen. "EmpiricalValidatioA of Tutor-Training Procedures," Paper.presented at the. California Education AssociationAnnual Meeting, San Diego, April 29-May 30,19,71.

35. Redl, John J. A Systematic Investigation of Three Facetsof Programmed Instruction: Tutorial Assistance. of tudy,Explanation of Incorrect Answers, and the Spacing ofH19-Difficulty, Frames. Technical ReportNo. 4)Florida State University, Tallahassee, Febrti.6.2.v, l970..'

36. Heinkel, Otto A. Evaluation of a General Studies Programfo the Potentially Low Academic Achiever in CaliforniaJunior Collevs; Final Report. San Diego, California:San Diego City College, 1970 (ERIC publication ED

-039 881)

37. Rernandet, Norma G. "Variables Affecting Achievement ofMiddle School Mexican-American Students," Reviewof Educational Research, 43 (1), Winter, 1973, 1-39.

4!

,op

.38. Isgor, Hazel-Ahn. Tutor Training, Washington, D. C.:"Tutorial Assistance Center, 1966.

9,.. Kaye, Mildred. College Discovery and DevelopmentTutorial Program, Research Report ORF-68-14,

e. October, 1968, City University of New York(ERIC publication ED 035 706).

40. Kipps, Carol..

Basic Arithmetic in California: Public.

Junior Colleges, Unpublished Doctoral Dissertation,University, of California, Los Angeles, Los Angeles,California, 1966.

a

41. Knoell,.Dorothy M. "Are Our Colleges Really Accessible tothe Poor?" Junior College ,journal 39, October, 1968,9-11..

42. Knoell, Dorothy.. "Outreach to the Disadvantaged" in .

Understanding Students and Counseling in the Two-Year Colleges, State UCiversity of New York at ,'

4. Albany, Albany-, New York, June, 1969.

43. Knbell, Dorothy M. People Who Need-College, Washington, D.0 .

American Association of Jun or Colleges, 1970.

44. Knoell, Dorothy. The "New ttude t" in the Commu. Paper presented at the annual meeting of t e American

Educational Research Assoc tion, Chicago, Illinois,April 6, 1972.

45. Liette, Eileen E. "Tutoring: its Effects On Reading.Achievement, Standard-Sett ng and Affect-MediatingSelf-EValuating for Black Male Under-Achievers in.Reading," Report BR-0-E4-0211.U. S. Office ofEducation, Bureau of Research, Washington, D.June, 1971.

^fT

14 6 . Lippitt, Ronald and Peggy LippJ.tt. "Cross-Age Helpers,"National Education Association 11ournal, 62;

. March, 1968,.24 -26. a0

47. Losak, John. 4Do Remedial Programs Really Work?"Paper presented at"the annual meeting of the American,Educational Research Association, New York, February,1971 (ERIO'public'atLon ED 046 975)

r%

48. Lucas,. John A.; Gaither, Gerald and James R. Montgomery-. "Evaluating a:Tutorial Program Containing VolunteerSubjects," The Journal of Experimental Education;36-(3), Spring, 1968, 78-81.

/

., .

'

. N.

49. McClellan, Billie Francis. "Student Invovementin theInstructional Process through Tutoring." .

. June, 1971 (ERIC publication ED,055 046).

McDavid, J. "Some RelationSi4s Between Social Reinforce-ment and Scholastic Achievement," Journal ofCounseling Psychology, 23; 1959,-151-154.

51. 'MacMillan, Thomas F. An Evaluation of ,ExtendedOpportunity Services at Santa Barbara City College,Fall 1970, Mimeographed Research Report No. 8-71,Santa Barbara, California, April, 1971.

52. MacMillan, Thomas F. The Identification of PotentialDropouts, Mimeographed'Research Report No. 6-72,Santa Barbara, Californa,March, 1972.

53. McNamara, Margaret,' "Group Dynamics in UniversityTutorials," University Quarterly,J6, Spring, 1972,231-253.

54'. McWhorter, Kathleen T. and Jean Levy. "The Influence of aTutorial Program Upon -Tutors," Journal of Reading,

(e' 14, January, 1971, 221-224.

55. Medin, Julie. The Teaching of Developmental Mathematicsin Community Colleges, May, 1972. (ERIC publicationED 075 197).

56. Medsker, Leland L. and' James W. Trent. Beyond High School',Berkeley: Center for Research and Development inAigher Educatiorl, 1964.

57. Melaragno, Ralph J. and Gerald Newmark. "A TutorialCommunity Works toward Specified Objectives In .anElementary School," 'Educational Horizons, 48,,Winter, 1969,.33 -37.

58. Menges, Robert J.; Marx, Robert; and P. W. Trumpeter."Effectiveness of Tutorial Assistance for High-Risk Students in Advanced College Coursesy.Journal of Counseling Psychology, 19 (3),'1972,229,-233.

59. Mohan,. Madan. "Peer Tutoring as a Teafinique for Teachingthe Unmotivated," Research Report of the TeacherEducation Research Center, State University ofNew York', Fredonia, ,New Xork, January, 1972.

60. Mbore, William, Jr. Against the Odds: TheHigh Rik,Student in the Community College, San Francisco:.Jossey-Bass, Inc., 1970.' ,

46

4

61. Morrisoh, Jane L. and Reynolds Ferrante,'-pompensatoryEducation in Two-Year Colleges, RepOrt 'No. 2Y,'Center for the Study of Higher Education, UniversityPark, Pa., Aprils 1973. .

a

4649.

62. Newman,A. "Tutoring the Culturally Different, ", Improving College and University Teaching, 19,

Autumn,' 1971, 299-300 f.

63. Niedermeyer, Fred C. "'Effects of Training on theIhstmectional.BehaVior:S of Student Tutorsfl!Jodrnal of Educational Research, 64,'NdveMber,1970, 12,2-1e7..

64. Noce, James S.- .Research and Evaluation in Tutorial

Programs, -Washington, D. C.: -TutbriaY.AssistanceCenter, 1967.;

Pellegrini, Robert J. and Robert ra°' Hicks.' "ProphecyEffects and TutorialInstrUctlon for the Disadvantaged.Child,", American Educational-Research Journal, 9(3), Summer, 1972, 413-419.

66 Phillips, George '0.. Study Habits and Attitudes ofDisadvantaged Students in a College Reading andStudy.Skills,Program, December, 1969 (ERIC publicationED 036 419).

67. Pierce, Richard A. and Gwendolyn Norrell. "White Tutors /4

''for Black Students," Journal of College StudentPersonnel, 11, April, 1971, 169-172.

68.1' Riessman,'G.. W. et al. Children Teach Children, New York:Harper and Row, 1972.

69. Robertson, Douglas J. "The Effects of an IntergradeTutbring Experience on Tutor Self-Concept," Paper

, ° presented at the49th Annual Conference of the,a California Educational Researth Association, San4 Diego,

70. Robeson, Douglas J. and Vicki F. Sharp. The Effect ofFifth Grade Student Tutors on the Sight WordVacabularyAttainment of First Graders, May, 1971(ERIG publication ED 055 735).

71. Rossi,, Timothy P. "HELP: Students Teach Students,"Reading Improvement, 6, Fall, 1969, 48-54.

72. Roueche, J. E. ."The Junior College Remedial Program,"Junior College Research Review, 2, 1967, 3-9. s.

':117o/

73. Roueche, J. E. and A. S. 'Hurltkurt. "The Op Door College:The Problems of the Low Achiever," Journal ofHigher Education, 34, (8), November, 1968, 453-4560 '

74. Roueche, .John E. "Accommodating Individual Differences,"Community College Review, July, 1973, 24-29.

,

.75. Schenz, Robert F. An Investigation- -of Junior CollegeCourses and Curricula for Students with Mw Ability,Unpublished Doctoral Qissertation, University ofCalifornia, Los Angeles, Los-Angeles, California, 1963

76. Schoeller, Arthur. W., and David A! Pearson, "BetterReading through Volunteer Reading Tutors," TheReading Teacher, 23, April; 1970, 625-630.

77. Sharon, Amiel T. Effectiveness Of Remediation in JuniorCollege, Research'Bulletin RB-70-50, Princeton,J. J.. Educational Testing Service, September, 1970(ERIC publication ED 051 795)

-78% Shaver, James P. "Tutorial Students Two Years.Later: AReport on the Logan-Cache Tutorial Center for° Under-achieving Readers and Writers," Logan, Utah: UtahState University, Logan College of Education, Report .

Nuillber DPSC-66-2152; OCtober, 1970. .

I.

79. Silberman, H.; Coulson, J.; Melaragno, R.; and Gary .

Newmark. Use of Exploratory Research and IndividualTutoring Techniques for the Development of ProgrammingMethods and Theory, Santa, Monica, California, SystemsDevelopment Corp., 1964,

4

80. Stein, Richard S. Some Concepts Held by Los Angeles CityCollege Entrants on Probation Because of Low SCATScores, Mimeographed Research Report) Los Angeles:Los Angeles City ColTege, 1966.

81. Sykes, Abel Baxton, Jr. "The Effect ofTutoring, ReadingInstruOtion, and Financial Stipends upOn StudentAchievement, Atti-ition2, and Attitudes at ComptonCollege," Unpublished Doctoral Disertation,':University of California, Los Angeles, California, 1971.

82. Taylor, Ronald G. "Tutorial Services and, Academic SUccess,"Journaiof Educatienal Research, 62, (5), January,19694 195-197

83. Taylor, Ronald G.; Cartwright, Paul;"and Gary R. Hanson.'.Tutorial Programs for'Freshman Engineering Students:Effect on Grades and ttrition," The Journal ofExperimental Educationo' 38, (3) Spring., 1970,87-92.

-48.:

84. Thelen, Herbert A. "Tutoring by Students," The SchoolReview, 77, September - December, 1969, 229-244.

-85. Tildon; Charles; Bard, Harry; and Robert Wilson; Jr."The-Elack Student: The Community College,"Junior College Journal, 40, November, 1969.

86. Tillett, William S.; Porter, Pavid K.; and Susan E. Joiner."The Peer Teaching Program of Community College-Studies," March, 1972 (ERIC publication ED 060 837).

87. Wessman, Alden E. "Scholastic and Psychological Effectsof a Compensatory EducationProgram for DisadvantagedHigh School Students: Project ABC," American .

Educational Research Journal, 9, (3), Summer, 1972,

88. Williams, Robert L. "What Are We Learning.from CurrentPrograms for the' Disadvantaged ?" Journal ofHigher Education , 40, 1969, 274-285.

89. Wilson,_5alph: "The Effects of Special Tutoring andCounseling on the Academic Success of Negro Freshmenat Southern State College," Mimeographed report,Southern State College,,Maenolia,-Ark., September, 1970.

. 90. Wright, Robert M. "'The Effects Organized Tutoring and ,

Advising by Upperclassmen with 'Predicted Unsuccessful'Freshmen," National, Center for Educational Researchand Development, Washington, D.C. Report No. BR-04-095,July, 1971.

91; Wylie, R.. C. The :Self-Concept. Lincoln, Nebraska:.University of Nebraska Press, 1961.

92. Yuthb,s, Ladessa J. "Studerkt Tutors in a Colege RemedialProgram," Journal of:Reading, 14, January, 1971,231-234.

93..i4e1,'Perry A. "Self-Concept and the 'Disadvantage' oft . Ethnic.Group Membership and Mixture," Review of

Educational Research, 41, (3), June, 1971, 211-225.

UNIVERSITY OF CALIF.

LOS. ANGELES

. NOV7

CLEARINGHOUSE FORJUNIOR COLLEGES