COMPARISON OF SELF- MONITORING AND A COMBINATION OF SELF- MONITORING, SELF - REINFORCEMENT, AND SELF-PUNISHMENT OF STUDY TIME ON TEST PERFORMANCE Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY CARMEL ANNE MYERS 1977 EilIIIH ‘IL‘ A“. ,: ‘ I/II/I/Il/I/l IIlI/I/I/IIII/l/I/I III/II/II/II/I/I/II LIL}. ~' R g; 3 1293 10394 5733 Michiga State University This is to certify that the thesis entitled COMPARISON OF SELF-MONITORING AND A COMBINATION OF SELF-MONITORING, SELF-REINFORCEMENT, AND SELF- PUNISHMENT OF STUDY TIME ON TEST PERFORMANCE presented by Carmel Anne Myers has been accepted towards fulfillment of the requirements for Ph.D. Department of Coun— seling, Personnel Services, and Edu- cational Psychology degree in Date 7 "q" 77 0-7 639 CONEARISON r. SELF-NONI“: pUxIsmms ABSTRACT COMPARISON OF SELF-MONITORING AND A COMBINATION OF SELF-MONITORING, SELF-REINFORCEMENT, AND SELF- PUNISHMENT OF STUDY TIME ON TEST PERFORMANCE BY Carmel Anne Myers Two self-control procedures were compared in the present investigations--self-monitoring (SM) and self- monitoring plus a combination of self-reinforcement and self-punishment (SM+C). A control group was also included. Different instructions comprised the Operational defi- nitions for the self-control techniques. Self-monitoring was operationalized as the daily tallying, recording, and graphing of study time for a designated course. Self- reinforcement and self-punishment were operationalized as positive and negative self-ratings of study time. The control group was given was given a task irrelevant to the investigation. Two independent replications of the experiment were conducted in a chemistry class and in a calculus class with 149 and 80 subjects respectively. The dependent variable of primary interest was academic performance, operationally defined as the sec-re on the secor itiependent varia} reported study ti: conitions (SM an also provided rat result of the pre their condition. istered to stude: in an effort to inStructions . In chemi first midterm e) students in the nific‘mtly bett. However, in cal obtained , Thus H‘yPOthe S i S l k I OfStUdents in s . upermr to co 2 Hypoth. self-monitOr a Punishment in ence super-10r simply monitor 0t Supported C HYPOtI Ondit. 10“ W 01.13 Carmel Anne Myers score on the second midterm exam. As a result of the independent variable of self-monitoring, information on reported study time was available for the two experimental conditions (SM and SM+C). Students in the SM+C condition also provided ratings of their reported study time as a result of the presence of the independent variable for their condition. Finally, a questionnaire was admin— istered to students in the two experimental conditions in an effort to assess the degree of conformity to instructions. In chemistry, analysis of covariance, using the first midterm examination as a covariate, indicated that students in the experimental condition performed sig- nificantly better than students in the control group. However, in calculus, no significant difference was obtained. Thus, only partial support was obtained for Hypothesis 1, which predicted that the academic performance of students in the experimental conditions would be superior to controls. Hypothesis 2 predicted that students assigned to self-monitor and administer self-reinforcement and self- punishment in the form of self-ratings (SM+C) would evi- dence superior performance relative to students who simply monitored study time (SM). This hypothesis was not supported in either class. Hypothesis 3 predicted that students in the SM+C condition would evidence higher reported study times tan studen not support Rep classes, he the conflic with regard dents' aver 40 minutes week EXperi Study time a day over to differ c Stu Ported this their StUdy the fact th in calculuS time! althc QUe ‘ v elasseS 0r 'c dlgferEHCeS then, to di It level is an use of self Carmel Anne Myers than students in the SM condition. This hypothesis was not supported in either class. Reported study time did differ between the two classes, however, providing a potential explanation for the conflicting results obtained from the two classes with regard to student test performance. Chemistry stu- dents' average study time increased from approximately 40 minutes a day to 2 1/2 hours a day during the three- week experimental period. Calculus students' reported study time increased from 17 minutes a day to 37 minutes a day over the experimental period. The classes appeared to differ considerably in difficulty level. Study time ratings from the SM+C conditions sup- ported this interpretation. Students in chemistry rated their study time as adequate only 26% of the time, despite the fact they studied a great deal. Conversely, students in calculus rated their study time positively 48% of the time, although their reported study times were much lower. Questionnaire results gave no indication of dif- ferential conformity to instructions either between classes or between conditions within each class. The differences between the two classes cannot be attributed, then, to differences in following instructions. It is suggested that course or task difficulty level is an important variable affecting the successful use of self-control techniques. Further investigation of this faCtOJ of significant ditions gives effects resul' P‘dnishment pr 50“ 0f covert and tangible Carmel Anne Myers of this factor is recommended. In addition, the absence of significant differences between the SM and SM+C con- ditions gives some support to the view that self-monitoring effects result from a covert symbolic self-reinforcement/ punishment process. Recommendations for further compari- son of covert and overt reinforcers as well as symbolic and tangible reinforcers are made. Other suggestions for further research are included. COMPARISON OF SELF-MONITO PUNISHMENT . A 1“ Partial Department 0 f COMPARISON OF SELF-MONITORING AND A COMBINATION OF SELF-MONITORING, SELF-REINFORCEMENT, AND SELF- PUNISHMENT OF STUDY TIME ON TEST PERFORMANCE BY Carmel Anne Myers A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services, and Educational Psychology 1977 COpyrig CARMEL 1977 Copyright by CARMEL ANNE MYERS 1977 This dissertation is dedicated to Gary, my closest friend and my husband. ii ACKNOWLEDGMENTS As chairman of my doctoral committee, an advisor, and, a teacher, Dr. Walter Hapkiewicz has contributed to this dissertation in many ways. I am grateful for his contribution to my understanding of social learning theory and his sound advice about my overall academic program. Most specifically, his interest, supervision, and support have been essential to the successful completion of this dissertation. Dr. William Schmidt's advice, direction, and assistance with the data analysis of this research have also been essential. Furthermore, his courses in statis- tics, both the content and his exacting standards, pro- vided the foundation of my understanding of statistical analysis. His expertise, combined with his sincere interest in students, commitment to teaching, and sense of humor, made his courses an academic high point. I am grateful to have had the opportunity to work with him in class and on this dissertation. Dr. Eldon Nonnamaker's membership on my doctoral committee has been very important to me. As a member of iii his staff fror many practical His advice and staff member a Dr. R: in? and acaden touchstone for he made clear i ideas and aCtir of social learr for the assumpt ideas and Viewt most limited tC So arid is refleCte FOSSible tO giv bution. his staff from 1966 to 1975, I had the opportunity for many practical learning experiences that are unparalleled. His advice and support have been valuable to me as a staff member and as a student. Dr. William Brazill has provided a model of teach- ing and academic excellence that will always serve as a touchstone for me. In his courses on intellectual history he made clear the importance of the relationship between ideas and actions--a viewpoint that is a basic principle of social learning theory. I learned, as well, to look for the assumptions that provide the parameters for ideas and viewpoints and that our freedom of thought is most limited to the extent we remain unaware of these parameters. So much of his teaching pervades my thinking and is reflected in my dissertation that it is no longer possible to give complete acknowledgment to his contri- bution. iv I. II. TABLE OF CONTENTS RATIONALE AND STATEMENT OF THE PROBLEM . Introduction . . . Factors Contributing to the Subordinate Position of Self- Control . . . . Self-Control and Academic Performance. DEFINITION OF TERMS AND REVIEW OF THE LITERATURE O O O O O O O O 0 Definition of Terms. . . . . . Relationship to Social Learning Theory The Self- Control Contingency. . . . Self- Control Designs . . . Self- Control Techniques as Independent Variables . . . . . . . . . Self-monitoring . . . . . Reliability of Self-monitoring . . Self- reinforcement . . . . . . Self-punishment . . . . . . . Self-Control Techniques as Dependent Variables . . . . . . . . . Direct Instruction and Prior External Reinforcement . . . . . . . Modeling Effects . . . . . . . Subject Characteristics. . Characteristics of the Task, Reinforcer, or Situation . . . . . . . Studies on Self-Control and Academic Performance . . . . . . . . Self-monitoring . . . Self—monitoring and Self-reinforcement. Self-reinforcement . . Self-reinforcement and Self-punishment. Page 28 28 29 33 36 39 4O 45 52 57 61 62 63 67 69 71 72 79 89 94 HI. IV. METHODS The Pr Logi The Hypothr Design Dependc subject Materia COVe1 Inst: REPQ; QUQS‘ Proced Proj Proj RESULTS Page The Present Investigation . . . . . . . 97 Logical and Theoretical Continuity . . . 97 The Specific Comparisons. . . . . . . 100 IIII. METHODS . . . . . . . . . . . . . 105 Hypotheses . . . . . . . . . . . . 105 Design . . . . . . . . . . . . . 106 Dependent Variables . . . . . . . . . 106 Subjects . . . . . . . . . . . . 107 Materials . . . . . . . . . . . . 109 Cover Letter . . . . . . . . . . 109 Instruction Sheet . . . . . . . . . 110 Report Forms . . . . . . . . . . 112 Questionnaire . . . . . . . . . . 112 Procedures . . . . . . . . . . . . 114 Project Development and Pilot Study . . . 114 Project Implementation . . . . . . . 115 Iv. RESULTS 0 O C O O O O O O O O O 0 12 1 Academic Performance. . . . . . . . . 121 Experiment 1--Chemistry . . . . . . . 123 Experiment 2--Ca1culus . . . . . . . 125 Reported Study Time . . . . . . . . . 127 Self-rating of Study Time . . . . . . . 135 Questionnaire Responses. . . . . . . . 138 Conformity to Instructions . . . . . . 139 Student Opinion Items. . . . . . . . 142 Correlations between Reported Study Time and Midterm II . . . . . . . . . . 142 'V. DISCUSSION . . . . . . . . . . . . 143 Academic Performance. . . . . . . . . 143 Reported Study Time . . . . . . . . . 148 Self-rating of Study Time . . . . . . . 152 vi Questionnaire Responses . . . . Conformity to Instructions . . Accuracy of Self-report . . . Student Opinion Items . . . . Related Issues . . . . . . . Demand 0 O O O O O 0 Prior External Reinforcement. . The Monitored Behavior. . Proportion of Self- reinforcement Self-punishment . . . . . Isomorphism . . . . . . . Conclusions . . . . . . . . APPENDICES APPENDIX A. CALCULUS MIDTERM EXAM 2 . . . . B. CHEMISTRY MIDTERM EXAM 2 . . . . C. CALCULUS MIDTERM EXAM 1 . . . . D. CHEMISTRY MIDTERM EXAM l . . . . E. COVER LETTER . . . . . . . . . SM INSTRUCTION SHEET . . . . . F G. SM+C INSTRUCTION SHEET. . . . . H . CONTROL INSTRUCTION SHEET. . . . I. SM REPORT FORM . . . . . . . J O SM+C REPORT FORM. 0 O O O C O K. CONTROL REPORT FORM. . . . . . L. SM QUESTIONNAIRE. . . . . . . M. SM+C QUESTIONNAIRE . . . . . . LIST OF REFERENCES . . . . . . . . vii Page 153 153 153 155 158 158 159 160 163 164 165 169 170 174 175 179 180 181 182 183 184 185 186 187 188 Table L M. 1L M. Studies c formanc BXPerimer. EXPerin I LeaSt Squ Standar l Adjusted in Chem AdeSted in Cale Average D; by Day, by week AVerage De OVQI tht C0nditi‘ WEekly Av‘ Conditi< AVQrage Rd dition ; Average Sc COUrSe é SseSSir OUrSe a LIST OF TABLES Studies of Self-Control and Academic Per- formance . . . . . . . . . . . . Experimental Design and Number of Subjects by Experimental Condition and Class . . . . Least Squares Estimate of the Contrast and Standard Error by Hypothesis and Experiment Adjusted and Observed Midterm II Mean Scores in Chemistry by Condition . . . . . . Adjusted and Observed Midterm II Mean Scores in Calculus by Condition . . . . . . Average Daily Reported Study Time in Minutes by Day, Condition, and Course . . . . . Average Daily Reported Study Time in Minutes by Week, Condition, and Course . . . . Average Daily Reported Study Time in Minutes over the Three-week Experimental Period by Condition and Course . . . . . . . . Weekly Average Rating of Daily Study Time in Condition 2, SM+C, by Course . . . . . Average Rating of Daily Study Time in Con- dition 2, SM+C, by Course . . . . . . Average Score on Questionnaire Items by Course and Condition . . . . . . . . Average Score on Questionnaire (Items 1-13) Assessing Conformity to Instructions by Course and Condition . . . . . . . . viii Page 15 106 123 125 127 128 129 130 136 137 140 141 Figure L Reported COmbin LIST OF FIGURES Figure Page 1. Reported study time by week and course for combined SM and SM+C conditions . . . . . 134 ix CHAPTER I RATIONALE AND STATEMENT OF THE PROBLEM Introduction Self-control techniques, an outgrowth of behavior modification and social learning theory, have been utilized to obtain behavior changes in a variety of sub- jects and settings. Increased interest in these tech- niques is a recent phenomenon. The present investigation compared one technique, self-monitoring, with a combi- nation of techniques, self-monitoring, self-reinforcement, and self-punishment. This investigation also compared these self-control techniques with a control group. Such Comparisons were required for theoretical and practical .reasons which will be discussed later. College student <=1assroom test performance in a calculus and chemistry <=lass was the dependent variable of primary interest in 1:his examination of the effect of two different self— place apprOpriate limits on our ability to generalize--ac example, we ge individual behl of individuals too willing to definitions of as a whole. we as the self-adn accurate reSpon Proceed to talk Whole works or forcers are bot l events may Ope responSeS’ in C generalize--across subjects and across settings. For example, we generalize from studies with children to all inndividual behavior, and from case studies to behavior (of individuals in groups. Similarly, we have been all txao willing to generalize from specific operational qumeu .ooccfiuowuwm umou mcfi I>oumfia as obfiuomuuo ONCE >Hucoofiuacwflm no) ucofiwou0ucflou «mcuouxo msad one succuucoo xocomcfiucoo uozu ocapcflu 0 ca nouasmwu umou mmmHu msofi>oum co woman ommuo>mn3odwn who: 0:3 macho beaconuucoo ca mucocsum o no mamxamcmou .m Houucoo .m cosmos mama c“ ucoaoUHOu neaou accumuxo nus: was mango uuoauucoo >ocoo xoofiozoxma >u0uosc acflucoo mafia OUCEOMHOQ ummu 00.»qu cm EOMM vCflHOuficglwaom .N whma aaauo>o qu museum m mucoucsao> ucocsum meau0uacoeiuawm .m .ocmon onu Ca wocoumwuwc Mm .H wucmEDOuqu uwmu .a ovmfifloo mm uncomfiuoafiou QSOLO n maMOuacoavaom a HOumfium msoum Houucou wudsvocmcfi CD can canccofiumosv muasmou Ham .wuo: acuucoo nanosecoou uo ucmsumouu o: .m seduosuumcH maawxm >osum :ofiuchnEoo qu nmm .N :0wu cowmmsowuo Houucoo madnesum mumELOm cofimmsomap ucuumflvmu umafidm macho .N ucosnmwcamuuaom can ambush: :uon ocsusc pwchubo coauusuumcw nauseouOuCMwuuuaom vbma can mosqaczoou no omwuo>o mucopaum wooaaoo couuwu) .H oanOuacoeuuHom .mwuumm :Owuwcfinsoo uOu IMMN .a ucfioa spasm .H nooucsao> mm uncomaucmfioo macho m "uo sawumcunfioo a exocom assuaux uo aces bassoou0ucaou umwmmo Hmcuouxo .m ucmsmuu0wcwou nuaom mafia ocfluOuecofi xmmu co Esau .v muow>mcin cououmHCchmuuawm pan twacm uo cofluccan mono: mcflnwuommc 4 .m mmouuu mafia cocfifiucumpiuawm msam mnma I600 mecca moancwuo> mono: coauoc . .N roman oamwuass ocfiuODACOBIuHom .scxao unoccuaeo Has new IMMN moocoucom g .~ muocauo cum «a immacuo H waNOuacoaiuHom a cucandm moswacnooh acuucOUuuaom Amvoanmfium> Amvosmwccumk new couuoaom uoouum unaccomoo muomflndm cobweb Houu:00cuaom Amvuocusd mocmauomnom oweocmod was Houucoolmaom mo mowcsum H OHQdB 16 uaouusoo cozu o>fluoouuo whoa can o>auuwuuw >H~czvc who: masons mcwuuom 32953.18 .33..wa .v 333 5 2.3 .v \ >Huoouuoo mascum co>aom maoanoum Houucoo ocean oucououuwc_mm .m uo womucvouom .m ucoEUUMOucwou 0c .n . Aauxosc ucoaoou0u:«ou new mucosa :0au3H0m awanoum moumpcnum posse mCOEs vocououwfic.mm .N uoouuoo mo ouau .N uuwuoc hmaacuuuxo .N maouucoo coca o>auoouuo one! can o>au ucoaoououcwou noomuo >afiosvo who: posse uOu mvucflcoum mnma nucuop xaaucuouxo can mcofiusaom cages cocafiuouocluaoa .H unaccoun unuauououcaou .>Hdoq.o vocwlhouovluaoamex .H iguana uoouuoo a .H auocuuo uc~ vm tecoauudmaoo macho m vosaauouovauaom a vounu4~0h accusuou ham unadum mafia usoavoqucaou tuaon .uonuuufl ucuBoUHOucuon meanowum> unaccomoc radon uOu ucufl such uOu insane locuOucuou aacuouxo Hoconaouo can ucos aauVOHQ coxou .ucosoouOu noUMOucfiounuHon no sea» doocom youuc :0 ca Inflow accumuxo uacanfioo noun: coCHMD usmuso uaaocuoa .N macanoum Macawuoao coxOu mcavsuocu mnoa issue can ucoswou0ucfiou flan caeuooon mousse m cud: .hhdoq.o a Accuouxo umvcs vacwduno qu>mn®n sud) axon mwumsumwo .H 0H0 Hoax euro @ ulddbuo a OCMuOUwcoauuaom .cmsnduo unequapCOU mcquuacoaiuHom nus: caduceucoo omwsum neaom ncsoo mo HHON ecu c“ IOOMOquvu ascuouxw scaucm>uouca Hmcofimmooo "ouo: . . vacuum 0» 6060a «nod Hmoo>asm6 .N uso ocwxaouuusum .N onwcum nonooou .muuut mmsao :« .mosunulnqu .H muooauo cum N Il>vsum once mcfluouficoaiuaom .GOUOun $23208. Houuaooufiam 333325 333538 you 633.51 vacuum 28:38 383.6 .528 83:8..38 3:853 i.uaooo H manna Ti:- .l.-.I ..I—._I s . .-I I a...) l.:..;7 .3::.I-..aqu—. —-:-\ I V — :1 .-.:...vh=&u «u. ~€~:..E HCHLHSN ha «1: I. :2. CS...) .62 3C .E..:—IEII («hove—I5 ul.£~ .IlI-IuIG—i... «I. a Z. .I M. -.I~. ~;..h uXU o -..lI—.-h 0 Funk“.- -..I- ..:-.,..I.u - 1M“), liq; ~3 ~5l I: . ah! «.00hh.7. 1:1. 2. .II.. \ Quail . .— .. I Ii I .Arafl-£€ -..I-.> u...u-..~..I—.Xh xu «Sin-DIS .N 0:.1'Eyah3u: n EL Erik 2:. - ‘1an u I 56: «E3 til; ~€I . ~ IIECIun-rh‘avu «Susan? A. CLIVQHO UEH 'N l «L... 5.... «iUCOUv H. SHOE?! En» a t 1: BL 5:580! uEIIIDHOh 2‘ IN «.9: ‘E& e finvlh ~02. a I C u... .‘w‘ Misfire-L. a .I. . 5:5; I n ~ 6.1... t “var-film ~Ik R.» fiPls sq.‘ 17 mango Houucoo 09 can oocano ou uOAuoasa use o>wuoouuo xaausucu>o unmawouowcwou pocafi auouocIunm can codes Iuouch>HHECLEuxoLMMN aduOOMQ oaonl u0u momuouocw ucoofiuwcmflm “vaguOchoE Iuaom uOumemmmwmmmmmw muouuo couscou cud) couoaoonmu OCHMOufi:OEIu~mmImoN xmou us wawu can mCOwusaom uowuuoU a pounce» mmsoum H~< .00ah0m cofiuoc«uxo :« monouou -uas ucouauacoaa.mm wanna aaucaafiuomxo weapon o>wuuvuwo saaudvo mum: onus was How» Icoo coca o>fluowuuw one! mus) cocdauouuv xaaccuouxo was nwcafiuouoquachmmx umdouo vocafluouoo aaaccuwuxo new Iuaom coozuon oocououwwv on uncoammou Houcma Iauwmxo wound uOMIMMN mUCMEHOHuUQ umuu nocwum ocfl>csum Esau muouuo caqu£uaum c xmuu be was» .m n:o«usH0n cause iguana uoouuoo t .H ucoaoouOu Isaou coxou o: .v and! lookoucwou coxou confiauouovIoocuzu .n uses IoouOucaou coxOu vocafiuouovluaon .N acoaoouou:«ou coxou cocashouwv maufio >~umcuouxo .a ocuuo cum mNH IIcoawuaoaoo macho onoouocw haauu pouowuucoo 33am UCMHOuwcosIuHom In>csum omen uomucsao> acousum ounscauo mcwsmnuu mafia xoonvoou gun: ocwuocuouaa ouoom uouuuoo cu no xoonuoou and: mHdncu0uxo .N usuaouu0ucaou new unhapcaun vocaluoumvluaoa .H muopaum cum vN Incouwuomfioo macho n ohma uCQBuONOquoquHom .c:>HU ocwuouficoaIuHom woma .xom mnma .ocficumu ocauouacoauuaom a gash vhma cuspsuua ucnlouu0usaou .hhuuq.o vocaluouovluuom u voanlddoh moawficnooh HOuDCOOum~om Mow oouuoaox uoouum Aweoanaaua> ucovcoaoo uuoofinsm I avenue A.ucooc H manna .mvommwczooe HouucoOIuHom Anvuonuss a... _ $23 I: ~ 3 #1.). ll'u CG .1le ..I....n.......e. ....>.-.- u:rnv»:;u.—:!A~ >«23 leuLhChIa l.- ..5 upon. LCNI-MGIN ”Filalalmlflvqu .4... «I.-I.I.IJINI~.UI.nu ...L u...~u.ln.h~n a..nv-u.- « INCOUV Ina-Ulnar Vifl IS-uiihlu‘ 3 t. ..:.. £22. In. MC» 7: q 33.3 I: Tm u.~..vl-$L ~3.~:w a a In ~ ..:.. “EPA” “Phrguc.v-V‘-~IIIN ._-.:~ Iu~au £..) I893 Pia C1: u... ~ E ‘3‘? tovw\ A!!! IIa-aruI~° N NIHONN’H :3.I-.~ «.39 U-.I~.I~I.:~ 41w. —..- hungry .£ nun-I URNOUMONEflOHINHOQ ?\3 H airy-t; .IL. 0 .2.) 7... «I» l.vn~s:< 18 vouuomuu aocaum van uflwu undue ca usuao>oumaa Aacuo>o .vaAMom HOHucoo weed Iowan mo mocwmnu can wozdwcnoou no nodumcwn I300 we own 0» can osvwccuou Houucoo umucuaauomxw nus: macammso Imap haxooa can .madxw .mouoc uo 30w>ou can Iuawm ou cucoou museum .m uoouc=~o> o:«u0uacosuuamm mood cuM3.MNMNMNMHNMMM mafia >csum .H ucocsum cocaaoo Iu>05ua onuo washouficoanuaom .Ucosdacaoo .538 mou accuuusuumcw :ofiuomaowu voww qum caucuacoa .N tacos mason nod>cnwn musomoup Acqucmuoa co«umm«owu «bag as omcouocfi as nouasmou ovaonma mwuoEOLQOm Inna caucuficoa .A .adcmoz ocauouflcofiIuHomLMMN cowummwowuuom mmoao Hoonom saw: bu acmuumasoo macho m ocwuouwcoEIwHom a coauuoo nmwucoOCAu Icoo Houucoquaom douucoo Hmscw>wccw saw) Hucuouxu have: can mwwocomcwucoo bonamuno UUCMEMOqu mmnao Hmeuwuxo saw: muma vo>oums« pecan» cm no mmcao ocwumcuouaa ocwa .oonm w Icama moszcnuou mo use vo>uwmbo Iowan dint coauwcoa usoauouOusaoquaom nuaosa HouucoquaomIWMN xmou so was» muocmum vcm m IIasouo H meauouwcoaIuamm .::>HU qu>~£0n ouowumoummo uOu mcwmsu scam usuaooMOMCAmquaom was ucoswou0ucwou nuawm no“: ocwa Iowan mcwumcuouam whoa >aco mcwwso mafia aces muoomuo tam .m¢m¢ vowwquE .musona Ioou0ucaoquHom wouImo xmau :0 mafia o>wuasumwo o IInsoum a unusmououswouuuaom a asaao megawanooe HoquOUIunm Am.oanofium> Amoomwaczuma uou couuomom uowuum ucovcoaoo muuonnSm coamoo Houu:OUIuHom Amvuonus< 1.»:ooo a manna -P\ “\\- (In—Luca... .:_..~R~ 5;: l~...~u:3-J UZqOIIh lfiruiu NO :Swu :(Zu t>_ ..:..»... ..:..: Cw NCO; this. Isa—£53.. to?! NON-.233 Q3~DEq «mu ..nzuhna —J:C A92) IMO—Jr?!” £50 u~nsE.-\.P.:EIM~.:K .N U—a.vE.§..h»4LC~U~I\~—y% \°-‘m:~.~u 60lh0>¢ ups—Can ItIhD i £b¢ .£uh n-~ IICCD‘L-Inclrv £280 a. Otwhcewtnlllh~9m i l‘hI-Qt . I -IIIIII I I IIIIIIIII II I II I III. I IIIII ~.tl».-...3d...~.2.:.-:5. I III II- lc-JifiZSZ 53:!!4 ~..~.:.... \nir. alab..£e2< «II» {-411 -(> ~::.Z..w._.v.~ AQUEOUV H NHQNE CC«.-=:..E .m. Ergo-Tulug L ~SI II.....N IIed-Jflaa..m..-.:I.I~.- ~.. ~ ..:.. VII h —.-.n.. I I ~..& w...—-.~.—.‘: g ...LL..— 19 .poospou xuaaanoaum> u< mo momma Hmcflu no>aoowu mmmao uo oome>wmsaucoo:« aouucoo ucoa Iuoouu 0: van uo«>unon weapon m:«u0uwcosluaom ou o>aumHou vo>ouafia xaucoofimwcmwm 08H» >05um mewuouacoaluaomimmx .ucomnd acomfiuugaoo Houucoo madaumoummu .u0>®3on .mouoom Uw>ouafiw Lug: nauseoOmmu ocacocou uo ucaaoouOu newoquHoqunoo>ws o usuaoUHOucflou 0: mo COAuoNHHuSumwocoo Qua: Edanonm ou mac COMumosv ca mufismou ”ouoc >aco H .me cw .HOuucoo vac ucoeoouOucacu 0: cu o>auaaou ucoswouOucficu consumacfiavs xaac Icuouxo UchuHom uOu maucmufluwcuam U0>ouafia mason moonuot xuozLMMN xcsum uwzuao as condense mwocououuav.mm naouucoo can» o>auoouuo muse coflmmsomwc macho can ucoammqufi uHouImo .H onusoo we new one x003 nuoa wo no mocha onusoo .N uwuumav mo x003 nub no no ovuuo onusoo .H xnouco>cu maaaxm susum co choom .N o>usm co ouoom .N oonuo>o ucfiom opnuo .H omauo>n unwom ououo omusoo >mo~ozo>mm >u0uosc0uucw as ea mucoucsuo> ucocsuu omoHHOU so amazon .Haaxm snag» a new auoou::ao> unopsun comaaou 0m Baum loam maaaxn >03um a MOM mucoucs~o> ucocaum ouodaoo mm N .mxw [duo noun aaawxa >vsuu a uOu mucouCSHo> ucmtsum mgwaaou nv MIammw mascaou ca Moog who) on: auvcaum cum a 5m :3“. n: «ouucoo uoducoo o: .n u0w>mnon assuao “OuwcoEIuHon .N u0a>mcon >v3un uouacosIuHoa .H couwudoaoo macaw EamoOMQ mauwxm xvsum aaaxn 05 uofioco assoaou uo >uw~usv uo ucoa Ioquucaoquaom mafia aduoOuQ uaawxa hosua .H uIcomau¢QEoo macho N Houucoo unusuaouu 0: .v ucua IoouOucMou 0c .m acosmOMOu Icuou Hmcuouxo .N ucoaoou0u Icaou vououma Iceschuaom .H coawuomsoo aboum Houucoo .m nuundn >csum uo coda Insomwc macho nosvwc I200» «0 :OAu Iucflnsoo saw) ucosooucuaIuHon .H N Incomquomfioo macho n ocwuouacoaIwHom ucoaaou0unwouuudom ucoaouuomcwounuaom «onucoo usasaaum ucosoouOucwoquaom ucduouwcoaIuaom Hams .ouazx a COQEOH. whoa .unoon ca> a consonn thd .umooa cm> a :omxooh mama .oaaaflaus a manuam mcswwcnooe HOLuCOUIwHom uou pouuomom uoouum .mvombmfiuc> unovcoawo nuuohnsm nuance 1.»:ooc a «Haas Amvoawasnowe Hahucootm~om .uvuosus< I .._...1;. ...l t....__..o:._.~ I3._..n:...~I- .1... ul...-. «1.! ..:..)SL .....al 7:... ..:..E:;...:_.~ I n..: . nth .. .2351; ...u.....h \. .E.ll.-L~C&: ...- ~12 Ta .1..- Zual n..-..q-.._~C «Hauling: NICHOuln- ..:«utlrld mutt-..:...uuzL ....>..~.—E. U~h~'.«£..>.<.— ..:..95; 6 lb.) Yanu‘ ’2‘ OJ.- IG:~.~:Z- ..s~ . In I— ~32; n(—2 VII“ ..:..:- UCE‘Id-uqaailhflavu skh'MKQ ~ ~C-u..CUIu_§lIIOIN u01)2£ I)HUQEI1V UCOUIG~SU ‘— IIQSCHD N utmIIOIahCE:~ILIN-OIO i think-“luv”. III-31......II .ce.~. —.-.- ~.-.v.. .u-wnhmuil . II I - II II gnaw .-.~In\q --\.> I II.I IIIIII .. III I I I I. I II I . II I I I. II II IIwfl-v atrium-:45 7-“... III I I. ...h -.o- u a. 1‘5; o-El-u Lu- ...nI-.:E._E.- SOUUDIA .Au-un .unv‘I’H -- ~:.«.d .\ uaubn R's lav-s u-< A -UEUUv H OHQNH. 20 maflanoua Hmu0a>mron cue: vhma com an couscouLMMN uofi>uzwn umoququso Hugo vac hoax AH Inxcsun ammo mcwu0uacoaIuHom .xxwuoam: uoHuomsm mmsouo acauouwcofiIuHOmmex .m >omuaooo nuns .m neocouowuac unavauacowm mm .v >ocusooa .v Houusoo neocououufic ucoowwacown.mm .m boom» .m souuulu0ucd on .v 69333 2.3329. 96 noocououwac ucduwuwcoam Mfl .N macaw no noses: .N Izuwz xooncoou uoouuo mucush0muom .n ozu coocmnco m:«u0uwcoa Ammo. meduOuMcOIIuaou nnoa Inflow uaoscwusoo noduacasdxm vuooom ucouufiluousd .N .nuso: can 30w>ou be old» ouusvuuu how IDq>ou washouasoluuaon a .0663 ouoa ucomm nuuwflndm a now mucosa—5H2, nsgcaucoo ..n .930: mcauouwcoBIuHomLMMN .H xmou co oawu .H ucovsun oouauoo hm Iucouwuamfioo macho v ocwuouflcoaIuHom .aocoza: unvaouu0ucaou no ovsuwcoms you Ola on» voucavo can .mcumvcaum ucoaoou0u ucoum Isaak vomomfiflluaom Icoo saw: me: ousuacoms unocuOmac mvumccuum ucoaooqu ucoaoouONEAou can: co>o Hau0a>nnvn Inflow canons“ aaao moon vomoasa adducuouxm 0» cu“: comocmawp Icuouxu bouoasoo mouovcaun ucma .nwuuso noduwmsm pomomszwHomLMMN ounu oncommou oflfioccoo 0H0 umo> NH Il>csum mono IuouOMCfiwu vwmomsauwamm a uuw>oq cocacu qu>unon assuamucwua can oceanmaczm vcwcwouno ca o>wuoouum Iuaom\ucoa aaansvo who) monsooUOum IoouOmCAoqudwu umoo use cusses .ouoa use .ucoscmazsn Iuonuusu «ucmfioouOucwwu \ucoEQOHONCMOM Aucuouxo cud) vocaouno Amuwmnon aacuouxo.acwaonon OUCEOHHOQ Um>0unwfifi Ufiuunflsuxmm woman—Q o sud) Whoa vocwoucwafl mosvacnoou a as «Hausa manon o>uunshnwo ucoonoHooa 0H uImsouu H ucuflouu0ucaoulwaom a :dflusuu moauacnoos HouucoquHom Amooansaum> Amooawficzooe how vouuomem unamum unaccomoo uuoonnsm shaman nonucooluaom anyuonus< i.ucooo H manna ..1 .. .- .- EO~E .I 5a£¢2 .Aw...g¥. 7:... In::~vpf XI.) C: ‘,.:LC 23‘ a ISTEL? .N . ~ 5-2:.qu >~.:¢..qu_:?.l :an- Esau—..:...IC >52... v.25 Ion—LUICd I— ~‘x: Eta ...th ..:...IL .SIII‘N . d “I: - .‘v. . - m. ...:L. wan. u—M-W-IIILI— .vmu I:L runny-..:..»: «LSLL; Since-v. ‘ugmu .. «I-» .u _ 29.. ans) ...narzu‘..nx~ a... .u T; :29 .V as: dun‘hb 2.52M 1:. ~3huCCU ..:..; SEER? “Zine-ring CC .8 3.: MO Italy... DtuuLshaciLIu—fil .N II.~‘~.~‘.~ ‘59 Ill CCqubafnuliw OZ) ILGE...3~C5 l-_xz >23:- .a IUQ..UILPLOG¢€3 an. IIEI-I-SDIEC :30an V tun-II: a .0530» NLUUAAJ' IN. QMQN-H IOQxb!‘ 3: ~ ob! - «:30 Ihuva Q 2‘ ....lt..-..:c:-D.~ IN~CI n5! 02-hCe‘c5llunInn I. I I. I..I..I»I.rI-I~ ..H. «NI-:xltfi. ‘ R 0“ ~p-h gnu-r u HR 3 N \ CUHDQU 4 I5 ~5CXIKU! «I» ICCHI< 21 ucuamouOucwou Hucuouxo .oauouocH vouosuuncunuaon sad“ has». vacuooou .n omoouoca couusuumcanuHom .muaouuca vouosuunculuHom .30 >33 poo-H003 .N «as» ausun cookouou x0e) a undo: oH .omaouuca can» aon coacsum vouuauumcwuuHon $5 63835 as: on: onusoo mooH xvnun vwvuooou .H I029?a >u0uo=cOHucH “axon: n uo>o HnaH ocHouoooquHom Ou uoHuwmam no: Loan! unmouocfi pouoshumCHIuHom ou uoHuodsm ma: ucos oEHu xcoum co souu mucoucaHo> moocosvom ucwuwuuHc n .HQEHO a noONOuCHou Hacuouxome mo mouscwe no a unocsum wooHHoo mm IncomHHanOo macho m ocwuouwcoaIuHom .c .uoHHH: wouuo>o ucHom ocoum co cenwuuasoo uwoauoum acaoauwcwam >hoasbdoo> can :0: .uofiuomsm mHHme .coHuconoumaoo accuu can ocfiuoouucoo .voomu .ocwvoou Iuaom saw) wucfloa combo nImsoum ucaa new moocouwuuHc msouu uuqouu uvucoo ucuowuacoanuwmmmwmmmm .m nououm .m coca achoncoo .v beacon vandx xpsum pea uHcouo vousvou can Houuzoo acozuaz xuoz co ouoom ucHoa mocha becauoouu o: .n uocowc aHucooHuwcmwm )OH NO mucus mcwuomuucoquHon .N nu“) vouoHUOuuo :oHu Iouwsvou onu umfl coHuosuuncH coaxoom unauoouucoo mnma nonhum:« nHHme xpsun can on) uncoucsao> aanxm xvsum .H IwHon a :w acoEoUMOMEHou .sousgu a ucauoouucoquHoquo .H ouoon (mom .H ouascauouoocn on IISOoHuooaoo maouo v Iqum can onwuouqcoaIuHom ucHonkozu: noswccooe HouucounuHom AmvoanHum> Hmouamacnoos now couscous vacuum unaccomoo uoonnsm sounds HouucoquHom Huvuoaus< 1.268 H 393. .. . . . . ”a IHCHLI 1.3.2:. ...CE.Q.._. EILC-CL .-Q~:>3 ....:¢ J..E..Hu...?ul Cm. .N 33.50 6.21.9 .fi .~ .u '5'. nauh Ilhqoan I. .-.I.-3......... :S b..Ew-.-of..u.~ Ih'u u «a. .....>~.¢ t. u «1! >1: a... POLICZZS >~ ...G..-.u ..:..—l 05— h... ..:...-.» ~6IIM$N fwuq.<..lal..._.....—. a: a ......v. .u a v” .. . . II a.»& ..n--.v.—..= s .o-uu: unfit—>153 >72»! HE’S-THE.— Ca Chou-.23.. Cuw) III—23,; >?.~.!~.c>t.m >LCI....VCL.~:— 2' ECuh II-ovnsa::~:> LCSUZGI O3O-Cb CA. I ..:.. £21 I ICC! .H MEL—ECU 53.311 .n ..H H 3:. 3.3“: ~Auh~=p~na 0:05'IUNU .1. .N ~....— ..:... .v: . ~ ~o-Nd-.7v .N—Nzfl—w on WK 3 N b:-~:..:QEI\Hcm .Ivhlc-iz NICHvIZHQZZI-IVWHI II IIII I I I \IV l.-§b:‘ U...-. ...:L 3:.th D uauu 9-hprv..\ NE... :uvql.l~ .H Odo-«6.5 A OUF~OUV 22 .mucousum ou&uo«:«:d HOu voucmm Iona nouoom adxw so unt Io>ONQEH uouuwuo mo mocoow>u QEOm .Amucmusum voELONCH no Ho>oH cu. ommouocw umomown Umaonw x003 umuww ocwusc couuomwu wawu Hnsuoa sown ucmawuomfic aeocusvoucoo uuo>o can uuo>oo .u uoososvoucoo uu0>oo .n moocmEVomcoo on .a "mwocmawomcoo vououmwcwsda Iuawm mo nomxu m sud) commouo xocnpwou newuqa umoa who: QEHu >csum mo cusp IuOMCH 30H nsmuo> mounEHUmo muons mucoosun .N unau0uacoaIuHom .N now: sud: xwuuma .coHuuHschms nx~ a CH 09.0.3 xoonvoou :oHuufiuowca ucauouacofiiuaoa 30H\£man can no: Edam 00w>pa .mcHuODHCOEIMme no maOuucoo nooucsmo> maawxm >Usuw o .muv uuouuo oocnnco mmm can Icoc uuuucoo o: m oow>ca .uo>ozo: .moocwsvomcoo .muabdn >05»n «Hawxu avsun .m consumaCHEUaIuHom o>0umsa ou :uoocoo Houucoo whoa .mHouucoo Ou uoHqusm a nuaz onusoo >moH unusuaouu o: .N .couxom an) noun) 00H>co mHHme Ionoxcm zuouusuOHucH Houucoo .uaoz xusua ou uoHuomsm one: an noun mucouc5~o> unaucoo o: .H usuaoouougwoquHom .wvaoc>mmu: mmdouo mcfiuouficoa uHonLMMN .H deo omusoo .H ucovsun flooaaou hm IIcomHuamfioo msoum m mcwuouacoaluaom .wvudnoflm bouuomou unfinccoHumosu mascuo ou Howuco coHunsHm>o IuwuuHo coHumEhoucH mm .v unusquHUIumom .v oCHHOuHcoe ofiau uo>o Iunm can Houu xHucu0wwwcon was» >usum Icoo maasawum commouocw museum Lyon nsHQ oofi>pm “vandmmom macawuomaou mHwam >v=um .o uwzuo on no can m mGHHOuHcoaIuHom mmsoum cowzuon mouse sumo mama 00w>oo Iuouuwv vasowuacoau.mm .m ocfluouwcoanuHom .n mHHme >u=um .m osvwczoou aHouucoo Houucoo unasadum HouucoUIwHom on can nooucsHo>Icoc Esau 00H>pm uoouwo o: «mHouucoo uouucoo o: oH mHHme xcsum .v canu uwuuon xHucuoHu “uo«>anon oow>vu Iacmwm masons ucoaumouu xpsum o>ouasfl mHHme xvsum .n commaHHoo "HHmuo>o coco ou cuoocoo zuaz Houucoo abounds ucoofiwacuwu.mm .m ocaum onusoo .N onusoo abodonoxnm unusuowuu o: .m moa>cm mHwam >03»m ShouospOMucw acuucoo voocccco xHucsoauHcon cm scum mucoucsHo> uocucou o: .H Houucoo wadsawum mnma mcHuouwcosIuHouLMMN .H deo Hacwu onusoo .H unopsum mooHHoo om IIcOmHudosoo macho o mswuouwcofluuHom .mouucuam moamwczooe HouucoquHmm AmVoHQMHHM> Hmoumwfinnoob now vauuomom noomum ucoocomwo muoonnsm Cowman HouucoquHom .mvuozusd I‘lI Tyson: H oHnon. «Shy 23.0 . n. QCQEC-a~3h ICth «MEL-val...» .N VKMN \N“W.‘afihfl «Eur-Cu. Efliu U>q HUSNHO 8&3... «..:.. o.) u 9.05 .6 IiOIa.—3 —C> ~...xh.....~..-N >~u¢2~et .....|It0-.C..1H1h Uiuiunfilaru acresul I.:..._Iu~e.l .~ now—uhiulno ..:..» Ih~€lIIIN I‘Odaquhnn UdaU‘ZUHLI . COUNan 0' IICOIi-NNHIRVU .wUChO H UCEIDUHONC‘IININN Iii-JCIZU-LFI.n-D.mI.N..IN..I.L._I..I-I»Iu II I III III I lall IlI-I.~¢..w.w-EI>I I . IIII. I II III -II II I- II I, I IKE...» 331.~5..~.£..8.H I I. ...u Troy-.273.- .Lfishna «II-..:.....l. leLflqQ-li :5:le ~3-.:_ru-.t~nz< «lshaviuzp‘ A IUD-owns HI NHQNHD 23 cemHumaaou u0u msoum Houucoo mo accused was ucoauouowcHoquHow uo acqquHHusumoucoo CH uuuwxo fipooum .mQSOHo wanna HHn :H o>Huooumm ma: acoEoUHOwcHou IuHou uunu consHocoo asap muonusa "ouoc nouoom CH omuouocH acuoHuHcon vouuomou wasouo m HHa “unscho .uHuoaov no anon 301A cuao ou huHcsuuomao bosons» use-OUHOucwou Hucuouxo aono) vovsHocH onuaoo mHHme hpsuu "one: OGHMOuwcofl uOu ucuBoUHOucHou oatnomusoo uHHme xenon .n uoHu IH>Huoo HusunHH venouHcoI IuHmm “cu acolouu0wcwou IMme msHQ ownsou mHHme sways .~ QOHUH>Huon xusua VououHcol IuHoa uOu ucofioDHOu Echo IcwouluHou uzofioUHOOCHoquHon IOHQ aHHqu >vsum usHm onusoo van .ocHuouHcoaIuHom u0u whoom a u0u uncoucdHo> uHHme >vsun .H unwaoUHOucHoquHum .:0mxoan sophomou ucoamocun:0.mm .H manon >o=um (:mm .H ucocaum vooHHoo mv cooHuumfioo macho m OCHuouwcofiIwHom a umoon cm> UOHHOQ spoon chcosHucmxo nucoa ocHomou umou ucos m.v :H .mos m u0 cHuo .m Io>onoo uHCHOuHHaU .m poms Henson :H onmouocw ouoon mchcou mchoou cH aumo> auumoua HoQH o.~ :H couHSmou CH ucoao>oumaH whoa no H vcwzvn usufloou0ucHou Eduooum ouHuco .w Ho>0H Guano .N can umoHu uses coxOu omens mhmH .cocHuubo coHuo=pou Iumsflpo HucoHuHa NH saw) :OHaoU .nuHEECOMH owcHuuchs moschnoou Icouu CH mcHo muooq.o HouucoquHomImo .H uoH>onon 0>Humsuch .H hoax HHIh m Iomsoum H ocHHOuHcoEIuHom .NuHauxuse Houucoo .m ucosoouOu Houucoo coca 0>Huoowuo Icku Hocuouxo .N duos use o>Huowuuo mucoussHo> acoaoou0u HHHasvo unosDUHOucHou couoHQEoo ucovsun ucHquuHou .H van Hucuouxo one IuHoquo mBmHnoum uHuolcuHum a oooHHoo mv IicOuHuamloo QsOHO m ucoaouuousaouuuHom .HooHodm ummwcsuaa HouucoquHam .mvoHnaHua> Aacmsuchooe ham vvuuomom uoouum unoccwmoo nanomAsm :onoo HonucoquHom Hmvuonusc NI H.ucoov H loms 24 OHQdudmaoo >HouoHQBou uoc avoHuom coxou can ocHHomon "duo: .ouo .coHucmHoHuumm aduooum :oxou a mmmHo .oococcouuu you» c« nuoH>o£on on ohmH chHuom coxou wCHudc mmaHo .wEHu xcsun IcsHo> ucmpsum mtmt ucoawououcwoquHom .uHsnnnsa mummoMUCH ucooHuwconInmN IImuoH>o£mn owaocnus ovoHHoo wHua quvaun ouao mcwuouqc0lIuHom a :ufluwzz ucoano ocHHowon cu :uduou nmsounu :oHuao NH IHHmou abandons "ouoc uoH>onon IOH mo ocean can no» .N aduucauIIxonm .N :uH) nucHo conuau chad owuoHQEoo nucofi Inwv HHHacoHuoau .chomom Iconmo sums no as Eouu axon :onoc m4 ucafioquMGHoquHom a :xoum mow .H cOHuuoaouoII>cct .H o>Huoouom>n N Ivasum ammo ocHuouHcoaIuHom .sdauouwst ostcnooe HouucoquHum HmVoanHuo> Hmvoawwcnoms wow vouuoaom uuouum ucovcwmoo voonndm conwo HouUCOUIuHom Hm.uosu=¢ INI Tyson: .H OHQMB 25 investigations will be discussed in greater detail in the next chapter along with other studies and reviews related to self—control and academic performance that did not meet these specific requirements. A review of Table 1 permits three preliminary observations. First, various self-control techniques have been used successfully to improve a variety of academic behaviors. Second, these techniques have been found effective in different settings and with different subjects. The diversity in subjects, settings and behaviors is important to any generalizations which are to be permitted about self-control techniques and academic performance. A closer look, however, at the academic settings used for self-control investigations indicates that 15 of the 31, nearly half, have been conducted with college students. Within this subset, nearly all of the investigations have used subjects who were volunteers from an introductory psychology course, or volunteers for a special program in study habits improvement, or something similar. Our ability to generalize to college students' academic performance is correspondingly limited. Similarly, the remaining 16 studies included elementary, junior high school, psychiatric institutions and special after school pro- grams as settings. DeSpite the diversity, it appears 26 that too few studies have been conducted as yet to permit any generalization about academic performance across settings. Many questions remain about the effectiveness of various self-control techniques on academic performance in various settings. The climate is improved for self- control investigations and future studies will address many of these questions. The present investigation was an attempt to address one of these questions in a com- parison of two self-control techniques--self-monitoring and a combination of self-monitoring, self-reinforcement, and self-punishment. Subjects were drawn from a required college calculus and chemistry class to provide a dif- ferent group of subjects for a self-control investigation. College students are a particularly suitable subject group for self-control as Opposed to external control studies because many important academic behaviors of college students are not accomplished in a supervised setting like that of the elementary or secondary school classroom. It is interesting and important to know the range of applications of self-control principles with elementary and secondary school students, however, other more traditional behavior modification procedures, such as the token economy, or, contingent free time for task completion, are equally suitable for behavior change efforts. These types of external control contingencies Iii ' are not the majI setting self-co this po gation, 0f self 0f the tingeng descrik literat 0f sell as Well teChnic th5t e: 27 are not as easy to apply to college student behaviors, the majority of which occur outside of the classroom setting, nor, perhaps, are they as appropriate. Hence, self-control methods have particular applicability to this population. Before further discussion of the present investi- gation, a detailed explanation of the technical definition of self-control will be undertaken as well as an overview of the theoretical underpinnings. The self-control con- tingency and related experimental designs will be described. Following this discussion, the self-control literature will be reviewed including both the effects of self-control techniques on various outcome measures as well as the effect of other variables on self-control techniques. Attention will then refocus on the studies that examine self-control and academic performance, in preparation for discussion of the current investigation. CHAPTER II DEFINITION OF TERMS AND REVIEW OF THE LITERATURE Definition of Terms Self-control as a technical term has been well- defined by two theorists in the field--Carl Thoreson and Michael Mahoney. They assert that "a person displays self-control when in the relative absence of immediate external constraints, he engages in behavior whose proba- bility has been less than that of alternatively available behaviors" (1974, p. 12). They elaborate this definition with a description of three important features of clas- sical self-control phenomena. (1) they always involve two or more alternative behaviors (2) the consequences of those behaviors are usually conflicting, and (3) the self-regulatory pattern is usually prompted and/or maintained by external factors such as long-term consequences. (1974, p. 14) The emphasis upon the long-term consequence is an important one in this definition for theoretical and practical reasons. The conflicting sequences of the two behaviors usually involve a short-run consequence 28 29 that is immediately pleasant or reinforcing, but ulti- mately aversive, versus a long-run consequence that is pleasant or reinforcing but whose short—run consequence is less desirable than that of the alternative behavior. Two examples could be smoking versus not smoking, or not studying versus studying. Obviously, the reinforcement value of any behavior or response is both person and situation-specific. Relationship to Social Learning Theory It is important to stress this point of long- and short-run consequence conflict in order to respond to the unwarranted accusation that the self-control litera- ture runs counter to reinforcement concepts in behavior theory. Thoreson and Mahoney have explained succinctly that "viewing the self-control sequence at a molar level, the definition simply states that, given ultimate and sufficient incentives, a person will display response patterns whose immediate consequences may appear non- reinforcing" (1974, p. 4). The emphasis here is upon immediate consequences. The effects of delayed consequences necessitate consideration of mediating factors and symbolic processes in the reinforcement model. Mediators are anathema to radical behaviorists whose paradigms have traditionally excluded anything and everything between a response and 30 its consequence. Mahoney, Kazdin, and Lesswing (1974) describe radical or metaphysical behaviorism as "(1) a denial of the existence of the mind, (2) reduction of all experience to glandular secretions and muscular movements, (3) acceptance of almost exclusive environmental deter- minism, and (4) avoidance of conscious processes . . ." (p. 14). They observe that this conceptualization "is a corpse which is intermittently exhumed by behavioral critics who seem to delight in its logical inadequacies" (p. 15). Methodological behaviorism, on the other hand, is "characterized by adherence to some form of opera- tionism, microscopic determinism, logical positivism, and pragmatism" (p. 15). Although radical behaviorism in its extreme form is a rarity, the avoidance of con- scious processes and de—emphasis upon the role of thought and mind is certainly still with us. Social learning theory investigations have had a significant counter- balancing effect. Thoreson and Mahoney (1974) write that . . . while the radical behaviorist may be perplexed by the tenacity of self-controlling responses in the absence of observable environmental influences, the researcher familiar with social learning pro- cesses recognizes the significant mediating role of self-reactions in maintaining certain behaviors. (1974, p. 15) Social learning theory's primary spokesperson, Albert Bandura (1969, 1971, 1974a, 1977) has repeatedly stressed the role of mediating factors in human behavior 31 although he recognizes the importance of environmental effects, immediate and otherwise. Many of the things we do are designed to gain anticipated benefits and to avert future trouble. . . . The widely accepted dictum that man is ruled by response consequences thus fares better for anticipated than for actual consequences. Consider behavior on a fixed-ratio schedule (say, 50:1), in which every fiftieth response is rein- forced. Since 96% of the outcomes are extinctive and only 4% are reinforcing, behavior is main- tained despite its dissuading consequences. As people are exposed to variations in frequency and predictability of reinforcement, they behave on the basis of outcomes they expect to prevail on future occasions. (1974a, pp. 859-860) The individual's self-reactions and self-generated con- sequences (including self-reinforcement and self-punish- ment) constitute a way in which the gap between a behavior and its long-run consequence can be bridged. Given two competing behaviors, in which one behavior (not studying) has short-run positive and long-run negative consequences, while the other (studying) has short-run negative and long-run positive consequences, mediational factors can enable the long-run positive consequence to exert an influential effect in the present. Self—imposed conse- quences can serve as such mediational factors. Thoreson, Mahoney, and other self-control theorists operate within a modified learning paradigm that acknowledges social learning processes such as mediation. In the foreward to Thoreson and Mahoney (1974) Bandura writes: 32 Behavior theory has been undergoing major changes. For years man was viewed mainly as a respondent to environmental influences which automatically shaped and controlled his actions. On closer inspection man proved to be more active and the environment less autonomous. Influences that were believed to affect behavior automatically, in fact have limited impact unless consciously mediated. The manner in which environmental events are cognitively transformed, reduced, and elaborated determines what will be learned and how it will be retained. . . . people play an active role in producing the reinforcement con- tingencies that impinge upon them. Thus, behavior partly creates the environment and the environment influences the behavior in reciprocal fashion. . . . By functioning as an agent as well as an object of influence, man has some power of self- direction. Nothing typifies more clearly the operation of reciprocal processes than the phe- nomenon of self-control. (pp. v-vi) The scientific technology of self-control has emerged from an integration of the cognitive and behavioral factors in the social-learning paradigm, according to Thoreson and Mahoney, and the "key to self-control lies in understanding how internal and external events function together" (1974, p. viii). A basic and extremely important assumption of social learning theory and cognitive behavior modifi- cation is that internal or private events obey the same laws as do observable behaviors. Homme (1965) coined the word "coverant" to refer to covert operants or "operants of the mind." Homme notes that he is not the first to argue this viewpoint on private events. He quotes Skinner, "We need not suppose that events which take place within an organism's skin have special 33 properties for that reason" (cited by Homme, 1965, p. 501). Nonetheless, Homme calls for attention to be paid these private operants and observes that "the tech— nology of self-reinforcement has lagged markedly behind the rest of operant conditioning's technology" (1965, p. 503). He concludes First, everybody is an organism and obeys the same laws of nature including the laws of reinforcement. Second, the occurrence or nonoccurrence of coverants can reliably be discriminated by at least one organism, the one to whom they are private. Third, the organism to whom the events are private can control the presentation of rein- forcement in the form of permitting the occurrence of high probability behaviors. (1965, p. 510) Thus, within social learning theory, reinforcement (self or external) can be applied to covert events, and further- more, such reinforcement can itself be covert. This is a dramatic departure from the prior emphasis upon external behavior, overt consequences, and external control. The Self-Control Contingency An elaboration of relevant aspects of social learning theory and cognitive behavior modification is essential to an understanding of self-control techniques. Description of the basic assumptions, such as the cor- respondence between external and internal events, is important also. Similarly, familiarity with the technical definition of self-control is necessary. This definition has been outlined and it is important to note that the 34 terms "self-regulation and self-management" are often used interchangeably for self-control (Thoreson & Mahoney, 1974, p. 15). It is important, now, to focus on the self-control contingency and to distinguish the controlled response (CR) from the self-controlling response (SCR). Skinner first used these terms in his early dis- cussion of self-control (1953, p. 230) and Mahoney and Thoreson have elaborated them. Controlled responses are those which we seek to modify--to accelerate or decelerate. (Controlled responses are also referred to as target behaviors, outcome measures, or dependent variables.) The CR can be modified by self-controlling responses (independent variables) and there are two major types of SCR's--environmental planning and behavioral programming. Environmental planning entails the use of a method by which "the individual plans and implements changes in relevant situational factors prior to the execution of a target behavior” (Thoreson & Mahoney, 1974, p. 16). This type of SCR has been referred to, also, as stimulus control or antecedent control because it seeks to change the stimuli which are antecedent and presumably initiate or influence the behavior's occurrence (i.e., a place for study is selected and no other activities are permitted there). snc par tor. the 35 Behavioral programming SCR's involve self- administered consequences that are contingent upon and follow rather than precede the behavior to be controlled (CR) (i.e., an individual might tear up a dollar after smoking a cigarette). Thoreson and Mahoney present a partial list of such SCR's. These self-administered consequences serve as mediators between the behavior and the long-run consequence, whose effect is weaker. 1. Self-observation: the recording, charting and70r display of information relevant to a controlled response (e.g., charting one's weight). 2. Positive self-reward: the self-administration or consumption of a freely available reinforcer only after performance of a specific, positive response (e.g., treating one's self to a special event for having lost weight). 3. Negative self-reward: the avoidance of or escape from a freely avoidable aversive stimu- lus only after performance of a specific, positive response (e.g., removing an uncom- plimentary pig poster from one's dining room whenever a diet is adhered to for a full day). 4. Positive selfjpunishment: the removal of a freely available reinforcer after the per- formance of a specific negative response (e.g., tearing up a dollar bill for every 100 calories in excess of one's daily limit). 5. Negative self-punishment: the presentation of a freely avoidable aversive stimulus after the performance of a specific, negative response (e.g., presenting oneself with a noxious odor after each occurrence of snacking). (1974, pp. 21-22) Mahoney and Thoreson note that investigations have typi- cally utilized behavioral programming SCR's (self- administered consequences) rather than environmental 36 planning strategies (antecedent control). The present review concentrates on studies using behavioral pro- gramming. Self-Control Designs It is important to consider one final aspect of self-control investigations--experimental design. (Ref- erence to such designs was included in Table 1 in Chap- ter 1.) According to Thoreson and Mahoney (1974), two designs predominate; namely, empirical case studies and empirical group studies (p. 30). The empirical case study examines one subject at a time. The most popular design used in such single-subject designs is an "operant reversal" or ABAB design. A subject target response (CR) is monitored under a baseline condition (A), an experimental or intervention condition (B) a reversal to baseline (A2), and then a reintervention (Bz) condition. Demonstration that behavior covaries with the presence or absence of intervention is deemed sufficient to indicate treatment effect. Problems with use of this design occur when reversing a behavior is not desirable (i.e., tantrum behavior or weight loss) or not possible (return to baseline for cognitive skills). A second type of empirical case study design involves two or more behaviors and is called a multiple- baseline design. Data are collected on both behaviors simultaneously during baseline. During intervention on of wh ac li tt- rs uQ: 37 only one behavior is initially subjected to intervention and the other remains unmodified. Sequential treatment of each behavior occurs and changes in the behaviors which covary with the introduction of the experimental condition are concluded to be causally or functionally related. Comparisons can be made, then, across behaviors, across situations, and across subjects. When designs like ABAB or multiple—baseline are used across subjects they are, in essence, group designs with a particular type of format--that of a single-subject case study. More typically, empirical group designs compare different experimental conditions assigned to different groups. Treatment effect is demonstrated when the depen- dent variables or outcome measures for each group differ significantly from one another. A comparison of such a design with an empirical case study investigation of an identical experimental issue is helpful to understanding the essential differences. A case study investigation of the effect of a contract to study 6 hours a day, for example, would begin with a baseline record of the number of hours studied (A1). The period of experimental intervention (B1) is marked by the beginning of the con- tract. Then, a baseline period (A2) would again occur in which the contract was not in effect. Finally, an experimental reintervention (BZ) would begin when the contract was reinvoked. The effects upon reported study 38 time over these four phases would be compared. In an empirical group study, subjects might be randomly assigned to two conditions--contract and no contract. The effects upon mean reported study time for the two groups would be compared. A type of hybrid design occurring in the literature is one in which the design of an empirical case study is applied to one entire group and their mean performance on the dependent variable across the ABAB phases is examined. Examples of this design are included, as well, among the studies on self-control and academic performance which were outlined in Table 1. It is important to stress that both case studies and group designs have their strengths and weaknesses and address themselves to particular experimental questions. The bias in social science toward group designs and sta- tistical analysis is inappropriate. As Thoreson and Mahoney have argued, in "individual clinical instances in which neither generalizability nor treatment compari- sons are of prime interest, intensive study of the single subject offers a powerful as well as practical format" (1974, p. 34). There are other apprOpriate applications as well. When the questions under investigation are most apprOpriately addressed through an empirical group study, 39 Thoreson and Mahoney underscore the importance of ade- quate comparison or control groups. One of the more crucial factors in the evaluation of group-based research is whether adequate con- trol groups have been employed. For example, if one were to read a study reporting substantial weight loss on the part of subjects who charted and rewarded their dietetic progress, the inter- pretation of these results would weigh heavily on at least two comparisons: (1) How did the above subjects compare with subjects who simply charted their progress, and (2) how did they compare with subjects who engaged in neither self-charting nor self-reward? The results of a study are more easily interpretable when any and all possible independent variables have been isolated. If Factor A had no effect, but Factor B did (either in a separate group or when combined with A), then we have some indication that B is the active ingredient. (1974, p. 34) Thus, it is important to compare the effect of self- monitoring (also called self-observation and self-chart- ing) with self-reward (or self-punishment), either alone or in combination with self-monitoring, and, to compare both with a control group who did neither. Self-Control Techniques as Independent Variables The theoretical framework and technical defi- nitions of self-control have been outlined as well as the specifics of the self-control contingency. Furthermore, self-control designs have been described. It is important now to examine the self-control literature that pertains directly to self-monitoring, self-reinforcement, and self-punishment. First, self-monitoring contingencies must be assessed with regard to effects, choice of the 40 behavior to be monitored, and accuracy or reliability of reported measurements. Second, the ability of various self-control techniques (SCR's) to modify behaviors (CR's) is of critical importance and must be determined. Most importantly, self-reinforcement techniques must be assessed relative to other self-control methods (self- monitoring and self-punishment) and relative to external reinforcement contingencies. Bandura (1974b) acknowledged the importance of determining "whether self-administered consequences do in fact serve a reinforcing function by influencing response output" (p. 301). Their comparative strengths must be determined also. It is important, as well, to review those studies in which the essential question concerns "how behavioral standards for self-reinforcement are acquired and modified" (Bandura, 1974b, p. 301). Bandura has noted the need to distinguish investigations in which the self-control technique is the independent variable whose effects are observed, from those in which the self-control technique is the dependent variable. From the latter type of investigation we may be able to learn the basis for some of the individual variation we observe. Self-monitoring The act of self-observing has been described by several authors as a reactive measuring procedure (Kazdin, 1974a, 1974b; Lipinski & Nelson, 1974; Lipinski, Black, & 41 Nelson, 1975; McFall, 1970). However, Thoreson and Mahoney (1974), Kazdin (1974a, 1974b) and others argue that self-observation should be regarded as a self- control technique, a treatment or experimental condition in and of itself, rather than merely a reactive measure- ment. Thoreson and Mahoney write: Behavioral self-observation stresses the detailed counting, charting, and evaluation of particular responses, either overt or covert. A growing number of studies has provided evidence that both the systematic counting and charting of certain actions are associated with positive changes in behavior. (1974, p. 134) Kazdin observed that the act of self-recording led to behavior change (1974a) as did Kanfer (1975). Several other studies have noted this effect on a variety of dependent variables such as stuttering (Lanyon & Barocas, 1975); cigarette consumption (Karoly & Doyle, 1975; McFall & Hammen, 1971; Rozensky, 1974); weight loss (Romanczyk, Tracey, Wilson, & Thorpe, 1973; Stollak, 1967; Stuart & Davis, 1974); fingernail-biting (Horan, Hoffman, & Macri, 1974); ecological acts (Hoon, 1976); attendance at swim practice and increased laps during practice (McKenzie & Rushall, 1974); verbal responding (Robertshaw, Kelly, & Hiebert, 1974); auditory hallucinations (Rutner & Bugle, 1969); compulsive behaviors (Jason, 1976); obsessive thoughts (Frederiksen, 1975); phobic behaviors (Leitenberg, Agras, Thomson, & 42 Wright, 1968); tics (Hutzell, Platzek, & Logue, 1974; Thomas, Abrams, & Johnson, 1971). Maletzky (1974) reported on five different case studies in which behavior counting was successful in reducing or eliminating undesirable behaviors. The case studies included a 52-year-old woman with a 30-year history of severe repetitive scratching, a 9-year-old boy with bizarre handwaving, and a 65-year-old woman with a lZ-year history of facial tics. The behavior of these three declined to 0 within a few weeks and remained at 0 at 6- or lZ-month follow-ups. Behavior in two other case studies involving severe nail-biting in a 20-year-old and troublesome out-of-seat behavior in an 11-year-old was reduced significantly. Maletzky notes that patients remarked that the wrist counter which was used served to remind them not to emit the behavior. It appears that self-monitoring of a response targeted for acceleration or deceleration has been associated with a change in the desired direction. Thus, self-monitoring appears to be more than a reactive measurement. Examination of two parameters of self-monitoring have been reported--continuous versus intermittent self- monitoring and timing of the act of monitoring. Frederik- sen, Epstein, and Kosevsky (1975) examined three pro- cedures for self-monitoring and found continuous 43 recording of cigarettes smoked to be superior to either daily or weekly intermittent procedures in achieving smoking reduction. Bellack, Rozensky, and Schwartz (1974) compared the timing of monitoring eating behavior in weight reduction programs. They found differential effects for prebehavior monitoring and postbehavior monitoring. They argue that "self-monitoring is not, in and of itself, a behavior change agent, but rather it provides information which may or may not then be used to modify future behavior. The usefulness will vary with the content, the time, and the nature of the monitoring" (p. 529). The importance of the particular behavior selected for monitoring has been noted by Romancyzk (1974) who found monitoring daily weight and caloric intake to be as effective in achieving weight loss as a full treatment group with a therapist. He reports that the group that simply monitored daily weight evi- denced no losses and concludes that the act of self- monitoring is not the critical variable, but the presence of frequent and immediate feedback contiguous to eating. McFall (1970) and Gottman and McFall (1972) likewise stressed the importance of the particular behavior selected for monitoring. Individuals who mon— itored smoking increased their smoking while those who monitored not smoking increased not smoking (McFall, 44 1970). Orne (1970) is critical of this study for a variety of reasons, including the implicit cues from the experimenter, who was also the instructor, as to the desired behavior. He is also critical of the selection of classroom smoking behavior as the depen- dent variable because of the lack of generalizability. It is one thing to demonstrate that a patient does not eat with the therapist, and quite another to demonstrate changed eating behavior, Orne quips. In another study, monitoring the frequency of class participation increased this behavior, whereas, monitoring nonparticipation increased that behavior (Gottman & McFall, 1972). The authors were able to reverse this effect with the same subjects by changing the behavior monitored. Wade (1974) reported opposite effects. He found that the performance of college stu- dents who monitored correct or incorrect math responses was superior to those who simply received performance feedback, and, to controls. However, a decline in per- formance over time was obtained for subjects who recorded their correct matches. It appears that care must be taken in the selection of the behavior to be monitored and there is some evidence that the behavior selected should be the one targeted for an increase. 45 Reliability of Self-monitoring Yet another interesting facet of the self—monitor- ing literature concerns the accuracy or reliability of the act of self-monitoring. Some of the studies mentioned above employed independent observers to corroborate the self-monitoring data. Others did not. Simkins (1971b) is critical of investigators who do not obtain relia- bility estimates of the self-monitored behavior. He also notes that the use of a wrist counter or similar gadget to facilitate recording does not necessarily assure accurate and reliable responding. The use of a wrist counter does not guarantee that behaviors incompatible with the behavior of pressing the counter may not be prepotent, so that the subject does not press the counter each time the event or behavior occurs. Of course, if the behavior is a "private" one, the discrepancies, if any, would never be discovered. There are two behaviors required of the subject; one is the behavior under investigation, the other is pressing the wrist counter. (p. 86) One response to Simkins is that this criticism can be made of other ways of collecting data--there is the behavior and then there is the questionnaire item; there is the behavior and then there is the ”independent" observer's observation, etc. Returning to the issue of reliability estimates, Nelson and McReynolds (1971) replied to Simkins by acknowledging that reliability data should be collected when possible. However, when the type of behavior under 46 investigation is not amenable to the collection of reliability data, such as covert responding, investiga- tions should continue. A distinction must be made, they argue, between determining the effects of self- recording procedures and controlling for them through the use of a comparison group which only monitors. Simkins (1971a) acknowledged the merit of this idea but persists in calls for reliability estimates "in any study dealing with the self-recording of behavior" (p. 600). It is important to note that this would effectively eliminate much research on covert responding. An excellent discussion of traditional methods of reliability assessment concludes that such methods are themselves unreliable. Five trained undergraduate observers with at least three months experience in recording were observed in a two-phase study. Two assessors with two years of experience as trained observers were used for reliability checks. In the preexperimental phase one assessor strictly adhered to the specified code and one assessor adopted a slightly modified code which resulted in slightly higher fre- quencies of certain behaviors. These differences were never discussed but were apparent when different assessors were used for reliability checks with each observer. This was the observers' only opportunity to note the modified code. 47 In the experimental phase the assessors were also present daily but as during the preexperimental phase they were not ostensibly involved in reliability checks on most days. They had other tasks to perform, such as developing a unique code for a given child. In actuality they were performing reliability checks each day. On some days, observers were not told reliability checks were being performed. On other days, they were told and, furthermore, they were told which assessor was making the check. The findings were quite interesting. When reliability checks were being conducted covertly observers recorded lower frequency of behaviors--75% of the frequency recorded during overt reliability assess- ment. Romancyzk, Kent, Diament, and O'Leary (1973) note that "both the frequencies and reliabilities of behavior recorded were consistently increased by knowledge that reliability was being assessed" (p. 180). Furthermore, the authors report that "observers shifted their obser- vational criteria to match the idiosyncratic criteria employed respectively by the two assessors" (p. 180). One implication of this study is that the use of dif- ferent assessors with different experimental conditions can result in a serious confounding variable. The overall implications are also serious. As the authors indicated "it is generally assumed that the 48 process of reliability assessment is non-reactive; that obtained reliability coefficients reflect the general adequacy of the observational process at times when reli- ability is not being measured" (p. 176). It appears that this is not the case. Traditional methods of obtaining accurate observations of behavior and estimates of the degree of accuracy are themselves seriously flawed. This should not be surprising. Cavior and Marabotto (1976) reported no difference between self-monitoring and external monitoring, indicating that behavioral changes that accompany monitoring are due to the reactive process of being observed. Three important conclusions can be drawn. One is that monitoring affects behavior (and that includes monitoring ”monitoring behaviors" as in reliability checks). Two, traditional methods of assessing reli- ability are flawed. The inability to use such methods when behavior change is undertaken in a self-control paradigm should not preclude the use of the paradigm. Three, the reactive effects of monitoring can be used in a self-monitoring contingency to facilitate an indi- vidual's behavior change efforts. Kazdin's (1974b) remarks on the reliability issue are important. He notes that . . . the importance of SM [self-monitoring] reliability varies with the purpose for which SM is employed. When SM is used as an assessment 49 technique, reliability is exceedingly important, when it is used as a behavior-change technique, the consistency and accuracy of measurement are certainly less crucial and perhaps irrelevant. (p. 231) Herbert and Baer (1972) reported that behavior change was associated with self-monitoring despite inaccuracies. They write that "the inaccuracy of the feedback is not crucial to the effectiveness of the procedure. . . . However, there may be lower limits to the accuracy of self-recording below which the procedure is ineffective" (p. 148). Broden, Hall, and Mitts (1971) have concluded that accuracy in self-monitoring was not essential to obtain effects in a case study investigation of study behavior. They write that a correlation between the subjects' estimate and her actual behavior "was not necessary to achieve or maintain high study rates" (p. 198). They used an independent observer with whom comparisons were made. One possible explanation for these findings is that the act of self-monitoring and concomitant evalu- ation processes can occur, in certain circumstances, without the tangential act of recording and public dis- play. This comes very close to a completely intrinsic self-control process and seems to be a reasonable last step in the progression from external reinforcement to public self-reinforcement to a truly private self- reinforcement event. Inherent in this analysis, then, 50 is the notion that the effect of self-monitoring is due to an implicit self-evaluation process (including self- reinforcement and self-punishment). Stuart (1967) remarks that the recording of weight four times daily for an obese person "serves as a periodic, mildly aversive stimulus associated with over-eating" (p. 359). The self-monitoring effect is not due to some magical result of writing down one's behavior, but to systemati- cally subjecting it to scrutiny, evaluation, and change. Thus, self-monitoring effects are likely to be due to covert evaluative processes and the distinction between self-monitoring conditions and combination con- ditions (self-monitoring, self-reinforcement, and self- punishment) is the degree to which the latter contingen- cies are explicit. Bellack (1976) discusses this issue. A direct comparison between SM and SR is not pos- sible. Subjects instructed to SR must SM first. Subjects instructed to SM can freely administer covert SR. . . . In the present study, the com- parison is between two self—monitoring conditions: one accompanied by intermittent covert SR and the other by fairly continuous overt and covert SR. While it cannot be concluded that SR in isolation is more effective than SM in isolation, the clinically applicable differentiation of the two procedures seems clear. The explicit use of SR critically augments whatever effects SM has. (pp. 73-74) Bellack goes on to say that "the question of whether SM affects behavior directly or simply allows for the occur- rence of SR cannot be fully answered, until and unless there is adequate control over the administration of {Iii he wI Sc as 19 0t '10 an no tr 35 ad Is In I? 51 covert SR" (p. 74). It may not be important to answer this question or to ask it. What may be important is to distinguish when self-monitoring plus covert reinforcement is sufficient and when it must be augmented with overt self-reinforcement. Investigations which seek to compare self- monitoring with the techniques of explicit self-reinforce- ment or self-punishment are not as conclusive as those which demonstrate the effect of self-monitoring alone. Some investigations reported that self-monitoring was as effective as these other techniques (Lanyon & Barocas, 1975; Romanczyk, Tracey, Wilson, & Thorpe, 1973), whereas other investigations have not (Mahoney, 1974; Mahoney, Moura, & Wade, 1973; Thoreson & Mahoney, 1974). Thoreson and Mahoney provide a fair, though tentative, conclusion about relative effectiveness. They write that "as a treatment technique the effects of self-observation are often variable and short-lived. Unless supplemented by additional behavior change influences (e.g., social reinforcement), self-monitoring does not offer promises in the long-term maintenance of effectual behavior" (p. 63). This is a reasonable statement that can be interpreted to mean that most of us have not developed our covert self-reinforcing contingencies as well as our overt ones. Thus, explicit self-reinforcement/ punishment strategies should be expected to enhance self-monitoring in most cases. 52 Self-reinforcement If it is surprising to some that the "mere" act of systematic self-observation is productive of positive change it is equally surprising to others that self- administered reinforcement or punishment is also effec- tive, and as effective as externally administered rein- forcement, in many cases. Thoreson and Mahoney (1974) note that "positive self-reward, in which a person presents himself with a freely available positive rein- forcer that is contingent upon his performing a certain action, has been the subject of a host of laboratory studies" (p. 135). "Generally," they conclude, "self- reward has been shown to be comparable in effectiveness to reinforcement that is externally administered" (p. 136). In a comparison of self-reward and experimenter reward on the number of arithmetic problems completed by college students, Speidel (1974) found them equally effective and concluded that self-reward had reinforcing effects. Glynn (1970) and others reported that self- and external reward were equally effective (Bandura & Perloff, 1967; Bellack, Schwartz, & Rozensky, S. M. Hall, 1973; Johnson, 1969; Johnson, 1970; Johnson & Martin, 1973; Kanfer & Duerfeldt, 1967c; Liebert, Spiegler, & Hall, 1970). Kanfer and Duerfeldt (1967b) found that external and self-reward were equally ineffective in facilitating recall following 53 overlearning of a paired-associate task. They write that . . . under conditions of high certainty by S [the subject] of the appropriateness of his response, the increments associated with feed- back by E [the experimenter] or g himself may play a less important role than the necessary interference with continued responding. (p. 196) Lovitt and Curtiss (1969) found higher academic rates when students determined their contingencies rather than teachers. They further noted that this effect was independent of reinforcement magnitude in this case study investigation. Marston (1967) reported self-reinforcement to be superior to external reinforcement in a visual- motor task. Rozensky and Bellack (1976) found self- reinforcement more effective than external control in a weight loss program. 8. M. Hall (1972) reported the reverse. Balfour (1974) reported that they were equally effective for weight loss. Bolstad and Johnson (1974) noted that self-regulation procedures were more effec- tive than externally managed procedures in the modifi- cation of disruptive classroom behavior. In noncomparative studies, self-reinforcement techniques are typically found effective in obtaining behavior change. Todd (1972) reported the successful use of self-reinforcement to control depressive thoughts and Jackson (1972) found similar success in decreasing reported depression through self-reinforcement of task 54 performance. Miller and Clark (1970) reported that a combination of self-administered candy and verbal self- reinforcement was most effective in facilitating new learning on a discrimination task. Self-reinforcement enhanced the matching performance of elementary school children, also (Montgomery & Parton, 1970). With regard to weight loss, Horan and Johnson (1971) did not report conclusive results with self- reinforced coverants. However, Balfour and Christenson (1975) reported that a behavioral therapy treatment group, which included self-monitoring and self-reinforcement in a self-contracting format, was more effective in obtain- ing weight loss than either a will-power group or a no- treatment control. In a case study, Martin and Sachs (1973) reported that a combination of self-reinforcement and stimulus control resulted in significant weight loss in an obese woman. The effectiveness of self-reinforcement appears to be well-documented, however, the notion of self- reinforcement is not without controversy. Catania (1975) argues that there is little evidence that self-reinforce- ment raises the likelihood of self-reinforced responses despite numerous studies previously reviewed. He argues that self-reinforcement is really only a process of dis- crimination--the individual is aware that a certain performance is adequate. He discusses an example in 55 which a student purportedly used self-reinforcement techniques to facilitate study behavior. Suppose that the student is highly competent in discriminating an effectively completed assign- ment from an ineffectively completed one. The problem at this point is that the student would self-reinforce the completion of the assignment only if its completion were already sufficiently important, that the completion alone would be sufficiently reinforcing anyway. . . . What then is the point of self-reinforcement? (p. 197) It seems that Catania has missed the point. It is pre- cisely because the completion of the assignment is not sufficiently important for it to be "reinforcing anyway" that an additional contingency is added. In the case where the completion is sufficiently reinforcing, we still have an example of self-reinforcement in which the student's self-evaluation process enters in. Catania argues "self-discrimination remains; the rest is mythical" (p. 198). Self-discrimination is necessary, but not suf- ficient; self-evaluation must be present as well. Catania's criticism is flawed at the point at which he fails to see self-evaluation as a kind of self- reinforcement and self-punishment process. Bandura (1976) includes a comprehensive analysis and critique of Catania's position. Goldiamond (1976a, 1976b) objects to the use of the term "self-reinforcement" and contends that operant reinforcement can only occur when there is independence between the individual who emits a response and the 56 individual or entity (such as a computer) that determines that a response requirement has been met. He asserts that "a reinforcement contingency presupposes such independence, absent in self-reinforcement" (1976b, p. 509). Goldiamond does not appear to be arguing that the kinds of behaviors we are discussing do not occur. He simply regards "self-reinforcement" as a misnomer, and, an important one, due to subsequent inferred simi- larity to the operant paradigm. He notes that at any time the subject in a self-reinforcement contingency can drop the contingency and take the reinforcement. This is not the case in operant reinforcement and there- fore, self-reinforcement is not operant reinforcement, although it is somewhat parallel. Thoreson and Wilbur (1976) and Mahoney (1976) offer interesting discussion of Goldiamond's paper but they do not address his basic point. Gewirtz (1971) takes a different approach in his criticism of the concept of self-reinforcement by attri- buting behavioral effects obtained under such conditions to external reinforcement on a high variable ratio schedule. Stuart (1972) makes a similar point asserting that ”the behaviors commonly ascribed to self-control can be functionally analyzed as a special subset of operant responses which are, in fact, under situational control" (p. 130). These conceptualizations have merit 57 in that many authors attribute the genesis of self- reinforcement to external reinforcement training through direct instruction or modeling which is then adopted. In his discussion of internalized control over behavior, Aronfreed (1968) contends that an act is internalized "to the extent its maintenance has become independent of external outcomes--that is, to the extent that its reinforcing consequences are internally mediated with the support of external events such as reward and punish- ment" (p. 18). There is certainly a relationship between self-reinforcement and external reinforcement that needs examination. Self-punishment The self-punishment contingency is used far less often and less conclusive evidence is available. In their investigation of weight gain, Gulanick, Woodburn, and Rim (1975) found self-punishment inferior to self- reward. Mahoney, Moura, and Wade (1973) found the same in their comparison of techniques to reduce weight. They note that “an empirical comparison of self-reward and self-punishment strategies is complicated by such factors as control for frequency of application (which is in turn altered by their relative effectiveness)” (p. 407). Weingartner (1971) failed to obtain differ- ential results in a comparison of self-administered shock and placebo on hallucination behavior of schizophrenics. 58 Conversely, Axelrod, Hall, Weis, and Rohrer (1974) found self-punishment to be effective in reducing smoking behavior in two case studies. Self-punishment was opera- tionally defined in one case as the requirement of shred- ding $1.00 for each day when cigarette-smoking exceeded the maximum specified. In the other case, 25¢ was donated to charity for each cigarette smoked. Similarly, Mahoney (1971) reported the successful use of self- punishment in reducing obsessive thoughts to zero in a 22-year-old veteran. He was instructed to snap a heavy gauge rubber band on his wrist whenever such thoughts occurred. Berecz (1972) reported the successful reduction of smoking through self—administered shock for imagined and actual behavior. Moser (1974) found a combination of self-monitoring and covert self-punishment to be effective in eliminating hallucinatory behavior in a male paranoid schizophrenic. Bellack, Glanz, and Simon (1976) reported self-reinforcement and self-punishment operationalized as positive and negative self-imagery to be equally effective in obtaining weight loss. Horan, Hoffman, and Maori (1974) found self-punishment to be TEES effective than self-reward in their attempts to modify nail-biting. In a comparison of combination of techniques including self-reinforcement and self-punishment, self- punishment, and a control, Tyler and Straughan (1970) 59 obtained no significant differences in weight loss. Harris (1969) also used a combination of techniques and then assigned one-half of the experimental group to an aversive control condition. She found that weight loss occurred in all groups, but those undergoing aversive control lost the most. One reason that there are relatively few self- punishment studies is because they are often mislabeled "negative reinforcement." The Harris (1969) study just mentioned used a punishment contingency but described it as negative reinforcement, and this is common (Bandura & Walters, 1972; Haynes & Kanfer, 1971; Kanfer & Duer- feldt, 1967c; Kanfer & Duerfeldt, 1968; Kanger, Duer- feldt, & LePage, 1969; Kanfer & Marston, 1963b; and Marston & Cohen, 1966 provide a few more examples from studies presently reviewed). It is this author's opinion that the long-standing negative bias in psychology and education toward the punishment contingency and the con- tinued confusion over conceptual and operational defi- nitions of the term have so confused and distorted the issue that a meaningful discussion of research results on punishment, self or otherwise, is difficult (Myers, 1975). From a conceptual and operational standpoint self-reinforcement and self-punishment conditions are. very difficult to separate. For example, the act of 60 rewarding oneself for X performance implies not rewarding oneself for Y performance. The act of not rewarding one- self is clearly an example of positive self-punishment-- the withholding of a freely available reinforcer after the performance of a specific negative response (Thoreson & Mahoney, 1974, p. 22). That is, rewarding yourself for studying for 10 hours implies you will not be reward- ing yourself for less than 10 hours. The latter is an example of a punishment contingency. Thus, upon careful examination of many investigations reportedly utilizing only a self-reinforcement contingency, one finds that both reinforcement and punishment contingencies are integrally present and operational. In conclusion, then, it seems apparent that the positive effects for self- reinforcement studies discussed earlier imply positive effects for self-punishment as well. Certainly the effects have not been untangled. The issues relating to self-monitoring, self- reinforcement, and self-punishment in the self-control literature have been outlined. Self-monitoring and self- reinforcement have been shown to be effective in modifying behavior. Studies which explicitly include self-punish- ment are fewer in number and results are mixed. The effects of self-monitoring appear to be enhanced by self-reinforcement contingencies, however, the difficulty of actually separating the two conditions has been 61 described. A reconceptualization has been offered placing what is usually called simply self-monitoring, and the combination of self-monitoring plus self- reinforcement, on a continuum of covert-overt reinforce- ment. This continuum could also be referred to as covert-overt self-punishment as well, and the inter- relationship of self-reinforcement and self-punishment has been described. As may be apparent, many of the investigations which use self-control contingencies are therapeutic interventions and attempts to control behavior problems such as stuttering, weight gain, nail-biting, and smoking. The self-control techniques in these investigations are independent variables whose effect on behaviors is observed. The self-control field is large and some overview of the general effects of these SCR's--self- monitoring, self-reinforcement, self-punishment is appropriate and necessary for a thorough understanding of associated theoretical and methodological issues. Self-Control Techniques as Dependent Variables An overview of studies that examine the effect of other independent variables on self-control behaviors is also important to a complete understanding of the field. Such independent variables include direct instruction in self-control; prior external reinforcement; 62 model standards; social comparisons; model character- istics; subject characteristics (i.e., age, sex, socio- economic status, degree of learning, personality vari- ables); task, reinforcer, and situational character- istics. Studies that include self-control techniques as dependent variables will now be reviewed. Direct Instruction and Prior External Reinforcement Mischel and Liebert (1967) write that "an impor- tant aspect of self-control is the manner in which indi- viduals regulate the self-administration of rewards and punishments that are available to them without external constraints." They go on to note that individuals "learn to evaluate their own performance and make self-reward contingent on criteria such as the attainment of particu- lar performance levels. Likewise, failure to achieve these criteria may result in self-punitive behavior or self-denial" (p. 673). Two very important ways in which this learning takes place are through direct instruction and prior external reinforcement. Heldebrandt, Feldman, and Ditrichs (1973) indicated that instruction and rules for self-reinforcement affected subsequent self- reinforcement behavior. Marston (1969) reported that differential reinforcement of self-reinforcement affected subsequent self-reinforcement rates. Many 63 authors have noted the effect that prior external rein- forcement plays in establishing self—reinforcement rates (Bass, 1972; Kanfer & Duerfeldt, 1967a; Kanfer & Duer- feldt, 1968; Kanfer, Duerfeldt, & LePage, 1969; Kanfer & Marston, 1963a; Karoly & Kanfer, 1974; Marston, 1964b; Marston & Kanfer, 1963). Modeling Effects Another important way that individuals learn to evaluate behavior and make contingent self-reward is through the effect of models. Bandura and Walters (1972) discuss the acquisition of self-control through modeling. Demonstrations that inhibitions and self-evaluation responses may be learned without the mediation of direct reinforcement are consistent with common- sense thinking. Socialization agents, for example, parents and teachers, frequently make use of exem- plary models and from time to time reward or punish children in front of others in the expec- tation that the positive or negative reinforcement will influence the future behavior of observers. (p. 296) Numerous studies have reported the effect of models on self-reward behavior. Modeled instruction has been shown to be as effective as direct instruction about performance standards for self-reward (Liebert & Allen, 1967; Liebert & Ora, 1968; Liebert, Hanratty, & Hill, 1969). Bandura and Mischel (1965) reported that children readily matched either the high or the low performance standard set for self-reinforcement by their particular model. Marston (1965) reported children imitated a 64 model's self-reinforcement criterion and Herbert, Gel- fand, and Hartmann (1969) report modeling effects for children's self-critical behavior as well. Bandura and Kupers (1964) obtained modeling effects for self—approving and self-critical behaviors. Kanfer and Duerfeldt (1968) found that subjects matched the modeled rate of external reinforcement but self-administered less self-criticism (incorrectly labeled negative reinforcement). Helde- brandt, Feldman, and Ditrichs (1973) reported that self- reinforcement rates were a function of both instruction in a rule for reinforcement and a model's behavior. Acceptance of the rule was enhanced when the most recent model's behavior and the rule were concordant. A group of children with a demonstrated preference for immediate but less valuable rewards were shown to reverse this behavior when they observed either a live or a symbolic verbal model exhibiting the opposite behavior (Bandura & Mischel, 1965). Prior direct instructions by the experimenter about self-reward criteria were either adhered to or violated by 9- and lO-year-old children who emulated models obeying or disobeying the experimenter (Hill & Liebert, 1968). Powerful modeling effects are evident when behavior is reversed and direct instruction is thwarted. Both self—reinforcement and the reinforcement administered to another were affected by a model's 65 standard as college students adopted that standard (Marston & Smith, 1968). Mischel and Liebert (1966) also found that the most stringent standards of self- reward were obtained with 4th and 5th grade children when models imposed a stringent standard on themselves and on the child. In cases of discrepant standards, the child emulated the standards actually imposed on him/her, and furthermore, imposed this standard on others. The authors' comments are interesting. The study demonstrated that consistency in the standards which an individual is trained to use for himself and those he observes used by social agents facilitates the adoption and transmission of these standards and pointed to some variables that can determine the performance levels which the person adopts for his own self-reward and for reinforcing others. . . . There is abundant clinical evidence that for troubled individuals the inappropriate regulation of self-administered rewards and punishment often is a central problem. A host of deviant behavior patterns, such as psy- chopathy, masochism, depression, sadism, etc., may be construed as reflecting the inappropriate regulation of self-administered rewards and pun- ishments and the imposition of excessively harsh or generous standards on other people. The iso- lation of antecedents of self-control therefore seems to have particular importance. (p. 53) Discrepant standards were also examined by Masters (1968, 1969) in a social comparison framework. He found that preschool age children were affected by inequity manipulations. The children played a game in which the subject received fewer, more, or equal rein- forcement relative to a younger peer. Subsequently, the self-reinforcement rate was highest for children who 66 previously received less reinforcement than their younger peer. The rate was most stringent when rewards to both had previously been equal. Certain characteristics of models have been found to enhance or diminish modeling effects. Bandura, Grusec, and Menlove (1967) reported on the effects of three variables--mode1 nurturance, socialreinforcement of a model's high standard setting behavior, and presence or absence of a peer model who adopts a low standard of self-reward. They found that vicarious positive rein- forcement enhanced emulation of severe standards by children aged 7 to 11 years old, whereas exposure to a more nurturant model or conflicting peer standards reduced emulation. Masters (1971) reported an inter- esting finding that 4- and 5-year-old children admin- istered more self-reinforcement when the experimenter was female. This would certainly have implications for all studies using self-reinforcement as a dependent variable and needs further investigation. Adults were more influential than peer models in a study by Bandura and Kupers (1964). Models perceived by 2nd and 3rd grade children as more powerful exerted more influence on self-reward criteria (Mischel & Liebert, 1967). Bandura and Whalen (1966) reported an interesting interaction between model competence and self-reinforce- ment criteria. Children emulated inferior models 67 displaying low performance criteria but they did not emulate superior models displaying high criteria. They established a criterion nearer to their own ability level instead. Colle and Bee (1968), however, did not obtain differential effects on self-reinforcement for high—low model competence. Subject Characteristics In addition to the effects of modeled behavior and model characteristics, self-control behaviors are affected by characteristics of the subjects themselves. In a developmental study comparing 543 children in grades 2nd through 8th, Kanfer (1966) reported that inappropriate self-reward decreased with age. Masters (1973) made the same observation with 4- and 7-year-olds. Colle and Bee (1968) reported significant effects of socio-economic status (SES) with boys aged 8 to 13. The higher the SES, the higher the standard setting for self- reward. Haynes and Kanfer (1971) also reported that third and fourth grade boys, whose standing in their classroom was higher, evidenced lower self-reinforcement rates and more self-critical behavior. Conversely, Reschly (1973) indicated that neither sex nor ability level was correlated with self-reinforcement rates. However, self-esteem estimates were positively correlated with self-reinforcement rates for seventh graders (Reschly & Mittman, 1973). 68 Masters (1972) found that subjects who exper- ienced success on a task generally increased self- reinforcement, however, self-reinforcement following failure only increased if the self-reinforcement was not contingent on the task, or was contingent on a dissimilar task. In comparing the self-reinforced performance of college males, Kanfer, Bradley, and Marston (1962) reported that students receiving discrimination training for 50 trials administered more correct self—reinforcement than those who had 25 trials of training. Other studies also reported correct self-reinforcement is related to degree of learning (Kanfer & Marston, 1963b; Marston, 1964b). Marston (1964a) noted that individuals categorized as task-oriented rather than self-oriented or interaction- oriented demonstrate increased frequency of corrected self-reinforcement over trials, presumably due to increasing confidence with the task. Marston and Cohen (1966) observed that self-critical behavior (incorrectly labeled negative reinforcement) was n25 related to intro- punitive scores. In fact, moderate intropunitive scores were associated with more self-critical behavior than either high or low scores. They conclude that ". . . the tendency to be self-critical relates to a dimension that may be described better as ego strength than as intro- punitiveness" (p. 243). They also report that individuals 69 who experienced a frustrating insoluble task following acquisition training on a verbal learning task were more self-critical during the self-control phase of the verbal learning task. Bellack (1972) reported no significant difference, with regard to total self-reinforcement, between college students categorized as internals or externals on a locus of control measure. However, internals gave themselves more incorrect self-reinforcement and made fewer correct responses on a verbal discrimination task despite train- ing to criterion identical to that of externals. H. V. Hall (1973) reported no differences between internals and externals on rate of self-reinforcement. Characteristics of the Task, ReinfOrcer, or Situation Characteristics of the task and the particular reinforcer also have demonstrated effects on self-control behavior. Masters and Christy (1974) reported that long tasks resulted in more self-reward, regardless of diffi- culty level. Kanfer and Marston (1963a) found that appropriate self-reinforcement was lower in frequency relative to inappropriate self-reinforcement when the test list differed from the acquisition list in a verbal learning task. Reschly (1973) and Reschly and Mittman (1973) reported that the greater the task ambiguity, the lower the self-reinforcement rate for seventh 70 graders. Liebert and Allen (1967) indicated that the more explicit the rule for self-reward, the less likely a child was to deviate from the rule when performing alone. They note that it is important to distinguish between acquisition of self-reward criteria and per- formance. Sufficient attention during training is essential for acquisition; other social variables like rule structure and reward magnitude affect performance. High incentive, operationalized as the most desirable category of reinforcement, was associated with more adherence to standards for self-reward (Liebert & Ora, 1968). The authors write that . . . this finding strengthens the view that the incidence of behaviors commonly taken as indexes of self-control may be partially determined by characteristics of the performance situation, rather than solely by prior training or the action of the hypothesized internal moral agents. (p. 543) The importance of the performance situation is also observed in two studies examining self-monitoring accuracy. Epstein, Webster, and Miller (1975) reported that self-monitoring accuracy was decreased when subjects performed another concurrent operant behavior. They also reported that reinforcement for accurate self— monitoring could reduce error somewhat. Epstein, Miller, and Webster (1976) found errors doubled during self-monitoring of respiration when concurrent rein- forcement for lever-pressing was included. They note 71 that it is important to assess the concurrent environ- mental demands when using self-monitoring as a treatment. Thus, the importance of investigations of self-control procedures outside the laboratory setting is clear. Studies on Self-Control and Academic Performance It is apparent from the foregoing discussion that a variety of variables including characteristics of the task, the reinforcer, the performance situation, the individual, and a model affect self-control behavior in addition to direct instruction and direct reinforcement. The differential ability of these and other variables to influence self-control behavior is strong evidence that such behavior is learned. Bem (1967) concludes from her work with 3- and 4-year-olds that verbal self-control is a develOpmental phenomenon, but one that is based on learning, not merely attaining a particular developmental level. Improved self-control is thus an attainable goal. We will now return our attention to the specific self-control investigations on academic performance which seek to determine the relative strengths and weaknesses of the techniques of self-monitoring, self- reinforcement, and self-punishment. These studies, initially presented in Table l in Chapter I, will be reviewed in depth. Given that self-control can be learned, improved, and taught, the elucidation of the 72 relative strengths and weaknesses of the techniques is an important undertaking. The results of these studies on academic performance are particularly pertinent to the current investigations. Self-monitoring Several of the studies on academic performance included a self-monitoring condition. In a study com- paring study skills advice, stimulus control techniques, and self-monitoring, Richards (1975) found that college students' study behaviors were significantly modified by self-control techniques and that self-monitoring was an effective addition to study skills advice, whereas, stimulus control was not. Significance was reached for an outcome measure of final exam grade, but not for final course grade. The self-monitoring treatment, which is more relevant to this discussion, consisted of moni- toring daily the number of pages read and hours studied. Weekly record sheets were required as well as a cumula- tive graph for the 5-week treatment time. Self-monitoring groups increased their study time significantly during the experimental period, however, no comparison with other groups is possible. The self-monitored data were actually a part of the independent variable for these conditions. Johnson and White (1971) also reported that self- monitoring study behavior increased academic performance (I) 73 when college students were asked to observe, record, and graph their study activities using a point system for different activities. Course grades as of the sixth week were significantly higher for these students relative to students asked to monitor dating behavior and relative to a control group that did not monitor any behavior. Significant differences were not obtained for final course grade. Conclusions must be considered tentative with regard to this finding, however, because 80% of the class received a final grade of A and variability was much reduced compared with sixth-week grades. Gottman and McFall (1972) obtained increases in the behavior selected for self-monitoring in their investigation of self-monitoring effects on the class participation of 17 high school students identified as potential dropouts. Subjects monitored either partici- pating or not participating in discussion, after having been instructed to participate as usual. The behavior monitored increased and a cross-over effect was observed when the subjects' task was reversed. The successful use of self-monitoring to reduce arithmetic errors was reported by Fink and Carnine (1975). They used an ABAB design with a group of 10 first graders alternating baseline with feedback about the correct score on arithmetic problems followed by a return to baseline. The second interaction included feedback 74 plus graphing scores. In only the latter intervention did scores differ significantly from baseline. These results must be considered tentative, however, due to the absence of a replication of the final intervention. Broden, Hall, and Mitts (1971) reported success- ful results with self-monitoring in two different case studies. Lisa, an eighth grade student, monitored in class study time in history in a modified ABAB reversal design. Teacher praise was added to the third inter- vention after self—monitoring successfully enhanced study rate to approximately 80% over baseline of 27% in two reversals. Social reinforcement in the form of teacher praise brought study rate to 88%. With praise only, there was a slight decline to 77%. In addition to increased study time, Lisa's grade increased from a D- to a C and postchecks subsequent to the con- clusion of the experiment indicated increased study rate was being maintained. This case study, as previously mentioned, also offered support for the hypothesis that the actual act of recording is not necessarily the active component of self-monitoring. Lisa often forgot to record her behavior in the latter intervention stages. Furthermore, although means of overall subject-observer records were similar, estimates of the percentage of study on a day-to-day basis showed little correlation. 75 Unlike Lisa's case, Stu, an eighth grade math student, monitored "talking out" behavior that was targeted for a decrease (Broden, Hall, & Mitts, 1971). Some reduction was accomplished but it was not as dramatic, nor as well maintained. No attempt was made to bring the behavior under social control by pairing with teacher praise and, as has been noted previously, the behavior selected for monitoring might best be the behavior to be increased. It is important to note that in Lisa's case, occasional praise from her counselor was confounded with the self-monitoring conditions. Furthermore, unlike Lisa, Stu did not himself seek assistance with his problem behavior, but was referred. These additional differences may help account for the conflicting results obtained. Several other authors reported case study results obtained with self-monitoring. Maletzky (1974) reported that out-of-seat behavior was reduced by 80% for an 11- year-old girl with diagnosed behavioral problems. Fox (1966) reported successfully increasing the time a graduate student spent studying French through a combi- nation of self-monitoring and self-contracting for increases. The results are inconclusive with regard to self-monitoring because this technique was used in combination with other techniques. Similarly, Goldiamond (1965) used a combination of self-monitoring and weekly 76 discussions about class notes, exams, and study behavior with a college student volunteer. Overall improvement in study time and grades was reported, however, the specific effects of self-monitoring cannot be assessed. Case studies are often difficult to interpret conclusively because treatments often involve a combi- nation of techniques. Furthermore, ABAB designs that permit modified treatments at each intervention and do not utilize adequate baseline periods are also incon- clusive. Although the overall behavior improvement may occur, the functional relationship to experimental manipulation is difficult to establish. Such problems are not limited to case studies, however. They extend as well to studies using a modified ABAB design with a group and to group comparison designs in which inadequate control groups are present for comparison purposes. Mahoney, Moore, Wade, and Moura (1973) reported an attempt at a more refined examination of the self- monitoring effect. They compared the effects of two types of self-monitoring on academic performance. They found continuous self-monitoring to be more effective than intermittent self-monitoring with reference to the length of time college students spent in review efforts for the Graduate Record Examination. They also found that self-monitoring affected accuracy on sample problems differentially, enhancing accuracy on mathematical 77 problems dramatically and demonstrating no effect on verbal accuracy. They discuss this interaction and suggest that "quantitative performance is much more sensi- tive to such modifiable factors as care given to calcu- lations and so forth. Verbal tasks in the present study dealt mainly with the S's [subject‘s] vocabulary and con- ceptual abilities" (p. 68). Unsuccessful applications of self-monitoring were reported in two studies. Bristol and Sloane (1974) com- pared the test performance of college students assigned to three groups--a control group, a second group that recorded and graphed study time, and a third group that participated in contingency contracting of study time in an ABAB design. All three groups earned $5.00 for graphing and recording study time for an initial 18-day baseline period. Subsequently, the controls were told that they had completed all the requirements for the experiment. Group II received instruction to continue recording and graphing. They received praise for turn- ing in graphs and earned an additional $12.00 at the end of the quarter for having submitted all graphs. Group III had weekly meetings with the experimenter and earned up to $5.00 a week for submitting daily task cards with the weeks' required study time. They also received the $12.00 for continued monitoring and graphing activities. 78 No significant differences were found on test performance among the three groups. However, within the contracting group, the reversal procedure established that contracting significantly increased the reported study time of students. Furthermore, differential effects were obtained on test performance for Group III in which below-average students' test performance improved significantly. The authors conclude that "self- recording of study time in the absence of differential consequences did not improve test performance" (p. 271). It is important to note, however, that the three treat- ment groups received different amounts of monetary rein- forcement. Conclusions about the relative effectiveness of self-monitoring in this instance must be tentative due to the presence of monetary reinforcement as a con- founding variable. Miller and Gimpl (1971) also reported unsuccess- ful results with self-monitoring in their comparison of self-monitoring, self-instruction, and external rein- forcement. They assigned 23 college student volunteers to one of three conditions. All three conditions recorded their study time during the first week. During the second week all three groups followed a procedure whereby they would give themselves instructions three times a day to increase study time a specified amount. They continued recording as well. During the third week 79 the three groups received differential assignments. Group 1 returned to recording study time, Group 2 con- tinued the self-instructions, and Group 3 received external reinforcement in the form of points earned toward their psychology course grade. Not surprisingly, Group 3 study time increased the most. Group 1 decreased somewhat and Group 2 increased somewhat. Thus, self- monitoring was least effective in this study, compared with self-instruction and with external reinforcement. Self-monitoring_and Self- reinforcement Several studies reported the use of self-monitoring in combination with self-reinforcement. Ballard and Glynn (1975) used designs with multiple baseline across three writing behaviors for third graders. Assessments were made of the number of sentences written, the number of action words, and the number of describing words as well as time on task and an external evaluation of the writing. Self—monitoring procedures during baseline were supple- mented with self-reinforcement in which the child earned points toward reinforcers for the specific writing behavior targeted for increase. Rates of responding, time on task, and external evaluations were all increased when self-reinforcement was added to self-monitoring. Turkewitz, O'Leary, and Ironsmith (1975) reported the successful use of a combination of self-monitoring 80 and self-reinforcement in a lZ-phase token program. Eight 7- to ll-year-old students were enrolled in a transitional adjustment class and were also behind one year or more in reading. The 12 conditions were baseline, goals, self-evaluation, tokens, matching 100%, matching 50%, matching 33 1/3%, no matching, fading backups 50%, fading backups 33 l/3%, fading backups 12 l/2%, no back- ups. During baseline, the childrens' progress in their readers was noted by one of the two teachers every 15 minutes. In the next phase goals were written on cards for each 15-minute interval and the cards were taped to the students' desks. In self-evaluation the children rated themselves on their goal cards every 15 minutes and could earn up to 5 points for their academic work and 5 points for their general behavior. Teachers also rated the children, but the ratings were not compared in this phase. In the token period, the children did not rate their behavior. The teachers used the same scale as was employed in self-evaluation and wrote the points on the children's cards during three of the four 15- minute intervals. The other lS-minute intervals con- stituted a control period. During matching the children resumed self-evaluation and received bonus points for exact matches and loss of all points for discrepancies of more than 1 point. Matching was faded in four phases 81 by progressing from 100% to 50% to 33 1/3% to 0% of the class being selected to match the teacher. In no matching the children received whatever points they assigned themselves. Backup reinforcers were faded similarly with 50%, 33 1/3%, 12 1/2%, and 0% of the students being selected to exchange their points for backup reinforcers. The authors were successful in reducing disruptive behavior with the introduction of the token phases and this reduction extended to the 15-minute control period. Low levels of disruptive behavior and a high degree of accuracy in rating were maintained, although some increase in disruptive behavior occurred during the no matching and no backup periods. The shaping and fading periods appear to have been effective in achieving a maintenance of the effects obtained during the token period. This study was unusually diligent in efforts to assure gen- eralization and maintenance of behavior change. From a practical standpoint, replications of effects through return to baseline and reintervention were obviously undesirable. From a research perspective, caution must be used in interpreting the results and inferring functional relationships between behavior change and the self-monitoring and self-reinforcement techniques. The particular goals, in this study, of maintenance and generalization of behavior change, are in direct conflict with Obtain. strate func The replicated . O'Leary (19 monitoring during an a lo‘Year-old gram and ma findings we accurate se reductiOn i even after inaEEd. Th the study a different 8 the finding c1y Use of Self 82 with obtaining the fine discrimination needed to demon- strate functional relationships in an ABAB design. The results reported by Turkewitz et a1. (1975) replicated an earlier study by Drabman, Spitalnik, and O'Leary (1973) who reported the effective use of self- monitoring and self-reinforcement in a token program during an after-school remedial class for eight 9- to lO-year-old boys. The major components of the token pro- gram and matching and fading phases were the same. The findings were also similar--relatively honest and accurate self-evaluation was obtained as well as a reduction in disruptive behavior which was maintained even after all checking of self-evaluations was elim- inated. The same concerns about replication within the study are present, however, similar results by two different studies offers some additional support for the findings. Glynn and Thomas (1974) reported the successful use of self-monitoring and self-reinforcement techniques with a third grade classroom in which students had not had prior specific training with external reinforcement. Using a modified ABAB design, a period of self-monitoring and self-determined, self-administered reinforcement for on-task behavior followed an initial baseline. Inter- mittent tape-recorded signals varying from 1 to 5 minutes signaled when students were to record whether they were on task. Each time in an acti were given duri The 4- the second inte appeared to be to be confused task. A chart the desired beh and indicated t teacher, Stay 1 0f the Chart sa books, read ins of self-mOnitOr resulted in a S authors note th with a Change 1 appear“ to o bs fACtOr I Result 83 on task. Each check mark was worth 1 minute of free time in an activity room and between 10 and 12 signals were given during a lesson. The 4- and 5-minute intervals were removed during the second intervention because the reinforcement ratio appeared to be too large. Furthermore, students appeared to be confused about what exactly constituted being on task. A chart with two colors was added to indicate the desired behavior. One-half of the chart was in red and indicated that the student was to "look at the teacher, stay in your seat, be quiet," the green half of the chart said "work at your place, write in your books, read instructions on the blackboard." The use of self-monitoring and self-reinforcement with cueing resulted in a significant increase in time on task. The authors note that the addition of cueing was confounded with a change in reinforcement ratio, however, it appeared to observers that the chart was the influential factor. Results must be interpreted cautiously for this reason and because no replication of the finding was undertaken. The effects of information feedback, self-rein- forcement and self-monitoring of study behavior was investigated by Richards, McReynolds, Holt, and Sexton (1976). Volunteer college students concerned about Study habits were assigned to one of eight conditions--a no treatme six study groups for nation fee for self-e instructic and instri consequem alized as day and re Student. recording lected We maniPulat; ing, A m. mOnitorin‘ advice Wh R mental 9r UhinfOrme of Study rEpOrted - found the their Stu time to t fig . .erlmen 84 no treatment control, a study skills advice group, and six study skills/self-monitoring groups. The latter six groups formed a 2 x 3 matrix in which high and low infor- mation feedback was crossed with three types of instruction for self-administered consequences-—no instructions, instructions to self-administer covert consequences, and instructions to self-administer covert and overt consequences. High information feedback was operation- alized as monitoring the exact number of pages read each day and recording this on cumulative graphs kept by the student. Low information feedback consisted of self- recording 0, 1-50, or over 50 pages read on graphs col- lected weekly. Exam grades showed that neither of these manipulations enhanced the effectiveness of self-monitor- ing. A main effect of treatment was obtained with self- monitoring/study skills groups superior to study skills advice which was superior to controls. Richards et a1. (1976) also collapsed the experi- mental groups into two categories--informed and uninformed. Uninformed students were those whose baseline estimates of study time were most discrepant from the times reported during the first week of the experiment. They found these students who had previously over-estimated their study time substantially increased their actual time to the level of informed students over the experimental period. The authors conclude that 85 "behaviorally uninformed students derived more benefit from self-monitoring than informed students“ (p. 318). This finding seems reasonable, although no other study has reported such a comparison. Also reasonable it seems are the findings on high-low feedback and the addition of self-administered consequences. Several authors have noted that reliability of self-recording is not essential to obtaining behavior change. Hence, fine-tuned manipulations of recording could be superfluous, if indeed they were accurate at all. Secondly, Richards et a1. (1976) note that "self-admin— istered consequences (e.g., covert self-reinforcement or self-punishment) may be implicitly involved in self- monitoring a valued behavior" (p. 36). To add covert consequences to the covert consequences of self-monitor- ing may not enhance the effect. To add overt consequences such as a tangible reinforcer has enhanced the effect of self-monitoring in some instances. However, there was no attempt to determine whether students actually followed instructions and conformed to this condition. The over- all findings reported do give support to the effectiveness of self-monitoring, however, in this instance, self- reinforcement did not increase the effect. Van Zoost and Jackson (1974) examined the effects of a combination of self-monitoring and self-reinforcement procedures added to a study skills course. The opportunity 86 to earn back $7.00 of a $10.00 deposit through completion of program activities was available to all students. Forty-three college students were assigned to one of three conditions in addition to the study skills class-- self-monitoring and self-reinforcement of specified study activities, self-monitoring and self-reinforcement of specified library activities, and, self-reinforcement of self-monitoring. The self-reinforcement consisted of the opportunity to earn up to 75¢ for every two weeks for eight weeks for the point activities on the cards. This enabled experimental subjects to earn up to $10.00 of their deposit back. Controls were paid $3.00 for completing final tests and questionnaires in addition to the $7.00 for class participation. The authors reported no significant differences among groups on the Study Habits score of the SSHA and add that ”monitoring study behaviors did not improve reported study habits relative to recording other behaviors or not recording at all" (p. 217). However, they did report significant gains for all three groups on the Study Habits score. They conclude that "the effectiveness of self-administered reinforcement in altering and sustaining complex behaviors overtime has again been demonstrated” (p. 217). An adequate control group was not present to permit such a conclusion about self-reinforcement which is confounded with the study 87 skills course. One of the conceptualizations of self- reinforcement utilized in this study--self-reinforcement of an ineffective self-monitoring behavior--and the incomparability among the three groups with regard to how the $10 deposit was potentially regained contributes further to the difficulty of interpreting the results obtained. McReynolds and Church (1973) compared the Work Methods and Study Habits scores on the SSHA for 39 col- lege student volunteers who met the requirement of a low grade point average and a reduced load. Students were assigned to a study skills group; a self-contracting group utilizing self-monitoring, self-reinforcement, and self-contracts; and a no treatment group. Students in a regularly offered Counseling Center course on reading speed, comprehension and vocabulary were also used for comparison purposes. The authors report sig- nificantly higher scores on the two scales for students in the study skills class and the self-contracting group. The results were equivocal for grades, with significant gains in grade points obtained by these two groups, but, no significant differences among the groups in grade point average. Two case studies also reported the effective use of self-monitoring and self-reinforcement. Wasserman, Brown, and Reschly (1974) taught self-management 88 techniques to two hyperactive children assigned to an emotionally disturbed class. Dramatic changes in the target behaviors of Ricky and Andy were obtained when self-monitoring and self-reinforcement procedures were applied. Ricky earned 10 minutes of free time for com- pleting all arithmetic assignments. Andy earned 1 minute of free time for each lO-minute interval in which he recorded no tantrum behavior. If there were no out- bursts in the entire 90-minute recording period he earned 10 minutes of free time. The results were so successful that after six weeks, the children were earning the privi- lege of a half day in the regular classroom, which was a powerful reinforcer for them. After a short period they reentered the regular class permanently on a part- time basis. The authors note that "for a two-year period prior to intervention neither child had been in anything other than a special education classroom environment" (p. 23). Again, from a research standpoint the absence of a replication of the effects necessitates caution in interpreting these results. In another case study utilizing a college student, Whitman and Dussault (1976) reported the application of Self-monitoring and self-reinforcement techniques in a token economy with an ABAB design. Simultaneous modifi- cation of 30 different behaviors was attempted. Point Values were established for achievements (earnings) and 89 for reinforcers (expenditures). The token period required close monitoring of the behaviors and rein- forcers were now contingent. The most important behaviors under modification were academic behaviors including a specified number of hours a week of study time, 100% class participation, and 100% class attendance. The reinforcer was time with a girlfriend. A marked improvement was obtained during the two token periods and a reversal did occur during the second baseline, although the behaviors did not return completely to the level of baseline I. Interestingly, examination of the graphs indicates that reinforcement expenditures remained about the same across the four phases, however, targeted behavior increases were obtained when these freely available reinforcers were made contingent upon the target behaviors. Self-reinforcement On the whole, self-monitoring effects have been reported, although, the addition of reinforcement has increased the effect in some cases. This finding is not unlike that observed in the self-control literature in general. Similarly, self-reinforcement techniques used in self-control literature examining academic performance and.study behavior, have been shown to be equal to external reinforcement, thus paralleling findings 90 reported for the field in general. Glynn, Thomas, and Shee (1973) concluded that self-determination of rein- forcement was effective in maintaining high rates of on-task behavior previously established and maintained with external reinforcement in a second-grade classroom. Similar results were reported by Felixbrod and O'Leary (1973) who demonstrated with second graders that self- determined contingencies and externally determined con- tingencies were equally effective relative to controls in enhancing time on task and number of correct solutions. Felixbrod and O'Leary (1974) replicated their 1973 findings in this study using third graders, however, they did not obtain significant differences among the three experimental conditions with regard to behavior in an extinction phase for either time on task or number of correct solutions. With the removal of contingent reinforcement, all children spent less time in the task setting. The authors noted a tendency, however, though not significant, for students in the externally imposed condition to spend more time on the task during extinction. Speidel (1974) found contingent self-reward and experimenter reward in the form of television viewing to be equally effective and more effective relative to controls in increasing the number of arithmetic problems completed by college students. Lovitt and Curtiss (1969) found, in two experiments, that self-imposed contingencies 91 were more effective in increasing academic response rate in a lZ-year-old student with diagnosed behavior dis- orders. Results of a third experiment indicated that reinforcement magnitude was not a factor. Even when the teacher specified reinforcement rates were set at the same level as that of the student specified rates in Experiment II academic response rate did not increase as it did under student specified periods. Glynn (1970) compared the test performance of 128 9th grade girls under conditions of experimenter- determined, self-determined, random, and no reinforcement. These conditions were preceded and followed by baseline conditions. In the final phase all students were under self-determined reinforcement. Glynn reported that self- determined reinforcement was as effective as externally determined and that both resulted in significantly improved test performance over the entire program rela- tive to the random and no reinforcement groups. Given the repeated demonstrations that self-reinforcement patterns are often similar to the reinforcement patterns experienced under external reinforcement, Glynn makes some interesting observations about the students in the chance reinforced conditions. These students generally performed at a level below that of the nonreinforced control. 92 The inconsistent experience of this class in terms of amount of reinforcement during the Token I phase, seems to have not only precluded performance increments during this phase, but also to have pre- vented subsequent self—determined procedures (which followed in Token phase II) from having any incre- mental effect. This is certainly an indication that the ability to apply self-determined rein- forcement is strongly influenced by the standards of externally determined reinforcement previously experienced. Hence, inconsistency of reinforce- ment can occur not only in terms of interspersing reinforcement with non-reinforcement as conse- quences of a given behavior, but also in terms of unpredictable amounts of reinforcement for a given behavior. These results suggest that parents and teachers, who function as major external rein- forcing agents for children's behavior, should be aware that one consequence of maintaining such inconsistent standards of reinforcement may be impairment of the child's ability to apply self- determined reinforcement procedures effectively. If such an ability is considered as one component of self-control, as Marston and Kanfer (1963) suggest, then inconsistent experiences of amount of reinforcement would have a debilitating effect on the development of an individual's ability to control his own behavior. (p. 131) It is important to recognize that the haphazard and unsys- tematic approach to reinforcement which many well-meaning individuals identify with freedom of the individual can result in diminished ability for self-control and self- determination. Jackson and Van Zoost (1972) assigned 47 college students in Experiment 1 and 35 college students in Experiment 2 to one of four conditions--self-administered reinforcement, external reinforcement, no reinforcement, and no treatment control. The three reinforcement con- ditions varied by the way in which students could earn 93 back a required $10 deposit. Members in external rein- forcement received money for completing the exercises in each session and the group leader evaluated their answers and paid them according to preannounced monetary value. They also earned 50¢ for attending each of six sessions and $1 at the end of the experimental period for completing the questionnaire. Students in the self- administered reinforcement condition received the same amounts of money for attendance and completing the questionnaires. These students, however, evaluated their own performance on the exercises and self- administered monetary reinforcement up to a maximum set by the group leader. The no reinforcement condition received 50¢ for attending each session and $7 at the end of the program after completing the questionnaire. The authors reported that no significant dif- ferences in grade point average were obtained in either experiment. The SSHA Work Methods score did improve significantly for external and self-administered rein- forcement groups in Experiment 1. In Experiment 2, all three experimental conditions improved their Work Methods score. This consistency can be accounted for, in part, by the fact that the "no reinforcement” condition included monetary reinforcement, equal to that of the other conditions in terms of total amount of money received. In fact, the probability for receiving the 94 entire $10 deposit was probably higher under "no rein- forcement” than under external reinforcement, in which the group leader assigned monetary value to performance on weekly exercises. In yet another investigation, Jackson and Van Zoost (1974) compared the study habits scores of those college students assigned to an experimental condition requiring that they self-assess and self-reward their teaching a study skill program in which they were enrolled to a friend or sibling. The teaching groups were superior in subsequent study habit assessment to nonteaching controls, indicating the effectiveness of the combination of self-reinforcement of their teaching behavior. No conclusions can be drawn about the unique effects of self-reinforcement because appropriate con- trol groups were not included. Self-reinforcement and Self- punishment In another study using a combination of self- control techniques, Harris and Trujillo (1975) reported that instruction in self-management which included teach— ing the principles of stimulus control, self-monitoring, reinforcement, punishment and various study skills led to improvement in grade point averages for junior high students compared with a no-treatment control. They also reported that group discussions on study problems 95 were effective. This study is one of only three investi- gations of self-control techniques and academic per— formance that included reference to self-punishment. Beneke and Harris (1972) taught a self-control procedure for modifying study behavior to college stu- dents which consisted of stimulus control procedures, self-reinforcement and punishment, and a study method, the SQ3R, outlined by Robinson (1970). Two instructional formats were used--written instruction and group dis- cussion. The combination of techniques resulted in a significant gain in grade point average for three semes- ters following the program in comparison with two semes- ters prior to the program for both experimental groups relative to controls. The control group, however, con- sisted of 15 subjects who chose to drop out of the pro- gram after the first meeting, thus their comparability is in serious doubt. Beneke and Harris introduced sub- jects to punishment . . . as an option for dealing with behavior incompatible with studying. Suggested types of punishment included fines, denying oneself a pleasant activity, performing good-for-you- but-unpleasant activities (calisthenics, house- work, etc.), and asking a friend or spouse for criticism. (p. 37) Kaufman and O'Leary (1972) also included a self- punishment component in their token program to reduce disruptive behavior and increase reading skills of 16 adolescent pupils in a psychiatric hospital school. 96 The program consisted of six phases--baseline, Token I with teacher reinforcement/punishment for appropriate behavior, baseline, Token II--reinstate Token I, Token III--modify Token I by changing classroom location, and, self-evaluation and self-reinforcement/ punishment. Reductions in disruptive behaviors were obtained with the onset of the Token periods and through- out the final phase. Furthermore, increased amounts of work were accomplished during the less disruptive token periods. The class had been divided in half, initially, and one of two procedures used during the token periods. Students either earned up to 10 tokens during 3 of 4 lS-minute periods or they lost up to 10 tokens which had been given to them at the beginning of a session. The number of tokens earned or lost by children displaying similar behavior patterns was held constant. These same reward-cost procedures were used during the self-control phase as well. There were no differences in the two groups during any of the token phases in level of dis- ruptive behavior or amount of work. The authors observed that "the failure to find differences between the effectiveness of Reward and Cost procedures may be due to the fact that the procedures were equally effective or the procedures were not really different" (p. 307, emphasis mine). This study was one 97 of only three investigations of academic performance that use self-punishment and is the only study to dis- cuss the essentially reciprocal nature of reward-cost procedures. One may argue, however, that since ultimately both reward and cost pupils obtained the same number of tokens for the identical behavior, irrespective of the class procedure, and that the amount always exceeded the number of tokens before class, then both procedures were in reality, Reward procedures. On the other hand, the Reward procedure may have contained certain elements that one would ordinarily associate with cost. Due to the potency of the token programs, there was very little disruptive behavior. Consequently, the teacher usually gave the maximum number of tokens during a rating (e.g., 10). . . . A less- than-maximum rating, therefore, by its sheer infrequency may be looked upon as loss, i.e., a cost. (Kaufman & O'Leary, 1972, p. 307) This integral relationship between reward and cost can also be observed with regard to each behavior. In the reward condition, reward is in effect for appropriate behaviors-~a token is retained or received. Simul- taneously, punishment is in effect for inappropriate behaviors--a token is either removed or 22E received. Withholding reinforcement is a type of punishment. Hence, the reciprocal relationship is aptly portrayed. The Present Investigation Logical and Theoretical Continuity It seems apparent from the foregoing discussion that self-control techniques have been successfully 98 applied to study behavior and academic performance. It is also apparent that such investigations vary on many dimensions. The subjects may be elementary, secondary, college, or adult groups, and, the designs may be group or case study investigations. The dependent variables may be grades, test performance, study time estimates, class participation, among others. The independent variables are also wide-ranging and include not only self-monitoring, self-reinforcement, self-punishment, but also, other self-control techniques such as stimulus control. Furthermore, the methods of operationalizing these independent variables vary considerably. Self- monitoring may include a daily log, charts, and graphs. Self-reinforcement can be food, money, positive self- evaluation, self-administered activity reinforcers such as coffee with friends, trips, breaks, and television. Self-punishment consists of fines, unpleasant activities, or the loss of any of the foregoing reinforcers. The studies also vary in what they investigate and compare-- self-monitoring versus self-reinforcement, self-monitoring versus no-treatment control, and so on. Finally, the results vary. Trends do appear in the data, however. The results appear to correspond, within this subset of investigations, to those described in the foregoing 99 general discussion of self-monitoring, self-reinforcement, and self-punishment effects. As McLaughlin (1976) has observed in a recent review, "a clear majority of the research reviewed has shown that self-control procedures can be implemented in public and remedial classrooms" (p. 654). Specifically, effects have been obtained with self—monitoring and self-reinforcement, and, furthermore, self-reinforcement has been found generally to be equivalent to external reinforcement in effecting behavior change. Studies reportedly utilizing self- punishment have been correspondingly few in number, however, discussion on the interrelatedness of self- reinforcement and self-punishment techniques has been reiterated with regard to studies on academic performance. As noted earlier, the effects obtained with self- monitoring can be explained as effects from a covert self-reinforcement/punishment procedure. From a prac- tical standpoint, self-monitoring contingencies (with presumed covert self-reinforcement and self-punishment) can be distinguished from procedures using overt self- reinforcement and self-punishment techniques. McLaughlin (1976) has observed that no classroom research has yet undertaken to isolate the effects of self-monitoring compared with other self-control techniques (p. 653). Within the larger field of self-control Kazdin (1974b) has stated that the unambiguous demonstration 100 of the efficacy of self-monitoring is more important than theoretical interpretations of the phenomenon (p. 246). Kanfer (1970) has noted that . . . from the standpoint of theory, the investi- gation of self-monitoring in the context of self- regulatory process, promises to add significantly to our understanding of a critical human charac- teristic, the capacity to adjust one's own behavior without continued dependence on environ- mental control. (p. 151) Jeffrey (1974) writes . . . a feasible strategy to isolate the effects of self-monitoring is to use an experimental design that separates the relative contribution of self-monitoring from other aspects of the treatment. A between-subjects group design with a no-treatment control group, a self-monitoring group only and a self-monitoring-plus-some- treatment group could be used to isolate the relative effectiveness of self-monitoring to behavior changes. (p. 183) This is the same comparison Thoreson and Mahoney called for (1974, p. 34) and precisely the comparison which was employed in the present investigation. The Specific Comparisons The present study was designed to provide information on the relative effectiveness of two self- control procedures. The self-controlling responses (SCRs) constituted the independent variable and three levels were present--self—monitoring (SM), self-monitor- ing plus a combination of self-reinforcement and self- punishment (SM+C), and a no-treatment control (C). As noted previously, the comparison is called for from a 101 theoretical standpoint. Thus far the effects of self- monitoring have not been isolated. Different instructions to the subjects comprised the operational definitions of the levels of the inde- pendent variable. The operational definitions utilized in this study were selected carefully. Self-monitoring was operationalized as the daily tallying, recording, and graphing of study time for a designated course. This is a fairly common operational definition for self- monitoring. Self-reinforcement and self-punishment were operationalized as positive and negative self-ratings. The interrelated nature of self-reinforcement and self- punishment is apparent in this particular operational definition. Clearly, these overt consequences appear to come very close to the presumed covert consequences which contribute to the self-monitoring effect. The goal in this operational definition was to make such positive and negative self-evaluative statements overt and explicit in one condition and to thereby permit a comparison of overt and covert techniques as well as self-monitoring and self-reinforcement/punishment process. Tangible self-reinforcers were considered but rejected in an effort to closely approximate hypothe- sized internalized reinforcement-punishment process. The use of self-ratings or combination reinforce- ment/punishment conditions is quite common, though not as common as the use of tangible consequences. In a 102 study comparing self-reward and self-criticism, Kanfer and Duerfeldt (1968) recount that the external use of evaluative responses like "good" or "correct," or symbolic nonvocal equivalents have been shown repeatedly to have reinforcing effects. They suggest that the results of their investigation indicate that "similar stimuli delivered by the subject following his own per— formance may have parallel reinforcing effects" (p. 267). Kanfer, Duerfeldt, and LePage (1969) observe that the facilitative effects of self-judgments have been demon- strated. They assume that the self-judgmental response has both informational and incentive characteristics. Bandura (1974a) has observed that self-administered consequences are often material, however, he expects this to change. Eventually changes in form, as well as source, of reinforcement will appear as the insufficiency of material outcomes is acknowledged. Most people value their self-respect above commodities. They rely extensively on their own self-demands and self-approval as guides for conduct. To ignore the influential role of covert self-rein- forcement in the regulation of behavior is to disavow a uniquely human capacity of man. (p. 863) Through the use of symbolic reinforcers, this study attempted to better examine the self-monitoring effect. In addition, it attempted to address a methodo- logical criticism noted by McLaughlin (1976).. Another methodological issue that plagues most of the research reviewed [on self-control in the classroom] has to do with subject selection bias, 103 that is, subjects who were selected for the study represent an extreme sample of all possible samples. (p. 656) This bias was noted in Chapter I with regard to the studies conducted with college students' academic per- formance. Generally students in such investigations were ones who either volunteered for a course or activity to assist them with study behavior or were selected because of a history of academic difficulty. The remain- der were typically introductory psychology students. The random assignment of such students to con- ditions enables accurate comparisons to be made about the techniques, however, the resulting generalizations are made appropriately only to the populations from which these students were a sample. It is quite plausible that self-control techniques are more suitable for certain groups. One study reported that self-control techniques were more effective for students who were below average on a previous classroom test (Bristol & Sloane, 1974) and another study noted that students who were relatively uninformed about their study habits improved the most when self-control techniques were used (Richards, et al., 1976). Thus, a broader base was sought for the present investigation. Subjects were obtained from two lower division classes--chemistry and calculus--for two independent replications of the experimental manipulations. 104 It is apparent that the present investigation addressed an issue of theoretical importance and further- more undertook to do so with consideration of the issue of generalizability. In the next chapter the specific hypotheses under investigation will be described as well as the experimental design and the specific depen- dent variables. The subjects, materials, and procedures will be described in much more detail. CHAPTER III METHODS Hypotheses Three hypotheses were investigated in this study. Hypothesis 1: Students in the experimental conditions (SM and SM+C) will demonstrate superior test performance relative to controls. That is, students who either self-monitor or self-monitor and administer self- reinforcement and self-punishment in the form of an overt self-rating will have higher test scores than control subjects. Thus, it is expected that the present study will repli- cate other studies that report that the use of such self- control techniques enhances academic performance. Hypothesis 2: Students in the SM+C condition will demonstrate higher test performance than students in the SM conditions. This effect is expected as a result of the observed trend in the literature indicating that the addition of overt self-reinforcement to self-monitoring enhances the self-monitoring effect. 105 106 Hypothesis 3: Students in the SM+C condition will report higher mean study time compared to students in the SM condition. Design The experimental design used to investigate these hypotheses included three conditions--self-monitoring (SM), self-monitoring plus a combination of self-rein- forcement and self-punishment (SM+C), and control (C). The same design was used in two different classes con- stituting two identical independent experiments. sketch of the design in Table 2 includes the number of subjects in each condition. Table 2 Experimental Design and Number of Subjects by Experimental Condition and Class Experiment 1 Experiment 2 Condition Total Chemistry Calculus 1 SM 46 28 74 2 SM+C 52 25 77 3 Control 51 27 78 Total 149 80 229 Dependent Variables The primary dependent variable was the second midterm examination scheduled in each class and a value 107 for this dependent variable was obtained for subjects in all three conditions. (See Appendices A and B for second midterm examinations.) In addition, daily reported study time was obtained from subjects in the two experimental conditions as a result of the presence of the independent variable requiring self-monitoring. Finally, a question- naire was administered to subjects in the two experimental conditions to obtain feedback on the experiment and to assess the subjects' degree of conformity to instructions. Subjects The subjects were 229 students in two classes-- an introductory chemistry class, Science 3A, and a second-quarter calculus class, Math 18, at the Uni- versity of California, San Diego. In chemistry, 151 students participated in the three-week project out of 197 for a participation rate of 77%. (Initial course enrollment was 239; 42 students dropped.) In calculus, 81 students participated out of 102 for a 79% partici- pation rate. (Initial course enrollment was 113; 11 students dropped.) In total, 232 students participated out of 299 for a 77.6% overall participation rate. Three of the 232 students were excluded from the data analysis because they were missing a score on the first midterm examination which served as a covariate. (See Appendices C and D for the first midterm examination.) Additional covariates which were recorded were Scholastic 108 Aptitude Test (SAT) Verbal and Mathematics aptitude scores as well as evaluated high school grade point average. The subjects in chemistry were 56% male and 43% female which includes a slightly higher percentage of males than the overall UCSD average of 52% male, 47% female. The subjects in calculus, however, were 37% male and 63% female. The higher percentage of females is accounted for, in part, by the existence of two calculus sequences at UCSD. One sequence, Math 2, is designed for those who intend to major in one of the physical sciences. The other math sequence, Math 1, is for those who will fill their year-long mathematics requirement with a calculus sequence, but do not intend to major in the physical sciences. Typically, women are underrepresented in the physical sciences accounting for their overrepre- sentation in the Math 1 sequence. Other possibilities exist for meeting the mathematics requirement, depending upon the student's college and major, including a pre- calculus sequence, a beginning statistics sequence, and a computer science sequence. Thus, students in the Math 18 course are in the second quarter of a mid-level difficulty sequence. The subjects in both classes were approximately 70% freshmen and 20% sophomores. The classes are 109 considered to be lower division courses, however, a few upper classmen were enrolled. Materials Considerable time and attention were devoted to the development of the seven forms, which constituted the operationalization of the independent variable, and to the development of two versions of a questionnaire that served as a dependent measure of the degree of conformity to condition. All forms are described below. Cover Letter A standard cover letter was used for all three experimental conditions (see Appendix E). In the cover letter the students were asked to participate in a research project being conducted by the Office of Aca- demic Support and Instructional Services (OASIS) at UCSD. They were told that the project would require them to provide information on a weekly basis for three weeks. They were asked not to discuss their particular project with other class members, who would have different pro- jects, to assure that the information they provided would reflect solely their point of view. Second, students were referred to an instruction sheet which outlined their particular project. The instructions were intended to be self-explanatory. They were told, however, should any questions arise, they were to call the experimenter 110 in her office in OASIS, or see her before or after class. Third, the cover letter explained that the information they provided would be confidential and that code numbers would be used on all materials to prevent identification of the student. They were assured that their professor would not receive information about their participation although he would receive summary information about the class as a whole. They were informed, as well, that their participation would have no effect on their grade. The cover letter was used, also, to inform students that each week they returned their OASIS project a token would be placed in a large jar and they would earn a chance on a $50 prize. In order to be eligible for the drawing, the student must have participated for all three weeks of the project. An addendum to the cover letter was added indi- cating that a 90% return rate for the entire class was essential to the project's success, otherwise, it would have to be repeated. If the return rate was not high enough, the drawing would be held in another class spring quarter when the project was repeated. This addendum was added to encourage participation. This probably facilitated the 77% return rate. Instruction Sheet The instruction sheet constituted page two of the packet and it existed in three forms: 111 Self-monitoring. Instructions to students in this condition specified that they were to log and graph on a daily basis the amount of study time they spent each day on their class, exclud- ing time in class. A sample of a week's log and graph was included (see Appendix F). Self-monitoring + self-rating. In this condition students were instructed to 109 and graph on a daily basis the amount of study time they spent each day on the class, excluding time in class. Additionally, they were to record a self-rating of the amount of time they spent on the class. They were to record a plus (+) whenever they regarded the amount of time to be adequate and a minus (-) when it was inadequate. This self- rating was to be based on their own opinion of adequate study time (see Appendix G). Control. Students in this condition were given a task irrelevant to the present investigation. They were asked to describe in a few sentences nonclassroom experiences they judged to be equal to or greater in importance compared with their course work (see Appendix H). 112 Report Forms Report forms existed in three formats as well. They had in common a small token in the lower right-hand corner that contained the student's code number. In addition to the student's code number, the report form number (1, 2, 3), and, the due date were written at the top of the form. 1. Self-monitoring. This form consisted of a blank graph with the seven days of the week on the abscissa and 250 minutes or 6 hours on the ordinate. A row of seven boxes under the days of the week was used for logging daily study time in minutes. Three c0pies were included in each packet (see Appendix I). 2. Self-monitoring + self-rating. This form was identical to the form above with the addition of seven boxes for self-ratings. Three capies were included in each packet (see Appendix J). 3. Control. This form was blank except for the identifying information and the token. Three copies were included in each packet (see Appendix K). Questionnaire Two forms of the questionnaire were utilized. Control subjects did not receive questionnaires. 113 l. Self-monitoring. This l3-item questionnaire included 9 items designed to assess the degree of conformity to condition; that is, it assessed the degree to which the subjects in the condition reported that they adhered to the instructions. In addition, 3 items asked whether the student thought the project had an effect on study time, whether he/she would continue graphing study time for this class, and whether he/she would use graphing in other classes. The 13th question was open-ended and sought feedback on the project (see Appendix L). 2. Self-monitoring + self-rating. This questionnaire contained 4 additional items intended to assess conformity to condition with regard to self-rating. All other items were identical to the self- monitoring questionnaire previously described (see Appendix M). Thus, each student received a packet of materials con- taining a cover letter, an instruction page, and three report forms. Questionnaires were distributed on the last day of the project to the students in the two experimental conditions. 114 Procedures Procedures will be discussed in two main cate- gories--project development and project implementation. Egoject Development and Pilot Study Development began in fall, 1976, when agreements were reached with the faculty in calculus and chemistry to use their courses between the third and sixth week of the Winter Quarter, 1977, for implementation of this research project. Initial discussions were held in November and the two faculty members involved were pro- vided with copies of the research proposal. Concurrently, further development of the forms was undertaken. Fourteen UCSD students and three junior high students participated in a two-week pilot test of the self-monitoring and self-monitoring plus rating forms to determine whether the instructions were self- explanatory, whether the accompanying graphs were clear, and whether the questionnaire was readily understood. Modification of all forms was made as a result of this pilot review. Following the review, an artist was contracted to design the graph forms in a most visually appealing and functional manner. The same artist's rendering was used to produce the photographic reduction for both types of graphs, insuring identical materials. The 115 only difference between the two graphs was the row of boxes for self-rating. Project Implementation Project implementation began in the second week of winter quarter with the fourth class meeting. For three consecutive class meetings the experimenter circu- lated blank class rosters at the lectures in an attempt to build an accurate class roster and to obtain current addresses and telephone numbers for all students in each class. Sign-up sheets were alphabetized and compared with the official class rosters as of the second week of the quarter when the period for late enrollment ended. Discrepancies were resolved by the experimenter in favor of the "unofficial" class roster. Few students questioned the circulation of class rosters because it is widely known that class lists are not completely accurate. The experimenter did explain the need for the rosters subsequently when a brief announcement was made in the lecture immediately pre- ceding the first exam. The experimenter announced that an OASIS research project would begin following the first exam and that materials would be distributed to students at the exam. It was, thus, necessary to have a complete and accurate class roster. In the announce- ment which took less than two minutes, the experimenter outlined the information in the cover letter and 116 highlighted the $50 drawing. The drawing was humorously received and students appeared to look forward to the project. The rosters were used to randomly assign students to one of the three conditions. Once students were assigned to the conditions, they were assigned code numbers which were recorded on a master log. The code numbers were written on the three report forms in each packet as well as the number of the form (1, 2, or 3) and the day and date due in class. Code numbers were written in the token as well. Large manila envelopes were prepared with the student's name printed in two-inch letters across the envelope. This arrangement was necessary to permit distribution to the particular student the materials for his/her assigned condition. These manila envelopes were arranged in alpha- betical order on the sidewalk outside of the lecture hall for chemistry and large alphabetical letters directed the student to the area where he/she would find a packet. Students picked up their packets as they entered or left the exam. Four additional OASIS staff were present to assure an orderly distribution to the students. The few students who did not pick up a packet received theirs in the mail within 48 hours. About 20% of the class were assigned to take the exam in another location to permit alternate seating in the 117 large lecture hall. Identical distribution procedures were handled by one staff member in this location. Distribution procedures for calculus were the same except that packets were arranged at the front of the class inside the exam. This was a much smaller class and distribution problems were lessened. Only the experimenter and one other staff member were required. Weekly contact was maintained with the faculty throughout the project. In addition, they were provided, in advance, a calendar listing the specific activities that would be undertaken in each class period. Every effort was made to avoid any infringement on class time. Thus, materials were distributed or collected before and after class. The experimenter was available throughout the project at each lecture meeting of both classes to answer any questions. (This required about 7.5 hours of time a week in class.) The only use of class time was the initial announcement and the $50 drawing at the project end. Total class time used was under five minutes. In addition to attending every lecture for three weeks--taking care to arrive early and remain 1ate--the experimenter wrote announcements on the chalkboard at every session. Announcements included the beginning date and due date of the week's project, compliments on excellent participation rate, and during the final week, an announcement about the $50 drawing date. Thus, 118 reminders were present in each lecture through the messages on the chalkboard, the experimenter's presence, and the presence of a large three-quart jar decorated with a $50 insignia and filled with increasing numbers of tokens. The project began on Monday, January 31 for both classes. Materials were distributed to chemistry students at their exam on Friday, January 28, and to calculus stu- dents at their exam on Monday, January 31. The first weekly collection was Monday, February 7. The experi- menter and one other staff member set up a card table, a large sign, and the jar. Armed with scissors, the tokens were clipped into the jar as students turned in their form on the way into the lecture. Students who indicated they forgot their forms were told they could bring them to the next lecture. They were quite apologetic. At the end of the first collection, all the forms were ordered by code number. That evening they were logged. A telephone follow-up was conducted the follow- ing day for all students who did not turn in a form. Students were generally quite appreciative of the phone reminder. Students without phones received notes in the mail within 24 hours. The same collection procedures were followed for the second collection date on Monday, February 14. 119 The telephone follow-up was limited to those students who participated the first week by returning project #1. The third and final collection was made on the day of the class's second exam. In chemistry this was Friday, February 18. Data for this third week were only available for five of the seven days in this class. In calculus the second exam was Wednesday, February 23. Project #3 ended for these students on Sunday, February 20, however, no class was held Monday, February 21 because of a University holiday. They returned project #3 at their second exam. In addition to collecting project #3 at the second exam, questionnaires were distributed to students in the two experimental conditions. The same distribution methods were used as for the distribution of packets. Collection, however, immediately followed the exam. Some students returned their questionnaires at the next lecture period as did some who forgot to bring project #3 to the exam. The $50 drawing was conducted in both classes on Monday, February 28. Students had been told that they must participate for the entire three-week period to be eligible for the drawing so it was essential that the roster be completely accurate prior to the drawing. The students also knew they had to be in class to win. 120 One token was drawn in each class, the student's name was read from the log of code numbers, and a $50 check was given to the student. CHAPTER IV RESULTS Academic Performance Analysis of covariance was used to analyze the scores obtained on the primary dependent variable, mid- term II, in both Experiment 1 (chemistry) and Experiment 2 (calculus). Midterm I was used as the covariate in both analyses. In Experiment 1 the correlation for midterm I and the dependent variable was .660, the multiple R2 was .435, and E (1, 145) = 11.63, p}< .0001. In Experiment 2, midterm I and the dependent variable correlated .618, with a multiple R2 of .381, and E (1, 76) = 46.86, p < .0001. SAT scores and evaluated high school grade point average (gpa) were not included as covariates in the final analysis because they did not add significantly to the variance accounted for by the use of midterm I alone. The decision to exclude SAT scores and gpa as covariates was made in a prior stepwise regression analysis. In Experiment 1 (chemistry), the stepwise regression probability levels were .0001 for the 121 122 contribution of midterm I as a covariate, and .644, .668, and .072 for SAT-Verbal, SAT-Math, and evaluated high school gpa, respectively. Midterm I account for 41.37% for the variance and the three additional covariates added only 1.47%. In Experiment 2 (calculus), the probability levels from the stepwise regression analysis were .0001 for the contribution of midterm I as a covariate, and .226, .518, and .415 for the addition of SAT-Verbal, SAT-Math, and evaluated high school gpa, respectively. Midterm I accounted for 48.04% of the variance and the three additional covariates added only .52%. Two orthogonal contrasts, based on Hypotheses l and 2, were examined in each experiment. The contrast associated with Hypothesis 1 is a comparison of the test performance of students in the two experimental con- ditions combined (SM and SM+C) with that of the controls. The contrast associated with Hypothesis 2 is a comparison of the test performance of students in the two experi- mental conditions with each other. Table 3 includes the estimates of the contrasts and associated standard errors which are discussed further in the results reported for each experiment. It is important to note that the con- trast for Hypothesis 1 considered the control group first and the contrast for Hypothesis 2 considered the SM+C condition first to give meaning to the signs of the con- trasts . 123 Table 3 Least Squares Estimate of the Contrast and Standard Error by Hypothesis and Experiment Experiment 1 Experiment 2 Contrast s.e. n Contrast s.e. n Hypothesis 1 —4.59 2.30 149 1.97 3.52 80 Hypothesis 2 —2.61 2.72 98 3.29 4.11 53 Experiment 1--Chemistry It was predicted in Hypothesis 1 that students in the two experimental conditions (SM and SM+C) would demonstrate superior test performance relative to con— trols, replicating earlier findings that the use of self- control techniques enhanced academic performance. A sig- nificant difference was obtained for midterm II when students' test performance in the experimental conditions was compared with that of the controls, E (1, 145) = 3.85, p < .052, using the adjusted means from the ANCOVA. The least squares estimate of the contrast was -4.59 with a standard error of 2.30. The confidence interval is -4.59 t 4.51. The estimate of the adjusted effect, -4.59, and its sign indicates the size and direction of the dif- ferences between the experimental groups and the controls on midterm II. The test scores for the experimental conditions were 4.59 points higher on the average when 124 compared with controls. The conditional standard deviation for midterm II in chemistry was 13.30, there- fore, the estimated 4.59 point difference between the groups is equivalent to one-third of a standard deviation (.35 to be exact). With regard to Hypothesis 2, it was predicted that students assigned to the SM+C condition, requiring both self-monitoring and self-rating, would evidence higher test performance than students in the SM condition who simply monitored their study time. No significant difference was obtained for midterm II when students in the two experimental conditions were compared, F (l, 145) = .928, p < .337, using adjusted means. The least squares estimate of the contrast was —2.61 with a standard error of 2.72. The confidence interval is -2.61 i 5.32. The estimate of the effect and its sign, -2.61, indicate that the size of the difference between the two experimental conditions is small relative to its standard error and that the students' performance in the SM condition averaged about 2.6 points higher than that of students in the SM+C condition. The 2.6 point difference is equivalent to .20 of a standard deviation difference on midterm II. The direction of the effect is opposite of that predicted although it is not sig- nificant. It is apparent that the addition of overt 125 self-reinforcement and self-punishment in the form of self-ratings did not enhance the effects obtained with self-monitoring alone in this experiment. Thus, it is clear that in Experiment 1, Hypothe- sis 1 was supported, whereas Hypothesis 2 was not. Table 4 includes the adjusted and observed means for the three conditions. Table 4 Adjusted and Observed Midterm II Mean Scores in Chemistry by Condition Control SM+C SM H n X' n X n Adjusted 70.59 51 73.88 52 76.49 46 Observed 71.25 75.42 74.28 Observed s.d. 17.21 17.83 17.85 Experiment 2--Ca1culus Neither Hypothesis 1 nor Hypothesis 2 was sup- ported in Experiment 2 using a calculus class. Once again, it was predicted in Hypothesis 1 that student test performance would be superior in those conditions using self-control techniques (SM and SM+C) when compared with controls. No significant difference was obtained, 3 (1, 76) = .346, p < .558, using adjusted means. The 126 least squares estimate of the contrast was 1.97 with a standard error of 3.52. The confidence interval is 1.97 i 6.90. The size and direction of the difference is evi- dent in the estimate of the contrast, 1.97. On the average, the control groups' test scores were about 2 points higher than that of the experimental condition. Although the difference was not significant, it was in the opposite direction of that predicted. The standard deviation for the adjusted mean on midterm II in calculus was 14.77, therefore, the 1.97 point difference is equal to .13 standard deviations. It was predicted in Hypothesis 2 that students in the SM+C condition, which included both self-monitoring and self-rating, would have higher test performance than students who merely monitored their study time in the SM condition. No significant difference was obtained on midterm II in a comparison of the two experimental con- ditions, E (l, 76) = .642, p,< .425, using adjusted means. The least squares estimate of the contrast was 3.29 with a standard error of 4.11. The confidence interval is 4.11 i 8.05. The size and direction of the difference is evi- dent in the contrast, 3.29. On the average, the students in the SM+C condition scored 3.3 points higher on midterm II than students in the SM condition. Although the 127 difference was not significant, the direction of the difference is as predicted. In Experiment 2, no support for either hypothesis was found. The use of these specific self-control tech- niques did not enhance academic performance. Table 5 includes the adjusted and observed means for the three conditions. Table 5 Adjusted and Observed Midterm II Mean Scores in Calculus by Condition Control SM+C SM i n Y n Y n Adjusted 73.23 27 72.90 25 69.61 28 Observed 75.15 69.96 70.64 Observed s.d. 18.97 19.97 17.09 Reported Study Time It was predicted in Hypothesis 3 that students in the SM+C condition would report a higher mean study time compared with students in the SM condition. Significant differences between experimental conditions were not obtained in either Experiment 1 or Experiment 2 when the average study time for the three-week experimental period was compared. Table 6 contains the daily average reported study time for each of the 21 days. Table 7 contains the average daily reported study time for each of the 1J28 Table 6 Average Daily Reported Study Time in Minutes by Day, Condition, and Course Experiment 1 Experiment 2 Day Chemistrya Calculus SM SM+C Total SM SM+C Total 1 (M) X 19.89 34.52 27.65 22.32 17.20 19.90 _ (31.77) (53.50) (45.02) (25.84) (26.70) (26.12) 2 (T) x 38.58 49.52 44.39 14.82 11.00 13.02 _ (56.88) (63.89) (60.64) (23.79) (19.69) (21.82) 3 (W) x 28.80 43.87 36.80 21.61 28.40 24.81 _ (42.72) (65.49) (56.20) (28.55) (39.34) (33.90) 4 (T) x 34.13 24.90 29.24 21.61 20.80 21.23 _ (47.52) (36.86) (42.23) (29.25) (30.30) (29.47) s (F) x 41.63 30.39 35.66 9.11 11.40 10.19 (55.46) (38.82) (47.44) (18.56) (25.23) (21.77) 6 (S) E 48.04 35.77 41.53 16.96 16.00 16.51 _ (60.35) (71.14) (66.25) (34.84) (25.82) (30.63) 7 (S) x 84.87 73.75 78.97 13.57 16.60 15.00 _ (99.44) (93.47) (95.99) (22.64) (36.59) (29.78) 8 (M) x 58.70 53.46 85.92 20.89 26.00 23.30 _ (66.11) (56.50) (60.94) (33.36) (40.83) (36.79) 9 (T) x 53.70 60.39 57.25 38.04 29.80 34.15 _ (81.68) (90.48) (86.08) (51.07) (43.60) (47.42) 10 (W) x 61.52 57.40 59.34 27.86 28.12 27.98 _ (61.39) (86.25) (75.26) (40.52) (44.23) (41.89) 11 (T) x 59.57 66.25 63.11 28.86 19.20 24.30 _ (78.68) (76.91) (77.42) (45.53) (32.65) (39.90) 12 (F) x 54.56 35.87 44.64 14.64 18.00 16.23 _ (68.26) (46.73) (58.29) (26.87) (34.85) (30.63) 13 (S) x 74.20 60.77 67.07 6.07 14.60 10.09 (107.10) (103.55) (104.90) (14.74) (33.41) (25.43) 14 (S) X 87.61 55.00 70.31 20.54 46.00 32.55 _ (109.42) (76.73) (94.43) (35.86) (58.67) (49.20) 15 (M) x 96.41 82.21 88.88 28.57 38.64 33.32 _ (98.45) (95.88) (96.85) (41.25) (50.03) (45.43) 16 (T) x 105.54 129.52 118.27 37.86 46.04 41.72 _ (106.11) (118.00) (112.64) (69.38) (58.00) (63.79) 17 (W) x 173.04 153.94 162.91 29.29 30.80 30.00 _ (100.42) (106.80) (103.77) (39.29) (36.53) (37.66) 18 (T) x 227.94 224.92 226.34 42.32 20.20 31.89 (133.79) (169.21) (152.84) (57.57) (37.21) (49.84) 19 (F) i 173.80 144.90 58.47 27.32 26.20 26.79 (92.65) (86.02) (89.90) (60.62) (48.80) (54.84) 20 (S) Y - - - 46.25 44.44 45.40 _ - - - (67.15) (71.87) (68.75) 21 (S) x - - - 48.39 52.80 50.47 - - - (58.52) (81.93) (69.86) Note. The n's are as follows: Chemistry, SM 2 46, SM+C = 52, Total = 98; Calculus, SE = 28, SM+C - 25, Total - 53. because midterm II was held at the end of the 19th day. Standard deviations are enclosed in parentheses. aDaily study time was not reported on the 20th and let days in chemistry 129 .mmmmnusmnmm GH owmoHosm mum maowuma>m© cummcmumn .mmmo m Eoum sumo co comma ma x003 pawn» map How mmmnm>m maxwm3 may .mocmn .wnumflamzo ca mmmp umam was nuom map so Umuuommu uoc mm3 mafia mpsum mawmam .3. n 13.2. .3 u o+2m .3 u 2m .mgsoamolwmb n 1.53. .mm H U+zm .mv n Sm .muumHEmno "m3oHHom mm mum m.c one .muoz Am>.m~v Ama.mmv Avm.mmv Amm.nhv Amh.mhv Amm.mhv wo.>m No.5m va.hm nm.oma oa.>va mm.mma M m Amm.mav Amm.mav on.mav Anm.mvv Aho.omv Abm.wwv mo.v~ mm.m~ av.m~ mm.mm mm.mm mm.vm M m Aom.vav hhv.mav Amm.vav Anm.mmv Am~.mmv bAmH.~mv vm.na vm.nH va.na mo.~v mw.av m~.~v M a Hmuoa U+zm 2m Hmuoe U+zm 2m msasoamu mwuumHEan x003 m ucmEHummxm H unwfiflummxm mmusou cam .COHuaocou .xmm3 >3 mmuscflz GA TEHB hpsum cmuuommm mawmo mommm>¢ h OHQMB 130 three weeks. The average daily reported study time is presented in Table 8 by condition and course for the entire experimental period. Table 8 Average Daily Reported Study Time in Minutes over the Three-week Experimental Period by Condition and Course SM SM+C Total Experiment 1-- 80.13 74.60 77.20 chemistrya (41.02) (37.69) (39.18) Experiment 2-- 25.57 26.77 26.14 calculus (17.67) (16.68) (17.06) Note. The n's are as follows: chemistry, SM = 46, SM+C = 52, total = 98; calculus, SM = 28, SM+C = 25, total = 53. aAverage for the experimental period is 19 days instead of 21. Four observations may be made from the tables. As is apparent in Tables 6 and 7, reported study time steadily increased over the three-week period. Table 7 contains the average daily reported study time for each of the three weeks. In chemistry, average daily reported study time increased, over the three-week period, by a factor of approximately 3.7 for the SM condition and 3.5 for the SM+C condition. Students averaged about 42 minutes a day during the first week and increased to about 2.5 hours a day during the third week. In 131 calculus the increase was smaller. Average daily reported study time increased approximately 2.1 times during the three-week period. Students reported daily study time of approximately 17 minutes during the first week and 37 minutes during the third week. Thus, a second observation is that reported study time increased much more in the chemistry class when compared with calculus students' reports. Thirdly, as is readily apparent in Tables 7 and 8 students in chemistry reported ' much higher average study time over the entire experi- mental period. During the first week, chemistry students reported they studied approximately 2.5 times the amount reported in calculus. By the third week chemistry stu- dents reported study time 4.2 times the amount indicated by calculus students. Chemistry students' lowest average study time was higher than the highest average time reported by calculus students. A fourth and final observation is that both daily and weekly averages are associated with high variability. This is particularly evident in Table 6 which includes daily study time. The standard deviations are often larger than the value of the daily average. Considerable differences existed among students in the amount of time they reported studying. In addition to such observations made from visual inspection of Tables 6, 7, and 8, a multivariate 132 repeated measures analysis of variance was performed on the reported study time data. Three questions were for- mally addressed with this analysis. First, was there an experimental condition by time interaction? Second, was there a significant difference between the two experi- mental conditions in the amount of study time reported? Third, was there a significant difference in time reported over the three-week period? In Experiment 1, chemistry, the interaction between experimental condition and time was not signifi- cant, multivariate E (2, 95) = .471, p < .626. Thus, the pattern of study time reported did not vary differ- entially between the two conditions. Furthermore, there was no significant difference between the two experimental conditions with regard to the study time reported over the three-week period, E (1, 96) = .47, p < .492. How- ever, with regard to the third question, there was a significant difference in the amount of study time reported over the three-week period, multivariate E (2, 95) = 96.10, p < .0001. That is, there was a sig- nificant increase in the amount of time chemistry stu- dents reported studying over the three-week period. Associated with the above procedures one can fit a polynomial model to the data points to determine whether the points are best represented by a linear equation or a higher degree equation. The step down E for testing 133 the fit of a higher order equation was 1.12, p < .292, which is not significant. The step-down E for a linear equation was 190.84, E < .0001, indicating that the best fit was a linear model. This implies that the data points in Figure 1, drawn from the totals for the experimental conditions listed in Table 7, for the chemistry class can be represented best by a straight line. In Experiment 2, calculus, the results were similar. The interaction between experimental condition and time was not significant, multivariate E (2, 50) = .31, p < .732. In addition, there was no significant difference between the two experimental conditions with regard to reported study time over the three-week period, E (1, 51) = .07, p < .800. However, a significant dif- ference was obtained in the amount of study time reported over the three-week period, multivariate E (2, 50) = 21.69, p < .0001. The best fitting model was, once again, linear with a step-down §.°f 43.16, E.< .0001; whereas, the test for a higher order equation was not significant, step-down E of .59, E.< .451. Figure 1 portrays the average weekly study time for the combined experimental conditions, SM and SM+C. It is apparent from the results of the repeated measures analysis and Figure 1 that the observations made pre- viously from visual inspection of Tables 6, 7, and 8 have Reported Study Time in Minutes Figure 1. 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 134 '/ I I I I I I / I I I / I I / I I I I o I I I Observed data .I Chemistry 0 Calculus A A Fitted curves Chemistry ————— Calculus l 2 3 Week Reported study time by week and course for combined SM and SM+C conditions. 135 been confirmed. Study time does increase steadily, the increases are greater in the chemistry class, and stu- dents in chemistry begin studying at a higher rate than that attained by calculus students as of the final week of the experimental period. Self-rating of Study Time No hypotheses were made with regard to daily study time rating. As in reported study time measures, daily study time rating is a by—product of the indepen- dent variable used for an experimental condition. As such it is only available for subjects in the SM+C con- dition which required both self—monitoring and a combi- nation self-reinforcement and self-punishment in the form of self-rating. Despite the fact that study time ratings are unavailable for the other two conditions, examination of the responses made for the SM+C condition are interest- ing, in and of themselves. Students recorded either a plus (+) or a minus (-) when they rated their daily study time. They were instructed to rate ggly whether they thought the quantity of time was adequate or inadequate. Positive ratings were coded as 2's and negative ratings were coded as 1's; hence, the midpoint of the range is 1.5. Average scores that are above 1.5 indicate an overall positive assessment of the day or week's study time. Conversely, scores below a 1.5 indicate a negative assessment. Table 9 136 contains the average daily rating of study time for each of the three weeks. Table 10 contains the average rating of study time for each day in the experimental period. Table 9 Weekly Average Rating of Daily Study Time in Condition 2, SM+C, by Course Experiment 1 Experiment 2 Week Chemistrya Calculus X s Y s 1 1.39 (0.25) 1.59 (0.27) 2 1.41 (0.25) 1.40 (0.29) 3 1.63 (0.28) 1.53 (0.24) Note. In Chemistry, 3 = 52; in Calculus, g = 25. aRatings were not obtained for the 20th and let days in chemistry. Two observations can be made from these tables. First, chemistry students tended to rate their study time negatively. On only 5 of 19 days (26%) did stu- dents in chemistry give themselves a positive rating. Four of these 5 days were during the third week when study time was averaging 2.5 hours a day. In calculus, positive ratings were given for 10 of 21 days (48%). These ratings were included in the first week when reported study time was a low average of 17 minutes daily, and, during the third week when reported study 137 Table 10 Average Rating of Daily Study Time in Condition 2, SM+C, by Course Experiment 1 Experiment 2 Day Chemistrya Calculus Y s i s 1 (M) 1.44 0.50 1.84 0.35 2 (T) 1.48 0.51 1.48 0.51 3 (W) 1.42 0.50 1.60 0.50 4 (T) 1.37 0.49 1.64 0.49 5 (F) 1.40 0.50 . 0.51 6 (S) 1.17 0.38 1.56 0.51 7 (S) 1.44 0.50 1.48 0.51 8 (M) 1.48 0.51 1.36 0.49 9 (T) 1.39 0.49 1.40 0.50 10 (W) 1.44 0.50 1.36 0.49 11 (T) 1.48 0.51 1.48 0.51 12 (F) 1.35 0.41 1.36 0.49 13 (S) 1.33 0.47 1.44 0.51 14 (S) 1.42 0.50 1.40 0.50 15 (M) 1.48 0.51 1.40 0.50 16 (T) 1.54 0.50 1.60 0.50 17 (W) 1.75 0.44 1.52 0.51 18 (T) 1.64 0.49 1.40 0.50 19 (F) 1.75 0.44 1.64 0.49 20 (S) - - 1.5 0.51 21 (S) - - 1.60 0.50 Note. In Chemistry, n = 52; in Calculus, g = 25. Values that are underlined are positive ratings. aIn chemistry midterm II was held at the end of the 19th day, hence ratings were not obtained for the 20th and let days. 138 time had increased to 37 minutes daily. A second obser- vation, then, is that the self-ratings differed between the two classes. Calculus students consistently rated their study time more positively despite the fact that their reported time was considerably lower than that reported in chemistry. Questionnaire Responses Questionnaires were distributed to all subjects in both experimental conditions at the time of the second class midterm. The questionnaire had two sections-- conformity to instructions and student opinion. For sub- jects in the SM+C condition, 13 of the 16 objective items were designed to assess conformity to instructions. These 13 questions included 4 on self-rating procedures. The last 3 questionse-items 14, 15, and 16--examined stu- dent opinion about the impact of the project on their study time and future plans about monitoring study time. Subjects in the SM condition had a 9-item section, out of a 12-item questionnaire, designed to assess conformity to instructions. The questions were identical to those asked of the SM+C subjects. (The 4 questions on self-rating were appropriately deleted because this condition did not use overt self-rating.) The remaining 3 questions on student opinion were identi- cal to those asked of the SM+C subjects. 139 In preparing for the analysis of questionnaire data, questions on the lZ-item SM condition questionnaire were renumbered so that identical questions also had identical item numbers for data processing and data display purposes. This is portrayed in Table 11. Items 6, 8, 9, and 11 were not included in the questionnaire received by subjects in the SM condition. (Appendices 8 and 9 include the actual instruments used.) Conformity to Instructions Students could select one of four responses to items 1 to 13--"always," ”usually,” "rarely," "never." In scoring the questionnaire, a 4 was given to the most desirable response and a 1 to the least desirable. For items 1-9, 12, and 13, a 4 corresponded to "always." Conversely, a 4 corresponded to "never" for items 10 and 11. This procedure permitted the computation of a mean score, indicating overall conformity to instructions, which is presented in Table 12. It is apparent that an overall high level of conformity to instructions was obtained in both experimental conditions in both classes. This is an important finding indicating that the dif- ferences in effectiveness of the experimental treatment of each class cannot be attributed to differences in following instructions. Examination of the average response to each item, included in Table 12, indicates, as well, that students 1.4() 666 666 m 666 .mwwonusouam :6 00606060 066 acowumw>0v vumcsdumo .8066 6:6 owvsaocw 66:6 macauwpcoo 666c06660mxm 0:6 «0 HmnEdc «£6 :0 upcmmuw STMw cm>6m 6 How 6660» .mm M U+zw "mm u 2m .msasoawu «Nm u o+zm .mv u 2m 6%66665060 unsoHHON we mum 6.: 0:9 .0602 166.61 166.66 166.66 166.66 166.66 166.66 .6666666 66666 66 66.6 66.6 66.6 66.6 66.6 66.6 m 6666 66666 66 66666666 666 666666666 666 6663 6 .66 166.66 666.66 .66.66 166.6. 166.6. .66.66 .66666 6666 66 6666 66.6 66.~ 66.6 66.6 66.6 66.6 M 66666 66 66666 666 666666 66 66666666 6663 6 .66 166.66 166.66 166.66 166.66 166.6. 166.66 .6666666 6666 66 666666 66.6 66.6 66.6 66.6 66.6 66.6 m 6 66 66666 6666 666 6666 6666666 6 6366666 6 .66 166.66 166.66 166.66 166.66 666.66 .66.66 .666 6666 66666 66666 666 66 66.6 66.6 66.6 66.6 66.6 66.6 m 6 66666666 6 666 66666 6 66 66666 6.6666 6 6663 .66 166.66 .66.6. 166.66 166.66 166.66 166.6. 66.6 66.6 66.6 66.6 66.6 66.6 m .666666 663 6666666 6666 66 666666 62 .66 166.61 166.66 I 166.66 166.6. I .6666 66666 66666 6666 66 6666666 666 vm.m 6m.m I ~m.m wm.m I x 00cm6666 uo: 66:00:66 60260 owmmusooco Oman H .66 I .0866 60366 166.66 166.66 166.66 166.66 166.6. 166.66 66666 66666666 666 66 6666666 666 66666666 666 mm.m vm.m ~m.m 66.m mm.m no.m M 0603 on: 66660 as :6 66:06:66 60560 cummusoosw H .oa 166.66 166.66 I 166.66 166.66 I .666 666 666 66666666 663 6666 66 666666 mh.m on.m I 6w.m mm.m I M 6623 Mo :o6c6mo G30 we so @6662 663 oc6umu a: .a 166.66 166.66 I 166.6. 166.66 I .666666 666 66666666 6 6663 66666 66.6 66.6 I 66.6 66.6 I m 6666 666 6666666 6 6666 66 666666 666 66666 6 .6 .>60 £060 mo new 166.66 666.66 166.66 166.66 .66.6. 166.66 6 666 66 66 66666666 66666666 6 66 6666 66 66666 66.6 66.6 66.~ 66.6 66.6 66.6 m 6666 666 6666666 6 6666 66 666666 666 66666666 6 .6 166.66 166.66 I 166.66 166.66 I 6 .666 6666 I 6 66 + 6 666666 6663 66.6 66.6 I 66.6 66.6 I m 66666 6666 66 66666 6 6666 66 666666 666 66666 6 .6 666.66 166.66 166.66 166.66 666.66 166.66 .6666666666 66666 ov.m 66.m wm.m m~.m oa.m m6.m m £663 uoomoua 6:6 we mco6mmaomwo umo60>6 H .m 166.6. 166.66 166.66 166.66 .66.66 166.6. .6666 66666 66.6 66.6 66.6 66.6 66.6 66.6 m 66 666666 66666 66 66 6666 66666 66666666 6 .6 166.61 .66.66 166.66 166.66 166.66 166.6. .66 66 666 .666666666 .66666666 6666663 .6666666 . 66.6 66.6 66.6 66.6 66.6 66.6 m 66666 6 6666 66666 6666 66666 66 66 66666666 6 .6 166.66 166.66 .66.66 .66.6. 166.66 .66.66 .66666 6666 66 ~6.~ 66.6 66.6 66.6 66.6 66.6 m 66666 6 6666 666 6666666 H 666 6666 .66666666 66 .6 166.66 166.66 166.66 166.66 166.66 6.66.66 .66666 6666 666 66.6 66.6 ~m.m 66.6 66.6 66.6 m 6666666 6 6666 66 666666 666 66666666 6 666 6666 .6 66666 6+zm :6 66666 6+26 26 unasoamo wuumwfiwno meuH 0666cc0666090 N ucmswuwaxm H ucwE660mxm 6co6u6psou can 666500 69 mEmuH ouwchowummso co wuoum 0m666>4 HH wanna 141 typically conformed to instructions. Excellent responses were obtained to items 3, 4, 6, 9, 10, ll, 12, and 13. Scores on these items were approximately 3.6 or higher. Students, thus, included the apprOpriate information in their reported study time (items 3 and 4), followed the rating procedure (items 6 and 9), conformed to a request that they not encourage other students not assigned the project to graph or rate their study time (items 10 and 11), and, reported honest responses (items 12 and 13). Table 12 Average Score on Questionnaire (Items 1-13) Assessing Conformity to Instructions by Course and Condition Experiment 1 Experiment 2 Condition Chemistry Calculus i s i 8 SM 3.49 (0.29) 3.48 (0.31) SM+C 3.56 (0.27) 3.58 (0.27) Note. The n's are as follows: Chemistry, SM = 46, SM+C—5‘52; Calculus, SM = 28, SM+C = 25. Good responses were obtained to items 1 and 5 with scores ranging from 3.1 to 3.5. Students indicated that each day they usually recorded the amount of time studied (item 1) and avoided discussions of the project with classmates in general (item 5). :-'_'=-: 142 Fair responses were obtained to items 2 and 7 with scores ranging from 2.3 to 2.9. Students indicated that they usually recorded the amount of study time at the end of studying or the end of the day (item 7); how- ever, they were between "rarely" and "usually" in their response to daily graphing. Follow-up conversation with students indicated that they typically plotted the graph at the end of the week. Student Opinion Items Students could select one of four responses to items 14, 15, and l6--"strongly agree," "agree," "dis- agree," "strongly disagree." These responses were scored 4, 3, 2, 1 respectively. Students typically disagreed with the statement that the project increased their study time (item 14). They also disagreed with statements about continued use of the procedures (items 15 and 16). Correlations between Reported Study Time andiMidterm II Correlations were computed between average reported study time and grade on midterm II in chemistry £_= -.033, n = 98; in calculus, £_= .004, n = 53. There were no significant correlations. ”I. I. I p ’ Jll‘”..u.3:’nl. CHAPTER V DISCUSSION The discussion of results obtained in the present investigation will follow the order of presentation used in the preceding chapter. Discussion will begin with the primary dependent variable of interest, academic per- formance, and the two related hypotheses under investi- gation. Attention will, then, focus on Hypothesis 3 and information on reported study time, followed by an exami- nation of the study time rating findings. A discussion of questionnaire responses will follow. Finally, conclusions about the present investigation will be drawn and recom— mendations for further research will be made. Academic Performance The conflicting results obtained when identical self-control techniques were used in two different classes is interesting in light of the earlier discussion of sub- ject selection bias in self-control investigations (McLaughlin, 1976). As noted earlier, the 15 self-control investigations examining the academic performance of 143 144 college students were limited to two basic groups of subjects. Subjects were either introductory psychology students or volunteers seeking assistance with study habits or other study behaviors. The present investi- gation, however, utilized a chemistry class and a calculus class. It is plausible that there are characteristics of various university courses or subject matter areas that interact with self-control techniques. For example, courses with a lower difficulty level could make the use of self-control techniques such as self-monitoring and self-rating superfluous because the course required little effort from a particular student or group of students. Or, a course could be so structured, including daily or weekly quizzes, that the effects of self-control tech- niques, designed to keep a student up to date in a class, would be unnecessary. The external structure and demands would be serving this purpose. In a similar example, a small seminar requiring informed participation could also make self-control techniques designed to assure daily preparedness unnecessary. Conversely, a student in a large lecture class with only one or two exams and optional attendance could find self-control techniques helpful when applied to completing reading assignments and attendance. In general, self-control techniques used in courses requiring mastery of specific content might be more helpful to students than when used in courses emphasizing 145 process rather than product, or a course requiring a creative product such as an essay or a research paper. In the latter case, however, self-control techniques designed to insure that a student begins writing early in the course, leaving time for review and rewriting, could be facilitative. These are just a few examples of aspects of courses that could differ and interact with self-control techniques--difficulty level, imposed structure, attendance requirements, class size and concomitant demands for par- ticipation, and type of academic product required. Mahoney, Moore, Wade, and Moura (1973) reported that self-monitoring was differentially effective in improving accuracy on sample problems on the Graduate Record Exam. No effects were obtained for verbal accuracy, however, accuracy was improved significantly in quantitative problems. It appears that the differential effectiveness of self-control techniques in different types of courses needs further investigation. It is clear from the present investigation that generalization about academic per- formance and self-control techniques across the full range of college level courses would be unwarranted. With regard to the hypotheses related to academic performance, conflicting results from Experiments 1 and 2 indicate only partial support for Hypothesis 1. In Hypothesis 1 it was predicted that students who utilized 146 self-control techniques would evidence superior test per- formance relative to controls. Evidence in support of this hypothesis was obtained in Experiment 1 using a chemistry class, however, significant differences between experimental and control groups were not obtained in Experiment 2. It was predicted in Hypothesis 2 that students in the SM+C condition would perform significantly better than students in the SM condition. No significant dif- ference in support of this hypothesis was obtained in either experiment. In chemistry, the students who simply monitored their study time (SM) evidenced slightly higher scores than those who monitored and used an overt self- rating procedure (SM+C). In calculus, the opposite find- ing was obtained. In both cases, however, the results do not depart significantly from results that could be obtained by chance. It is clear that the addition of self-reinforce- ment and self-punishment in the form of self-ratings did not enhance the self-monitoring effect when it occurred. Furthermore, it is apparent from the questionnaire results that students in both condition appeared to follow instructions well, eliminating the possibility of dif- ferential conformity to instructions. Bellack (1976) was previously quoted as asserting that "the explicit use of SR [self-reinforcement] critically augments whatever 147 effects SM [self-monitoring] has" (p. 74). This was certainly not the case in the present investigation. It is important to recognize, however, that self- rating, the particular operational definition selected for the self-reinforcement/5elf-punishment combination, was selected because of its approximation to the hypothe- sized internal evaluation process that is implicit in self-monitoring. Self-rating constitutes an overt mani- festation of the implicit processes hypothesized to con- tribute to the effects obtained with self-monitoring alone. Self-rating is a symbolic, as opposed to tangible, reinforcer. Perhaps there a£g_differences in the effec- tiveness of symbolic and tangible reinforcers that inter- act with a variety of characteristics of the subject, setting, or dependent variable. Such differences could account for the failure of the self-reinforcement/self- punishment procedure to augment the self-monitoring effect. It may be that Bellack (1976) is right, then, if self-reinforcement is defined more precisely to be tangible self-reinforcement. Certainly such reinforcers as money or points toward grades could be expected to exert a more powerful influence, relative to self— ratings. Bandura (1974a) indicated that he expects the insufficiency of material reinforcement to be recog- nized and the importance of covert self-reinforcement to 148 be acknowledged. "Most people value their self-respect above commodities," he writes (p. 863). It is not at all clear that they do. This is an empirical question and the differential effectiveness of covert self- reinforcement, overt symbolic reinforcers, and tangible reinforcers needs further investigation. In the present investigation, the failure to find differences between the SM and SM+C conditions is also a success in finding similarities. That is, it seems that the overt self-reinforcement/self-punishment procedures used in the SM+C condition do not differ sig- nificantly from self-monitoring procedures and their pre- sumed implicit self-rating components. This finding lends some support to the contention that self-monitoring effects result from more than merely recording behavior. Such effects could be due to a covert self-reinforcement,/ punishment process, much like that of self-rating. Reported Study Time Lessons can be drawn from conflicting results and evidence from the reported study time data provides a good example. The conflicting results obtained between Experiments 1 and 2 with regard to academic performance take on a new look when the reported study times for each class are examined. It would appear that the two classes differed markedly in difficulty level or amount of studying required. Study times were reported 149 by two-thirds of both classes, and despite the wide variability in individual study times, a course pattern was evident over the weeks. Clearly, students in the chemistry class began the experimental period with a much higher average daily study time relative to calculus students (2.5 times). (It should be remembered that the lst week of the experimental period was actually the fourth week of the course. It was also the week imme- diately following the first midterm exam.) By the third week, their average daily reported study time was 4.2 times higher than that of calculus students. Chemistry students were averaging approximately 2.5 hours a day, seven days a week. The study times reported by calculus students were never very high; neither did they increase very much even when a midterm exam was pending. Conversely, chemistry students' average times were quite high and their increases were dramatic. It appears that reported study time can serve as a kind of unobtrusive measure of student opinion about the difficulty levels of dif- ferent courses. Kanfer, Cox, Greiner, and Karoly (1974) have suggested that "more attention be given to a stage prior to execution of self-control in which promises, intentions, or performance criteria are developed since these events may determine later execution of self-control" (p. 606). 150 It is very possible that self-control techniques have little impact on a situation in which minimal self- control is required to obtain the desired results. If students have discerned that only a small amount of effort is required to attain their goal, which may be either a grade or knowledge of the subject matter, there is little more for self-control techniques to accomplish. A variety of situational and subject variables that affected the performance of self-control techniques were discussed in Chapter III. In particular, Kanfer and Duerfeldt (1967b) reported that both external reinforcement and self-reward procedures were equally ineffective in facilitating recall of material that had been overlearned. A similar circumstance may exist when minimal effort toward initial learning is required. Some of the conflicting results in self-control applications could be due to such uncon- trolled effects of course or task difficulty level. These factors are deserving of further investigation. With regard to the reported increases in study time over the experimental period, it is important to recognize that these increases cannot be attributed to the experimental manipulations--the self-control tech- niques. Many other factors could have influenced study time during the three-week period and they are confounded with the treatment. It is appropriate, however, to attribute the obtained differences in academic performance (in Experime of the prese performance. the control act of repox variable or conditions. that the con Hence, it is increased 0v similarly f C1358. and two Classes, to the eXpe a control gr None in reported 1 for Whom We accounting f academic per 151 (in Experiment 1) to the self-control techniques because of the presence of information about a control group's performance. Information on the reported study time of the control groups is not available, however, because the act of reporting study time was part of the independent variable or experimental manipulation for the SM and SM+C conditions. This investigation was designed precisely so that the control groups would not monitor study time. Hence, it is legitimate to observe that study time increased over the three-week period, that it increased similarly for the two experimental conditions within each class, and that it increased differentially between the two classes. The increases, however, may not be attributed to the experimental manipulation without information about a control group. Nonetheless, the differential rates and increases in reported study time, on the two-thirds of the students for whom we do have data, offer valuable information in accounting for the conflicting results with regard to academic performance in the two experiments. Furthermore, the fitted linear model from the multivariate repeated measures analysis depicted in Figure 1 permits the observation that the slopes of the lines are different for the two classes. This suggests the possibility that the experimental treatments found to be effective in chemistry enhanced the increase in study time that would be obtained It may be t1 hlwhich the tive, repres suggestions emphasized t. The in both clas aSSistance j the two eXpe Only availat r“-‘Cluired to the thrEe ex Class is pOS Inte from eaCh Cl only rated t time. Ful‘th 152 be obtained in a three-week period between examinations It may be that the trend observed in the calculus class, in which the experimental manipulations were not effec- tive, represents a type of baseline. Although these suggestions seem reasonable from Figure 1, it must be emphasized that they are only conjecture. Self-rating of Study Time The self-ratings provided by the SM+C condition in both classes also provide valuable information and assistance in accounting for the discrepancies between the two experiments. Self-ratings are intentionally only available for the one experimental condition required to self-rate. As such, no comparisons among the three experimental conditions within an individual class is possible. Interestingly, comparison of the self-ratings from each class indicated that students in chemistry only rated their study time positively about 26% of the time. Furthermore, they gave such self-ratings largely during the week they were studying approximately 2.5 hours a day. Calculus students rated study time posi- tively for the first week when it was approximately 17 minutes a day during the third week. They rated study time positively about 48% of the time. These observations give further support to the notion that students in calculus simply required far less study 153 time than their counterparts in chemistry in order to complete the course or meet self-established performance criteria. Questionnaire Responses Conformity to Instructions The questionnaire was designed primarily to pro- vide an independent opportunity to determine whether stu- dents in the experimental conditions followed their instructions. Overall, students indicated a high level of conformity, averaging about a 3.5 to a 3.6 on a 4.00 scale. There were no significant differences between the two conditions within a class; neither were there any significant differences between the two classes. This is an important finding. The failure to obtain effects for the self-control techniques in Experiment 2 cannot be attributed to any difference in conformity to instructions between the two classes. Explanations for the discrepant results must be, and have been, sought elsewhere. In one setting the techniques enhanced academic performance; in another setting they did not. In both settings, however, students did what they were asked to do. Accuracy of Self-report Two of the items designed to assess conformity to instructions provide some insight about the accuracy of the sel of study 1 averaged t "usually" number 14, I recorded averaged 3 Thus, it a] study time: would be re conformity No esfimates 15 Possible me flaWed. F0. Subject to recorded stw investigati vised Settil their Self-, polated to 1 (Bristol 5. deemed WOrt] The length- It gations on f 154 of the self-report measure. Item 13 stated "my record of study time was honest." Responses to this item averaged to be about 3.75 which is a midpoint between "usually" and "always" on the scale. A related item, number 14, stated "when I didn't study on a given day, I recorded 0 in the daily study time log." Responses averaged 3.98 and 4.00 for Experiment 1 and 2 respectively. Thus, it appears reasonable to conclude that the reported study times were accurate and honest. Otherwise, one would be required to conclude that there was tremendous conformity among subjects in a lying behavior. No other attempts were made to obtain reliability estimates for the self-report measures. A variety of possible methods were considered but found to be seriously flawed. For example, an experimenter could require each subject to have another person verify the student's recorded study time. (Who verifies the verifier?) One investigation had students study occasionally in super- vised settings, and, then, assessed the reliability of their self-observations on these occasions and extra- polated to unsupervised settings and self-reports (Bristol & Sloane, 1974). Such techniques were not deemed worthwhile. The reliability issue has been discussed at length. It is this author's opinion that to do investi- gations on either covert behaviors or behaviors in a 155 nonsupervised setting one must reconcile oneself to the unavailability of formal reliability indices. Three steps were taken to assure honest, accurate responding. Students were assured that the information they provided was anonymous, would remain so, and would not affect their grades. Thus, a safe environment for honest reporting was established. Second, the report forms were developed, pretested, and revised to be certain that they were easily used. Hence, unintentional errors in reporting were minimized. Finally, independent assessment, via questionnaire, was made of the accuracy of their responses. These factors, together with examination of the actual logs and graphs, the presence of different colored inks for recording, the high variability between and within students' reports, and the presence of a great many zeros in study time logs would seem to provide suf- ficient assurance that the data are reasonably honest and accurate. Self-recorded study time was not the dependent variable of primary interest, however, even if it had been, this author would be satisfied that these procedures provided satisfactory self-report data. Student Opinion Items Three questions were asked of students to which they responded on a continuum of "strongly agree, agree, disagree, strongly disagree." Responses were coded 4, 3, 156 2, 1 respectively. When asked whether they planned to continue recording and graphing study time, the responses averaged approximately 1.8 across both conditions and both classes. Thus, students uniformly disagreed with the statement. "I will continue to record and graph my study time in this class“ (item 15). They also uniformly disagreed with a statement about use of the techniques in other classes, "I will try recording and graphing my study time in other classes" (item 16). Responses to this item averaged about 1.8 and 1.9 in Experiments 1 and 2 respectively. Lastly, students in both conditions and both classes disagreed with the statement "I believe I studied more for this class as a result of this project" (item 15). Thus, even students in the chemistry class, for whom the experimental manipulation produced significant differences in academic performance, did not believe the procedures increased their study time. As we have discussed, the procedures may or may not have increased their study time. An assessment of this cannot be made. However, the procedure did increase their academic performance. Perhaps the improvement in their academic performance was due to increased study time without the students' awareness that it resulted from the procedures. The relationship of study time to academic per- formance or academic behaviors is not a clear-cut one. 157 Only four studies of the 31 reviewed reported correlations between reported study time and academic performance. Beneke and Harris (1974) reported no significant cor- relation between the number of hours studied and the number of lessons completed (E = .184) and between hours studied and three semester grade point average gain (E = .027). They did report significant correlation for study time and cumulative gpa (5 = .493, p < .01). Bristol and Sloane (1974) reported correlations of .45 and .41 for study time and test score, although they did not report significance levels. Johnson and White (1971) found a range of significant correlations between .42 and .82 for different groups using cumulative grade points and study time over an eight-week period. Kaufman and O'Leary reported significant negative correlations between disruptive behavior and the amount of work com- pleted (5 = -.50, p < .005) although this was not found in all phases of the experiment. The present investigation reported zero order correlations in both classes between performance on midterm II and an individual's average study time. This seems reasonable in that the study time required to attain a particular level of performance would vary with the ability and background of each student. The precise way in which self-monitoring and self-rating of study time actually affects academic performance is not clear. 158 This issue will be discussed further in the section on the monitored behavior. A very few students did indicate that they believed the procedures increased their study time. Such comments were made in the open-ended section of the questionnaire and to the experimenter in conversation. One calculus student remarked that it was one thing to know she wasn't studying very much, and quite another to have that fact staring at her on a piece of paper. How- ever, such comments were few in number. Related Issues Demand Bandura has indicated that awareness is an important factor in behavior change procedures. He has described "reinforcement as an unarticulated way of designating appropriate conduct" (1974a, p. 862). Furthermore, he has observed that "not surprisingly, people change more rapidly if told what behaviors are rewardable and punishable than if they have to discover it from observing the consequences of their actions" (1974a, p. 862). Nevertheless, a decision was made in this investigation to withhold information about the actual purpose of the investigation to avoid confounding the experimental manipulation with an expectation for behavior change. It is common practice in self-control studies to indicate that the procedures are intended to 159 facilitate behavior change. This is due in part to the emphasis on therapeutic interventions. Thus, it is impossible to know whether the actual technique results in the obtained change, or, the expectation of change results in change. In this particular investigation, students were given the impression that descriptive data were being obtained about these two classes, required of many lower division students. Such information included average study time and students' perceptions of what constituted adequate study time. Only the faculty member in the course knew that such information was secondary to a com- parison of self-control techniques. Even the course teaching assistants were unaware. Thus, students remained unaware of the potential for change that could result from the techniques. Therefore, the change that was obtained cannot be attributed to expectancy effects. Prior External Reinforcement As discussed in Chapter II, self-control studies sometimes bring the subject's behavior, first, under control through external reinforcement. Self-control techniques are then implemented to maintain change. This circumstance also contributes to an expectation for change that was not present in this investigation. No prior external reinforcement methods were used. Thus, 160 subjects were not aware that a change in their behavior was being sought. They were under the impression that descriptive information was being obtained, through time sampling, that would be helpful to OASIS in its work with students. Hence, the absence of prior external reinforcement and the presence‘of an acceptable rationale for the project, enabled expectancy effects to be con? trolled. The change obtained in Experiment 1 cannot be attributed to expectancy effects. The Monitored Behavior The monitored behavior is another important facet of the present investigation that deserves examination and discussion. It is important to highlight the fact that self-control investigations on academic performance are typically arranged so that a particular behavior with a presumed relationship to academic performance is monitored. Study time, disruptive behavior, or number of pages read are three examples of behaviors that are monitored, followed by assessments of academic performance. In the larger field of self-control investigations, the monitored behavior is often the actual dependent variable of interest (i.e., nailbiting, hallucinating, smoking). The dependent variable is monitored directly in such cases. Weight control studies provide a good example of both methods. In some studies, subjects monitor eating behavior or calorie intake, however, the 161 dependent variable of interest is actually weight loss. Such studies can also be designed so that weight is the actual behavior monitored (i.e., daily weight charts). Similarly, studies on academic performance can be designed whereby the dependent variable and the behavior to be monitored are the same. For example, 2nd graders monitor the number of correct arithmetic problems com- pleted each day, rather than monitoring the amount of time spent on arithmetic, and subsequently examining arithmetic performance. This is an important distinction because the effects of monitoring a behavior assumed to influence another behavior is dependent on the cor- rectness of such an assumption and the strength of the relationship. The present investigation is one in which the actual behavior monitored (study time) was not the primary dependent variable of interest (academic performance). Obviously, there are settings in which such a procedure is preferable or the only Option available. Academic performance is assessed far too infrequently in most college classes to be used as the monitored behavior. However, it will be important for future self-control studies to examine whether differential effects are obtained as a result of a distinction between directly monitoring and consequating the particular dependent variable of interest versus one with a presumed relation- ship to this variable. 162 One final aspect to be considered about the moni- tored behavior is apparent in reviewing the study time reported by calculus students. Although the average time ranged from approximately 17 minutes to 37 minutes a day over the three-week period, the variability of daily scores was very high. A great many students recorded zero study time. Thus, we have a situation where the behavior monitored is actually the behavior to be elimi- nated, namely, not studying. Previously reported investi- gations have indicated that it is generally desirable to have subjects monitor the behavior to be increased. That was certainly the intention in selecting the behavior of study time. However, the outcome in calculus was the opposite and the behavior to be increased was often not being monitored. For many students in calculus, study time behavior was so infrequent that the monitoring pro- cedure ended up as a procedure whereby "not studying" was actually monitored most of the time. It is impossible to reduce this behavior below zero, however, it may be more difficult to increase a behavior through monitoring or reinforcement when the baseline is nonexistent. It would seem important in future investigation to determine whether the baseline of the behavior of interest was so low that shaping procedures or some kind of contracting should be used first. Such modifications would not have been appropriate for this particular 163 research project, but from a practical standpoint, shaping and contracting procedures may be essential to establish- ing a behavior such that it can, in fact, be monitored and consequated. Proportion of Self-reinforcement anHSSelf;punishment In general, punishment techniques are regarded as inferior to reinforcement procedures in obtaining behavior change, despite a great deal of evidence to the contrary. In examining the data on the self-rating of study time, it is interesting to note the preponderance of negative ratings in chemistry. For this class, then, the amount of self-punishment, operationalized as a negative self- rating, was three times the amount of self-reward. Con- versely, in calculus the amount of self-reinforcement and self-punishment was approximately equal. We have discussed a possible reason for this discrepancy, namely differing difficulty levels and study demands in each class. Nevertheless, the outcome is such that chemistry students administered self-punishment approximately 74% of the time whereas calculus students administered self- punishment approximately 52% of the time. It would appear that the combination of self- reinforcement and self-punishment was effective when the proportion of self-punishment was high. This finding casts some doubt on the notion that self-punishment is an 164 inferior behavior change technique. Future studies would do well to consider the reciprocal nature of the self-punishment/reinforcement contingency and examine the overall amount of reinforcement and punishment present in any experimental period. Isomorphism The assumption that "symbolic activities obey the same psychological laws as do overt behaviors" (Meichen- baum & Goodman, 1971, p. 125) has been very important in the development of mediational learning theory and self- control. However, the notion that covert and overt reinforcement/punishment are equally effective in all situations is not essential. In the present investigation, some question has been raised about the relative strength of covert and overt symbolic reinforcement/punishment processes. The similarities between the effects of the SM and SM+C con- dition indicate that the addition of overt symbolic reinforcement/punishment process, in this instance, did not enhance self-monitoring effects which appear to result from similar processes that are covert. Yet, prior studies have indicated that reinforcement does augment the self-monitoring effect. It will be important for future investigations to attempt to assess whether such augmentation occurs with tangible reinforcers rather than symbolic. 165 One possible design to address such a question would compare a control group and subjects who self- monitored, self-monitored and self-rated (overt symbolic reinforcement/punishment process), and self-monitored plus administered a tangible se1f-reinforcement/punishment contingency. This would permit a comparison of two types of reinforcement—punishment processes-~symbolic and tangible. Furthermore, it would also permit a replication of the present investigation comparing presumed covert symbolic and overt symbolic self-reinforcement/punishment processes . Conclusions The results of the investigation have been analyzed and discussed. Partial support for the effec- tiveness of two particular self-control techniques in improving classroom academic performance was demonstrated. Conflicting results from a chemistry and calculus class were examined and partially explained through information on differential course study times and associated diffi- culty levels. Furthermore, questionnaire results indi- cating overall conformity to instructions between con- ditions and between classes eliminates the possibility of differential conformity to instruction as an expla- nation for the conflicting results. The failure to find differences between the two experimental con- ditions, SM and SM+C, was interpreted as some evidence 166 of success in finding similarities. That is, the effects of self-monitoring may be due to a covert self-evaluative process much like the overt self—rating procedure in the SM+C condition. The lack of significant differences between the two experimental conditions in both courses lends support to this conceptualization. Recommended areas for further research have been made throughout the discussion and will be reiterated in closing. The present investigation suggests the impor- tance of the following issues to a more complete under- standing of the self-control process. 1. Additional investigations aimed at isolating the self-monitoring effect should be undertaken using covert symbolic, overt symbolic, and overt tangible reinforcers as well as control groups for comparison. The differential effectiveness of symbolic and tangible reinforcers in self-control investigations of academic performance has not been examined. Likewise, the dif- ferential effectiveness of covert and overt symbolic reinforcers on academic performance needs investigation. 2. Self-reinforcement and self-punishment should be conceptualized as integrally related processes. Assessments of the pr0portion of self-reinforcement and self-punishment in a given investigation should be made. Attempts to determine the most efficacious pro- portions should be undertaken. 167 3. Subjects should be selected from a wider variety of subject matter areas to diminish the subject selection bias that exists at present in investigations of self-control and academic performance. 4. Careful attention should be paid to the actual behavior that is being monitored. Future studies should focus on the relative effectiveness of monitoring the behavior to be increased versus the behavior to be decreased. 5. Investigations should seek to determine whether direct monitoring and consequating of the primary dependent variable is superior to monitoring a presumably related behavior. Then, when possible, the preferable technique could be utilized. 6. Efforts should be made to isolate the factor influencing academic performance when study time is the behavior actually monitored. Academic performance-study time correlations are often small and rarely significant, thus, it is important to determine what the active ingredient is in monitoring study time that contributes to improvement in academic performance. 7. Where possible, comparisons should be under- taken to determine whether conditions like SM and SM+C actually result in increased study time. Such a study 168 would probably have to be conducted in a supervised setting in which unobtrusive assessment of a control group's study time could be made. 8. Course or task difficulty level should be varied systematically and its effect upon self-control techniques should be investigated. 9. Attempts to ascertain the importance of expectancy effects should be undertaken. Comparison should be made of a group which is told that the self- control procedures utilized typically enhance performance versus an uninformed group. The importance of subject awareness could be ascertained similarly. APPENDICES APPENDIX A CALCULUS MIDTERM EXAM 2 APPENDIX A CALCULUS MIDTERM EXAM 2 Mathematics 18 Hour Exam (2) 2/23/77 Each intergral is worth 6 points; you receive 10 points for stating your name correctly. Show all work and each answer. An integral which is done "in your ead"&for which there is no work to show must be accompanied by a derivative which checks the integra- tion. 1 k _ *“ _ l. f(1+2X) dX - 2. fm— dX - 8 cos x 3. f(1+-Sin x) cos x dx = 4. fV1+-sin x dx = x l 5°fx/1-(2x)2_d"= 6'N1- (2xTIdxz cos x 1 7' [V1 - sinzx dx = 8' f 25 + x7 dx== 9. I log x dx = 10. I log (x2) dx = 11. - sin: X +c052 X _ \/sin X “9 dx‘ 12.]e cosxax: Vsin x +1 13. f xexdx = 14. f 1 -: r77!“- ’ 8 15. £ (1 + 2x) dx = 169 APPENDIX B CHEMISTRY MIDTERM EXAM 2 APPENDIX B CHEMISTRY MIDTERM EXAM 2 Science SA - 1977 Name Second Exam Section Instructions: 1) Write your name and recitation section on all pages. 2) 'Read over all questions and work problems in order of ease. 3) The point value assigned to each question appears in parenthesis. u) Make sure there are no pages missing in your exam booklet. 5) Turn to Problem 6 only after you have finished with all others. This is an extra credit question. 6) log 2 = 0.3 Scoring: Number Points Available Points Scored 1 16 2 22 3 18 u 20 5 2H 5 + TOTAL ' 100+ 170 Name 1. 2b 171 Section (16 Points). Circle statements that are right and cross out statements that are wrong. a) A catalyst shifts the equilibrium towards more products. b) A catalyst shifts the equilibrium towards more reactants. c) The equilibrium constant can be changed by changes in temperature. d) The equilibrium constant can be changed by changes in pressure. e) An increase in pressure yields more products for the following reac- tion: 2A + B = “C f) The best conditions for maximum yield in the ammonia synthesis are low pressure and high temperature (ammonia synthesis is exothermic). g) The conjugate base A" to the weak acid HA is a weak base. h) A lMHClhaspI-ial. (22 Points) (16 Points). For the reaction A + as + 60 3 90 , you are given 0.9 mole A, 8 mole B and 2 mole C. a) (u Points). What is the limiting reagent? b) (6 Points). How many moles of D are produced? c) (6 Points). How many moles of A are left over? (6 Points). For 3A + B t 2C , K 8 0.5. At equilibrium, 2 moles of A and 1 mole of B are detected, How many moles of C are there? (Assme a 1 liter flask for the reaction) Name 3. 3a 3b 3c 4a 172 Section (18 Points) (6 Points). how many grams of acid H3A are needed to neutralize 90 g.of base B(0H)2? (Strong acids and bases) Molecular weights (mole mass): H3A 90 gmo'l, B(0H)2 60 ng“l (6 Points). How many grams of B(0H)2 are in 3 ”ters of a 0.6 N solution of B(OH)2? 1 liter of 0.6 N B(0H)2 is titrated to equivalence with 2 (6 Points). What is the molarity of H3P0g? liter Of H3P0g. (20 Points) What is the molarity of HCl? (6 Points). A solution of HCl has pH = 2.7. (8 Points). How many liters of H20 do you have to add to 1 liter of 1 M HCl to reach pH = 3? (6 Points). A 0.1 M acid RA has pH = 3. What is the equilibrium constant (1(a) of the acid? 173 Name Section S. (2“ Points) 5a (8 Points). *What is the pH of 1 liter solution prepared from 0.1 moles of NaOH and 0.1 moles of HA (Ka = 10's)? Sb (8 Points). What is the pH of 1 liter solution made from 0.01 moles NaOH and 0.11 moles HA (Ka = 10-5)? Sc (8 Points%. What is the pH of 1 liter solution made from 0.1 moles of HA (Ka = 10’ )7 ' 6. (+, extra credit). What is the of a solution that is 0.01 H in "H3 and 0.1 H in NHHCl? Assume Ka = 10‘ ° for NH.+ It NH3 + H+ APPENDIX C CALCULUS MIDTERM EXAM 1 APPENDIX C CALCULUS MIDTERM EXAM 1 Mathematics 18 Hour Examination (1) 1/31/1977 Each cuestion is worth 25 points. Show all work and your answers . 1. Find the following derivatives: d 2 + Tfi-(ex1)= ;%<1og(2x+1n= .% (log (eSln "n = z 7%- (sin (eyn = l d t _ Tfi? (e ) - 2' Ed? (sin (x + % )) = ; 3%- (cos (sin 0)) = __‘L dt (sin (nt)) = 1%? (sinzx + coszx) = ‘0 ?gr (cos (elog (sin x)) 3. A. Sketch carefully the graph of x + sin (nx) for 0 S x s 1. In particular, indicate the values of sin (nX) for x = 1 1 .l ‘1 .3 6, 4, 3, 2, 4, B. /, AC is 10 miles and g, , ACE is % C47 How long is AB? 4. A certain population of cells is 10° at the time t = 0. In two hours the population is 10’. Write the equation for the population y(t), where t is measured in hours. 174 APPENDIX D CHEMISTRY MIDTERM EXAM l APPENDIX D CHEMISTRY MIDTERM EXAM 1 Science 3A - 1977 First Exam Name Instructions: 1) Write your name and recitation section on all pages. 2) Read over all questions and work problems in order of ease. a) The point value assigned to each question appears in parenthesis. u) Make sure there are no pages missing in your exam booklet. 5) Turn to Problem 6 only after you have finished with all others. This is an extra credit question. 6) R = 0.08 liter atm K-l mole“ N = 6 x 1023 Scoring: Number Points Available Points Scored 1 18 2 20 3 20 A 22 S 20 6 + TOTAL 100+ 175 1a. 1c. 1d. 2a. 176 Name Section (2 Points) - Who's done it? Mass conservation (H Points) - What is the fusion reaction in the sun? (u Points) - How many protons and neutrons are there in 3He? (8 Points) - How many atoms are there in l g of element sOE? (8 Points) - If one gram of metal reacts with 2 g of ‘50, what is the equi- valent weight (equivalent mass) of the metal? (12 Points) - Assume 6 g 12C form a compound CO; with an unknown amount of oxygen. What is this amount of ‘50; in grams? What is the density of €02 gas at STP? 177 Name Section 33. (10 Points) - Assume an 8 liter glass container explodes when it reaches a pressure of 10 atm. At what temperature will the container explode when M g H2 are enclosed? 3b. (10 Points) - What is the molecular weight (mole mass) of a gas, 0.2 grams of which are enclosed in 0.8 liters at 300 K exerting a pressure of 10 atm? “a. (12 Points) - Assume 02 has a velocity (root mean square velocity) of 500 ms"1 at 300 K. At what temperature will it have a velocity of 2000 ms’ ? 178 Name Section (10 Points) - Assume a l molal aqueous solution of CaClz to have a freezing uh. point of -3°C. What then is the freezing point of a 0.1.molal solution of NaCl 5a. (12 Points) - An electric current separates on g Cu out of a solution of CuClz in 1200 seconds. For how many seconds will you have to let this cur- rent flow through an aqueous solution of CrCl; to separate 10.6 g Cr at weight Cu 64 and Cr 53? Sb. (8 Points) - Order the following ions in order of increasing deflections in a mass spectrometer (put strongly deflected ions right): 0+ , O2+ , He+ 5+ (Extra credit) - You want to lift SH kg with a hydrogen balloon. What volumel = 29 g mole' do you need at 300 K and 1 atm? Average molecular weight of air APPENDIX E COVER LETTER APPENDIX E COVER LETTER UNIVERSITY OF CALIFORNIA. SAN DIEGO blRKlLI'J' - mun. - must: - IO) AKLlLFs - Muir‘s“): - «As DIFLU - sAV FRANCIMO SANTA BARBARA - SANTA CRUZ OFFICE OF ACADEMIC SUPPORT LA JOLLA. CALIFORNIA 92093 AND INSTRUCTIONAL SERVICES January 28, 1977 Dear Class Member: Questions? Call OASIS--X3760--Carmel Myers Your participation in the project described on the next page is essential and valuable to me. The written description of your task is meant to be readily understandable. However, if the description is not totally clear, I urge you to call me at OASIS at 452-3760 if you have any questions about the task. Please do not talk with your friends and fellow students about the directions. Confidentiality Not only is your participation essential, but so is your confi- dentiality. I must request your cooperation during the next three weeks and ask that you do not discuss your particular task with other class members or friends. Different members of the class will be providing me with different types of information. It is important that the information I receive from you reflects solely your point of view and the task you've been assigned. As soon as I have summarized the information I have receive from you, I will prepare a written report and invite you to a meeting where we can discuss the project. You can earn $50.00! The project requires you to provide me with information each week. For each week you complete and return your assigned task, a token with your code number will be placed in a large jar. You must participate every week and complete all the tasks for your tokens to remain in the jar. At the end of the project, a drawing will be held and if your token is drawn you will receive a crisp $50 bill! Good luck to you and thank you for your assistance. Sincerely, Carmel Myers Assistant Director DM:j1 179 APPENDIX E SM INSTRUCTION SHEET APPENDIX F SM INSTRUCTION SHEET UNIVERSITY ()I" (T;\LII*‘()H.\'IA. SAN DIEGO "LHLHIV - Imus ° HHIKI - lus AMII ls ' lll\ LIIHIH - \«V DIN“) - \«V lReM [\(‘O sax‘la BARBARA - \\\I.»\ I my OFFICE OF ACADItMIC SUPPORT LA JOLLA. CALIFORNIA ‘L‘U‘Ii ANI) INSTRUCTIONAL SERVICES Students' problems in class sometimes result from inadequate amounts of study time. It is important for us to know the typical amount of time spent daily in studying for this class if we are to give reasonable advice and assistance to those who request it. You are part of a group of students who have been randomly chosen to assist in determining this average by recording daily the time you spend in this class. WHAT to record Please add up all the time you spend each day on reading, outlining, working problems, and so on for this class. Please do 225 count the time you spend in class in your daily summary. If you do not spend any time on a given day, record that also. Please make certain your record is complete. honest and accurate. WHEN to record Record your total study time each day as soon as you have finished all the work you will do that day in this class. If you do not spend any time, record that at the end of the day. WHERE Graph report forms are attached to this explanation. The steps to be completed are listed here. 1. Record on the line labeled "Log of Daily Study Time” the total amount of time you spent that day on this class in minutes. 2. Plot your time on the graph. Put a dot across from the number of minutes you studied that day on the line for that day. Connect the dots as the week goes on. An example is shown below. 90 —u 80 I 0 60 50 40 gr I 20 I0 0 IHR M T VV 8 E0120 II 70 II (LII o I 65[6) LOG OF DAILY STUDY TIME IN MINUTES HOW to return theygraph Your graph for the past week will be collected at the first lecture session each week. Specific due dates are written on the graphs. GENERAL INFORMATION It is important for you to know that the information you provide about study time is confidential and your graph reports are coded to assure such confidentiality. Your instructor will not see any of the information you provide in the graphs and your grade will not be influenced in any way by your participation in this investigation. Your full cooperation and assistance with this study are essential and appreciated very much. 180 APPENDIX G SM+C INSTRUCTION SHEET APPENDIX G SM+C INSTRUCTION SHEET UNIVERSITY OF CALIFORNIA. SAN DIEGO alnuu \ 'nxxn - mun - um \v uh ' sun-um: - H\ MIC“ ‘ “x IIANfN U sin—TA nAKuuu - 3A\TA «In I OFFICE or M‘ADEMIC summer LA JOLLA. CALIFORNIA ozw-u AND INSTRU‘TIUNAL SERVICES Students' problems in class sometimes result from inadequate amounts of study time. It is iqaortant for us to know the typical amount of time spent daily in studying for this class if we are to give reasonable advice and assistance to those who request it. You are a part of a group of students who have been randomly chosen to assist in determin- ing this average by recording daily the time you spend in this class. IRA? to record Please add up all the time you spend each day on reading, outlining, working problems, and so on for this class. Please do not count the time you spend in class in your daily wry. If you do not spend any time on a given day, record that also. Please mks certain your record is coqleta, honest and accurate. ‘ WHEN to record Record your total stay tin each day as soon as you have finished all the work you will do that day in this class. If you do not spend any time, record that at the end of the day. wan: to record Graph report forms are attached to this explanation. The steps to be ooqleted are listed here. 1. Record on the line labeled "Log of Daily Study Time' the total munt of time you spent that day on this class in minutes. 2. Plot your ti. on the graph. Put a dot across from the humor of minutes you studied that day on the line for that day. Connect the dots as the week goes on. An example is shown below. N 70 ID I HR 50 40 JG 70 IO 0 5O 20 70 O O 65 I5 STUDY HOW to return the grfl Your graph for the past week will be collected at the first lecture session each week. Specific due dates are written on the graphs. SELF MATH]! On your graph below the line labeled 'log of Daily Study Time“ is space for you to rate the mount of time you spent on this course each day. Your self-rating should be based on your own opinion and standards. For crawls. if you think the mount of time you spend on this particular class was adequate, you should give a positive (+) rating. If you do not think the time was adequate, you should give a negative (-) rating. Record one of tho syfiols below each day and please do not leave any days blank. positive + negative - It is important that your self-rating be based on the amount of study time for the day and not the quality of the time spent. An exapla is shown below. F60 ll so H 45 ll Ice" 76 [IzoJI 1:1 L00 0! DAILY STUDY Tm: m mNUTes [--II+IV+ ll+ll+I—JI+] SELF-RATING OF AMOUNT OF STUDY TIME GENERAL INFORMATION It is important for you to know that the information you provide about study time is confidential 8114 your graph reports are coded to assure such confidentiality. Your instructor will not see any of the information you provide in the graphs and your grade will not be influenced in any way by your participation in this investigation. Your full cooperation and assistance with this study are essential and appreciated very much. 181 APPENDIX H CONTROL INSTRUCTION SHEET 1:- APPENDIX H CONTROL INSTRUCTION SHEET UNIVERSITY OF CALIFORNIA. S.\.\' DIEGO sinuttx . O\\I\ - IInIsr - UH \\I.I.Ln - mtuumt - >\\ mrtu - nx Mushy“) 5\\T\ B\K.|a&.\ - S\\T.\ C912 OFFICE OF ACADEMIC SL'PPOKT LA JOLLA, CALIFORNIA 9209.3 AND INSTRUCTIONAL SERVICES We are attempting to learn about the kinds of activities or exper- iences which students have outside of class that are equal to or greater in importance to course work. You are part of a group of students who have been randomly chosen to assist in providing this information. Please use the attached report forms to provide the necessary infor- mation. Your form for the past week will be collected at the first lecture session each week. Specific due dates are written on each form. —....—.. It is important for you to know that the information you provide is confidential and your report forms are coded to assure such confidentiality. Your instructor will not see any of the informa- tion you provide and your grade will not be influenced in any way by your participation in this investigation. Your full cooperation and assistance with this study are essential and appreciated very much. 182 APPENDIX I SM REPORT FORM MORE THAN 360 MINUTES CLASS TOTAL TIME STUDIED IN MINUTES FOR NOTE: IF TIME STUDIED IN THIS ONE CLASS EXCEEOS sso MINUTES IO HOURS) LOG EXACT AMOUNT APPENDIX I SM REPORT FORM CODE NUMBER GRAPH NUMBER DAY AND DATE DUE IN CLASS 360 6 H R S 350 340 330 320 310 5 H R S 290 280 270 260 250 240 230 220 210 200 190 180 I70 160 150 140 I30 I20 I IO 4 HRS 3 HRS 2 HRS 80 70 I H R 50 40 30 20 FMJEJI WJI—TUI F I 1sz L06 OF DAILY STUDY TIME IN MINUTES 183 APPENDIX J SM+C REPORT FORM APPENDIX J SM+C REPORT FORM CODE NUMBER GRAPH NUMBER DAY AND DATE DUE IN CLASS MORE THAN 360 MINUTES 6 HRS 350 330 320 3 I 0 300 290 280 270 CLASS 5 HRS 250 240 230 220 TI 0 4 HRS 3 HRS 2 HRS TOTAL TIME STUDIED m MINUTES FOR 3 IHR TEL—TEL WJFTIF IL jFJ LOG OF DAILY STUDY TIME IN MINUTES [—TCjL JL JL I ]L__] SELF-RATING OF AMOUNT OF STUDY TIME NOTE: IF TIME STUDIED IN THIS ONE CLAQ EXCEEOS m MINUTES (6 HOURSI LOO EXACT AMOUNT 184 APPENDIX K CONTROL REPORT FORM APPENDIX K CONTROL REPORT FORM Code Number Report Number Day and date due in class Please describe in a few sentences an experience or activity for this week that you consider to be of equal or greater importance compared to your work in classes. 185 APPENDIX L SM QUESTIONNAIRE APPENDIX L SM QUESTIONNAIRE Code Number PLEASE CIRCLE THE RESPONSE THAT DESCRIBES YOUR BEHAVIOR DURING THIS PROJECT: Day and date due (Return with graph #3) 10. 11. 12. 13. Each day I recorded the amount of time Always Usually Rarely Never 1 studied for this class. In addition. each day I graphed the Always Usually Rarely Never time I spent in this class. I included in my study time totals time Always Usually Rarely Never spent reading, working problems, outlining, and so on. I excluded class time in my daily Always Usually Rarely Never totals of study time. I avoided discussions of the project Always Usually Rarely , Never with other classmates. I recorded the amount of time I Always Usually Rarely Never studied for this class as soon as I finished studying or at the end of' each day. I encouraged other students in my Always Usually ' Rarely Never class who were not assigned the ‘ project to try graphing their study time. My record of time studied was honest. Always Usually Rarely Never When I didn't study in a given day I Always Usually Rarely Never recorded 0 in the daily study time log. I believe I studied more for this Agree Agree Disagree Disagree class as a result of this project. Strongly Strongly I will continue to record and graph Agree Agree Disagree Disagree my study time in this class. Strongly Strongly I will try recording and graphing .Agree Agree Disagree Disagree my study time in other classes. Strongly Strongly Additional cements or suggestions. (Please use back of page as well). 1136 APPENDIX M SM+C QUESTIONNAIRE 10. 11. 12. 13. 14. 15. 16. 17. APPENDIX M SM+C QUESTIONNAIRE PLEASE CIRCLE THE RESPONSE THAT DESCRIBES YOUR BEHAVIOR DURING THIS PROJECT: Each day I recorded the amount of time I studied for this class. In addition, each day I graphed the time I spent in this class. I included in my study time totals time I spent reading, working problems, outlining, and so on. I excluded class time in my daily totals of study time. I avoided discussions of the project with other classmates. I rated the amount of time I spent on this class with either a + or - each day. I recorded the amount of time I studied for this class as soon as I finished studying or at the end of each day. I rated the amount of time I studied for this class when I recorded the amount. My rating was based on my own opinion of what amount of time was adequate for the day. I encouraged other students in my class who were not assigned the project to try graphing their study time. I also encouraged other students not assigned the project to rate their study time. My record of time studied was honest. When I didn't study on a given day I recorded 0 in the daily study time log. I believe I studied more for this class as a result of this project. I will continue to record and graph my study time in this class. I will try recording and graphing my study time in other classes. Additional comments or suggestions. (Please use back of page). 187 Always Always Always Always Always Always Always Always Always Always Always Always Always Agree Strongly Agree Strongly Agree Strongly Code Number and Date Due (Return with graph #3) Usually Usually Usually Usually Usually Usually Usually Usually Usually Usually Usually Usually Usually Agree Agree Agree Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Rarely Disagree Disagree Disagree Never Never Never Never Never Never Never Never Never Never Never Never Never Disagree Strongly Disagree Strongly Disagree Strongly LIST OF REFERENCES LIST OF REFERENCES Aronfreed, J. Conduct and conscience: The socialization and internalized control over behavior. New York: Academic Press, 1968. Axelrod, S., Hall, R. V., Weis, L., & Rohrer, S. Use of self-imposed contingencies to reduce the fre— quency of smoking behavior. In M. J. Mahoney & C. E. Thoreson (Eds.), Self-control: Power to the person. Monterey, Cal.: Brooks7Cole, 1974. Balfour, J. D. A comparison of the effects of external control in the modification and maintenance of weight. Journal of Abnormal Psychology, 1974, §§, 404-410. Balfour, J. D., & Christenson, E. R. Behavior therapy versus willpower in the management of obesity. Journal of Psychology, 1975, 22, 303-311. Ballard, K. D., & Glynn, T. E. Behavioral self-management in story writing with elementary school children. Journal of Applied Behavior Analysis, 1975, g, 337-398. Bandura, A. Principles of behavior modification. New York: Holt, Rinehart, Winston, 1969. Bandura, A. Vicarious and self-reinforcement processes. In R. Glaser (Ed.), The nature of reinforcement. New York: Academic Press, 1971. Bandura, A. Behavior theory and the models of man. American Psychologist, 1974, $2! 859-869. (a) Bandura, A. The case of the mistaken dependent variable. Journal of Abnormal Psychology, 1974, fig, 301- §U3 . (bY Bandura, A. Self-reinforcement: Theoretical and methodo- logical considerations. Behaviorism, 1976, 4, 135-155. 188 189 Bandura, A. Social learning theory. Englewood Cliffs, N.J.: Prentice Hall, 1977. Bandura, A., Grusec, J. E., & Menlove, F. L. Some social determinants of self-monitoring reinforcement systems. Journal of Personality_and Social Psy- chology, 1967, 2, 449-455. Bandura, A., & Kupers, C. J. Transmission of patterns of self-reinforcement through modeling. Journal of Abnormal and Social Psychology, 1964, 69, 1-9. Bandura, A., & Mischel, W. Modification of self-imposed delay of reward through exposure to live and symbolic models. Journal of Personality and Social Psychology, 1965, g, 698-705. Bandura, A., & Perloff, B. Relative efficacy of self- monitored and externally imposed reinforcement systems. Journal of Personality and Social Psychology, 1967, 1, 111-116. Bandura, A., & Walters, R. H. Acquisition of self-control through modeling. In R. C. Johnson, P. R. Dokecki, & O. H. Mowrer (Eds.), Conscience, contract and social reality. New York: Holt, Rinehart, & Winston, 1972. Bandura, A., & Whalen, C. K. The influence of antecedent reinforcement and divergent modeling cues on patterns of self-reward. Journal of Personality and Social Psychology, 19667I2, 373-382. Bass, B. A. Reinforcement history as a determinant of self-reinforcement. Journal of Psychology, 1972, g1, 195-203. Bellack, A. S. Internal vs. external locus of control and the use of self-reinforcement. Psychological Reports, 1972, El! 723-733. Bellack, A. S. A comparison of self-reinforcement and self-monitoring in a weight reduction program. Behavior Therapy, 1976, 1, 68-75. Bellack, A. S., Glanz, L. M., & Simon, R. Self-reinforce- ment style and covert imagery in the treatment of obesity. Journal of Consulting and Clinical Psychology, 1976, 44, 490-491. 190 Bellack, A. S., Rozensky, R., & Schwartz, J. A compari- son of two forms of self-monitoring in a behavioral weight reduction program. Behavior Therapy, 1974, 5, 523-530. Bellack, A. S., Schwartz, J., & Rozensky, R. H. The con- Bem, S. Beneke, Berecz, tribution of external control to self-control in a weight reduction program. Journal of Behavior Therapy and Experimental Psychiatry, 1974, 5, 245:250. ‘— L. Verbal self-control: The establishment of effective self-instruction. Journal of Experi- mental Psychology, 1967, 14, 485-491. M., & Harris, M. B. Teaching self-control of study behavior. Behavior Research and Therapy, 1972, 10, 35-41. J. Modification of smoking behavior through self-administered punishment of imagined behavior: A new approach to aversion therapy. Journal of Consulting and Clinical Psychology, 1972, 38, 244:?50. '__ Bolstad, O. D., & Johnson, S. M. Self-regulation in the modification of disruptive classroom behavior. In M. J. Mahoney & C. E. Thoreson, Self-control: Power to the person. Monterey, Cal.: BroOks/ Cole, 1974. Bristol, M. M., & Sloane, H. N. Effect of contingency Broden, contracting on study rate and test performance. Journal of Applied Behavior Analysis, 1974, 1, 271-285. M., Hall, R. V., & Mitts, B. The effect of self-recording on the classroom behavior of two eighth grade students. Journal of Applied Behavior Analysis, 1971, 4, 191-199. Catania, A. C. The myth of self-reinforcement. Cavior, Colle, Behaviorism, 1975, 3, 192-199. N., & Marabotto, C. M. Monitoring verbal behaviors in dyadic interaction. Journal of Consulting and Clinical Psychology, 1976, 44, 68—76. '— H. A., & Bee, H. L. The effects of competence and social class on degree of modeling of self-reward patterns. Psychonomic Science, 1968, $2, 231-232. 191 Davison, G. C. Self-control through "imaginal aversive contingency" and "one-downmanship": Enabling the powerless to accommodate unreasonableness. In J. D. Krumboltz & C. E. Thoreson (Eds.), Behavioral counseling: Cases and techniques. New York: Holt, Rinehart, & Winston, 1969. Dorsey, T. E., Kanfer, F. H., & Duerfeldt, P. H. Task difficulty and non-contingent reinforcement schedules as factors in self-reinforcement. Journal of General Psychology, 1971, 84, 323-334. Drabman, R. S., Spitalnik, R., & O'Leary, K. D. Teaching self-control to disruptive children. Journal of Abnormal Psychology, 1973, 83, 10-16. Epstein, L. H., Miller, P. M., & Webster, J. S. The effects of reinforcing concurrent behavior on self-monitoring. Behavior Therapy, 1976, 1, 89-95. Epstein, L. H., Webster, J. S., & Miller, P. M. Accuracy and controlling effects of self-monitoring as a function of concurrent responding and reinforce- ment. Behavior Therapy, 1975, 6, 654-666. Felixbrod, J. J., & O'Leary, K. D. Effects of reinforce- ment on children's academic behavior as a function of self-determined and externally imposed contin- gencies. Journal of Applied Behavior Analysis, 1973, g, 241-250. Felixbrod, J. J., & O'Leary, K. D. Self-determination of academic standards by children: Toward freedom from external control. Journal of Educational Psychology, 1974, 26, 845-850. Fink, W. T., & Carnine, D. W. Control of arithmetic errors using informational feedback and graphing. Journal of Applied Behavior Analysis, 1975, 8, 461 . '- Fox, L. Effecting the use of efficient study habits. In R. Ulrich, T. Stachnik, & J. Mabry (Eds.), Control of Human Behavior. Glenview, 111.: Scott, Foresman, 1966. Frederiksen, L. W. Treatment of ruminative thinking by self-monitoring. Journal of Behavior Therapy and Experimental Psychiatry, 1975, 2, 258-259. 192 Frederiksen, L. W., Epstein, L. H., & Kosevsky, B. P. Reliability and controlling effects of three procedures for self-monitoring smoking. Psycho- logical Record, 1975, 25, 255-263. Gewirtz, J. L. The roles of overt responding and extrin- sic reinforcement in "self-" and "vicarious- reinforcement" phenomena and in "observational learning" and imitation. In R. Glaser (Ed.), The nature of reinforcement. New York: Academic Press, 197i. Glynn, E. L. Classroom applications of self-determined reinforcement. Journal of Applied Behavior Analysis, 1970, 3, 123-132. Glynn, E. L., & Thomas, J. D. Effect of cueing on self- control of classroom behavior. Journal of Applied Behavior Analysis, 1974, 1, 299-306. Glynn, E. L., Thomas, J. D., & Shee, S. M. Behavioral self-control of on-task behavior in an elementary school. Journal of Applied Behavior Analysis, 1973, 6, 105-113. Goldiamond, I. Self-control procedures in personal behavior problems. Psychological Reports, 1965, ll, 851-868. Goldiamond, I. Fables, armadyllics, and self-reinforce- ment. Journal of Applied Behavior Analysis, 1976, 9, 521-525. (a? Goldiamond, I. Self-reinforcement. Journal of Applied Behavior Analysis, 1976, 9, $09-514. (b7 Gottman, J. M., & McFall, R. M. Self-monitoring effects in a program for potential high school dropouts: A time series analysis. Journal of Consulting and Clinical Psychology, 1972, 39, 273-281. Gulanick, N., Woodburn, L. T., & Rimm, D. C. Weight gain through self-control procedures. Journal of Consulting and Clinical Psychology, I975, 43, 536-539. '— Hall, H. V. Effects of direct and self-reinforcement as a function of internal-external control. Perceptual and Motor Skills, 1973, 31, 753-754. 193 Hall, 8. M. Self-control and behaviorist control in the behavioral treatment of overweight women. Behavior Research and Therapy, 1972, l9, 59-68. Hall, S. M. Behavioral treatment of obesity: A two year follow-up. Behavior Research and Therapy, 1973, ll, 647-648. Harris, M. B. Self-directed program for weight control: A pilot study. Journal of Abnormal Psychology, 1969, 14, 263-270. Harris, M. B., & Trujillo, A. E. Improving study habits of junior high school students through self- management versus group discussion. Journal of Counseling Psychology, 1975, 22, 513-517. Haynes, L. E., & Kanfer, F. H. Academic rank, task feedback, and self-reinforcement in children. Psychological Reports, 1971, 28, 967-974. Heldebrandt, D. E., Feldman, S. E., & Ditrichs, R. A. Rules, models, and self-reinforcement in children. Journal of Personality and Social Psychology, 1973, 25, 1-5. Herbert, E. W., & Baer, D. M. Training parents as behavior modifiers: Self-recording of contingent attention. Journal of Applied Behavior Analysis, 1972, 5, 139:I49. Herbert, E. W., Gelfand, D. M., & Hartmann, D. P. Imitation and self-esteem as determinants of self-critical behavior. Child Development, 1969, 49, 421-430. Hill, J. H., & Liebert, R. M. Effects of consistent or deviant modeling cues on the adoption of a self- imposed standard. Psychonomic Science, 1968, l2, 243-244. Homme, L. E. Perspectives in psychology--XXIV: Control of coverants, the operants of the mind. Psycho- logical Record, 1965, l5, 501-511. Hoon, P. W. Effects of self-monitoring and self-recording on ecological acts. Psychological Reports, 1976, ‘18, 1285-1286. 194 Horan, J. J., Hoffman, A. M., & Macri, M. Self-control of chronic fingernail biting. Journal of Behavior Therapy and Experimental Psychiatry, I974, 5, 307-309. Horan, J. J., & Johnson, R. G. Coverant conditioning through self-management application of the Premack principle: Its effect on weight reduction. Journal of Behavior Therapy and Experimental Psychiatry, 1971, 3' 243-249. Hutzell, R. R., Platzek, D., & Logue, P. E. Control of symptoms of Gilles de la Tourettes syndrome by self-monitoring. Journal of Behavior Therapy and Experimental Psychiatry, 1974, 5, 7l-76. Jackson, B. Treatment of depression by self-reinforce- ment. Behavior Therapy, 1972, 5, 298-307. Jackson, B., & Van Zoost, B. Changing study behaviors through reinforcement contingencies. Journal of Counseling Psychology, 1972, l5, 192-195. Jackson, B., & Van Zoost, B. Self-regulated teaching of others as a means of improving study habits. Journal of Counseling Psychology, 1974, El, 489- 4g3 O Jason, L. A. Rapid improvement in insomnia following self-monitoring. Journal of Behavior Therapy and Experimental Psychiatry, 1975, 5, 349-350. Jason, L. A. Self-monitoring in the treatment of nose squeezing and daydreaming. Psychological Reports, 1976, 55, 235-238. Jeffrey, D. Self-control: Methodological issues and research trends. In M. J. Mahoney & C. E. Thoreson, Self-control: Power to the person. Monterey, Cal.: Brooks/Cole,7i974. Johnson, S. M. Effects of self-reinforcement and external reinforcement in behavior modification. Proceed- ings of the 77th Annual Convention of the American Psychological Association,_1969, 4 (Pt. 2), 535- ‘536. Johnson, S. M. Self-reinforcement versus external rein- forcement in behavior modification with children. Developmental Psychology, 1970, 5, 147-148. 195 Johnson, S. M., & Martin, S. Developing self-evaluation as a conditioned reinforcer. In B. S. Ashem & E. G. Poser (Eds.), Behavior modification with children. New York: Pergamon, 1973. Johnson, S. M., & White, G. Self observation as an agent of behavioral change. Behavior Therapy, 1971, 2, 488-497. Kanfer, F. H. Influence of age and incentive conditions on children as self-rewards. Psychological Reports, 1966, l5, 263-274. Kanfer, F. H. Self-monitoring: Methodological limi- tations and clinical applications. Journal of Consulting and Clinical Psychology, 1970, 35, ‘IIa-lsz. "' Kanfer, F. H. Self-management methods. In F. H. Kanfer & A. P. Goldstein (Eds.), Helping_people change. New York: Pergamon, 1975. Kanfer, F. H., Bradley, M. M., & Marston, A. R. Self- reinforcement as a function of degree of learn- ing. Psychological Reports, 1962, l2, 885-886. Kanfer, F. H., Cox, L. E., Greiner, J. M., & Karoly, P. Contracts, demand characteristics, and self- control. Journal of Personality & Social Psy- chology, 1974, 52, 605-619. Kanfer, F. H., & Duerfeldt, P. H. Effects of pretraining on self-evaluation and self-reinforcement. Journal of Personality and Social Psychology, I967, 1, 164-168. (a) Kanfer, F. H., & Duerfeldt, P. H. Effects on retention of externally or self-reinforced rehearsal trials following acquisition. Psychological Reports, 1967, gl, 194-196. (b) Kanfer, F. H., & Duerfeldt, P. H. Motivational proper- ties of self-reinforcement. Perceptual and Motor Skills, 1967, 25, 237-276. (a) Kanfer, F. H., & Duerfeldt, P. H. Comparison of self- reward and self-criticism as a function of types of prior external reinforcement. Journal of Personalityiand Social Psychology, 1968, 8, 261-268. ‘ Kanfer, Kanfer, Kanfer, Karoly, Karoly, Kaufman, Kazdin, Kazdin, Lanyon, 196 F. H., Duerfeldt, P. H., & LePage, A. L. Sta- bility of patterns of self-reinforcement. Ps - chological Repcrts, 1969, 51, 663-670. F. H., & Marston, A. R. Conditioning of self- reinforcing responses: An analogue to self- confidence training. Psychological Reports, 1963, l5, 63-70. (a) F. H., & Marston, A. R. Determinants of self- reinforcement in human learning. Journal of Experimental Psychology, 1963, 55, 245-254. (b) P., & Doyle, W. W. Effects of outcome expectancy and timing of self-monitoring on cigarette smoking. Journal of Clinical Psychology, 1975, 5l, 351-355. P., & Kanfer, F. H. Situational and historical determinants of self-reinforcement. Behavior Therapy, 1974, 5, 381-390. K. F., & O'Leary, K. D. Reward, cost, and self- evaluation procedures for disruptive adolescents in a psychiatric hospital. Journal of Applied Behavior Analysis, 1972, 5, 293-309. A. E. Reactive self-monitoring: The effects of response desirability, goal-setting, and feed- back. Journal of Consulting and Clinical Psy- chology, 1974, 15, 704-716. Ta) A. E. Self-monitoring and behavior change. In M. J. Mahoney & C. E. Thoreson, Self-control: Power to the person. Monterey, Cal.: BrooKS/ Cole, 1974. 87b) R. I., & Barocas, V. S. Effects of contingent events on stuttering and fluency. Journal of Consulting and Clinical Psychology, 1975, 557 ‘786-793. Leitenberg, H., Agras, W. 8., Thomson, L. E., & Wright, D. E. Feedback in behavior modification: An experimental analysis in two phobic cases. Journal of Applied Behavior Analysis, 1968, l, 131-137. Liebert, R. M., & Allen, M. K. Effects of rule structure and reward magnitude on the acquisition and adoption of self-reward criteria. Psychological Reports, 1967, El, 445-452. 197 Liebert, R. M., Hanratty, M., & Hill, J. H. Effects of rule structure and training method on the adoption of self-imposed standard. Child Development, 1969, 19, 93-101. Liebert, R. M., & Ora, J. P. Children's adoption of self-reward patterns: Incentive level and method of transmission. Child Development, 1968, 55, 537-544. Liebert, R. M., Spiegler, M. D., & Hall, W. M. Effects of the value of contingent self-administered and noncontingent externally imposed reward on chil- dren's behavioral productivity. Psychonomic Science, 1970, l5, 245-246. Lipinski, D. P., Black, J. L., Nelson, R. 0., & Ciminero, A. R. Influence of motivational variables on the reactivity and reliability of self-recording. Journal of Consulting and Clinical Psychology, 1975, 357637-646. Lipinski, D., & Nelson, R. The reactivity and unrelia- bility of self-recording. Journal of Consulting and Clinical Psychology, 1974, 35, 118-123. Lovitt, T. C., & Curtiss, K. A. Academic response rate as a function of teacher and self-imposed contin- gencies. Journal of Applied Behavior Analysis, 1969, a, 49-54. Mahoney, M. J. The self management of covert behavior: A case study. Behavior Therspy, 1971, 3! 575- 578. Mahoney, M. J. Self-reward and self-monitoring tech- niques for weight control. Behavior Therapy, 1974, 5, 48-57. Mahoney, M. J. Terminal terminology: A self-regulated response to Goldiamond. Journal of Applied Behavior Analysis, 1976, g, 515-5I7. Mahoney, M. J., Kazdin, A. E., & Lesswing, N. J. Behavior modification: Delusion or deliverance. In D. M. Franks & G. T. Wilson (Eds.), Annual review of behavior therapy: Theor practice. New York, N.Y.: Brunner/Mazel, l9 4. 198 Mahoney, M. J., Moore, B. S., Wade, T. C., & Moura, N. G. Effects of continuous and intermittent self- monitoring on academic behavior. Journal of Consultingand Clinical Psychology, I973, 4I, 65-69. -—' Mahoney, M. J., Moura, N. G., & Wade, T. C. Relative efficacy of self-reward, self-punishment, and self-monitoring techniques for weight loss. Journal of Consulting and Clinical Psychology, 1973, 55, 404-407. Maletzky, B. M. Behavior recording as treatment: A brief note. Behavior Therapy, 1974, 5, 107-111. Marston, A. R. Personality variables related to self- reinforcement. Journal of Psychology, 1964, 55, 169-175. (a) Marston, A. R. Response strength and self-reinforcement. Journal of Esperimental Psychology, 1964, 55, 537-540. (b) Marston, A. R. Imitation, self-reinforcement and external reinforcement of another person. Journal of Personality and Social Psychology, I965, a, 255-26I. Marston, A. R. Self-reinforcement and external reinforce- ment in visual-motor learning. Journal of Experimental Psychology, 1967, 11, 93-98. Marston, A. R. Effect of external feedback on the rate of positive self-reinforcement. Journal of Esperimental Psychology, 1969, 55, 175-I79. Marston, A. R., & Cohen, N. J. The relationship of nega- tive self-reinforcement to frustration and intro- punitiveness. Journal of General Psychology, 1966, 13, 237-243} Marston, A. R., & Kanfer, F. H. Human reinforcement: Experimental and subject controlled. Journal of Esperimental Psychology, 1963, 55, 91-94. Marston, A. R., & Smith, F. J. Relationship between reinforcement of another person and self-rein- forcement. Psychological Reports, 1968, 53, 83-90. 199 Martin, J. E., & Sachs, D. A. The effects of a self- control weight loss program on an obese woman. Journal of Behavior Therapy and Experimental Psychiatry, 1973, 1,155-159. Masters, J. C. Effects of social comparison upon sub- sequent self-reinforcement behavior in children. Journal of Personality and Social Psychology, I968, IQ, 391-401. Masters, J. C. Social comparison, self-reinforcement, and the value of a reinforcer. Child Development, 1969, 55, 1027-1038. Masters, J. C. Effects of social comparison upon chil- dren's self-reinforcement and altruism toward competitors and friends. Developmental Psy- chology, 1971, 5, 64-72. Masters, J. C. Effects of success, failure, and reward outcome upon contingent and noncontingent self- reinforcement. Develppmental Psychology, 1972, 1, 110-118. Masters, J. C. Effects of age and social comparison upon children's noncontingent self-reinforcement and the value of a reinforcer. Child Development, 1973, 51, 111-116. Masters, J. C., & Christy, M. D. Achievement standards for contingent self-reinforcement: Effects of task length and task difficulty. Child Develop- ment, 1974, 55, 6-13. McFall, R. M. Effects of self-monitoring on normal smoking behavior. Journal of Consulting and Clinical Psychology, 1970, 55, 135-142. McFall, R. M., & Hammen, C. L. Motivation, structure, and self-monitoring: Role of nonspecific factors in smoking reduction. Journal of Consulting and Clinical Psychology, 197i, 51, 80-861 McKeachie, W. J. The decline and fall of the laws of learning. Educational Researcher, 1974, 5, 7-11. McKenzie, T., & Rushall, B. Effects of self-recording on attendance and performance in a competitive swimming training environment. Journal of Applied Behavior Analysis, 1974, 1, 199-206. 200 McLaughlin, T. F. Self-control in the classroom. Review of Educational Research, 1976, 55, 631-663. McReynolds, W. T., & Church, A. Self-control, study skills development and counseling approaches to the improvement of study behavior. Behavior Research and Therapy, 1973, 55, 233-235. Meichenbaum, D. H. Cognitive factors in behavior modifi- cation: Modifying what clients say to themselves. In C. M. Franks & G. T. Wilson (Eds.), Annual Review of Behavior Therapy: Theory and Practice. New York, N.Y.: Brunner/Mazel, 1973. Meichenbaum, D. H., & Goodman, J. Training impulsive children to talk to themselves: A means for developing self-control. Journal of Abnormal Psychology, 1971, 11, 115-126. Mendelsohn, M. B. Nonautonomous free man. Behaviorism, 1975, 5, 117-119. Miller, A. W., & Clark, N. Self-reinforcement established for a descrimination task. Journal of Genetic Psychology, 1970, 117, 1-6. Miller, A. W., & Gimpl, M. Operant conditioning and self- control of smoking and studying. Journal of Genetic Psychology, 1971, 119, 181:186. Mischel, W., & Liebert, R. M. Effects of discrepancies between observed and imposed reward criteria on their acquisition and transmission. Journal of Personality and Social Psychology, 1956, 5, 45-53. . Mischel, W., & Liebert, R. M. The role of power in the adoption of self-reward patterns. Child Develop- ment, 1967, 55, 673-683. Montgomery, G. T., & Parton, D. H. Reinforcing effect of self-reward. Journal of Experimental Psy- chology, 1970, 55, 275-276. Moser, A. J. Covert punishment of hallucinatory behavior in a psychotic male. Journal of Behavior Therapy and Experimental Psychiatry, 1974, 5, 297-299. Myers, C. A. Punishment: Problems in definition. 1975. (ERIC Document Reproduction Service No. ED 105 569) 201 Nelson, C. M., & McReynolds, T. Self-recording and control behavior. A reply to Simkins. Behavior Orne, M. T. From the subject's point of View, when is behavior private and when is it public: Problems of inference. Journal of Consulting and Clinical Psychology, 1970, 55, 143-147. Reschly, D. J. Consistency of self-reinforcement rates over different tasks; and sex, task success, and ability as determinants of rates of self-reinforce- ment. Psychological Record, 1973, 55, 237-242. Reschly, D. J., & Mittman, A. The relationship of self- esteem status and task ambiguity to the self- reinforcement behavior of children. Developmental Psychology, 1973, 5, 16-19. Richards, C. S. Behavior modification of studying through study skills advice. Journal of Counseling Psy- chology, 1975, 55, 431-436. Richards, C. S., McReynolds, W. T., Holt, 5., & Sexton, T. Effects of information feedback and self-adminis- tered consequences on self-monitoring study behavior. Journal of Counseling Psychology, 1976, 55, 3I6-321. Robertshaw, C. 8., Kelly, T. J., & Hiebert, H. D. The use of self-monitoring and contingency management techniques to increase verbal responding of an emotionally disturbed adolescent. SALT: School Applications of Learning Theory, 1974, 1, 31-357 Robinson, F. P. Effective study (4th ed.). New York: Harper & Row, 1970. Romanczyk, R. G. Self-monitoring in the treatment of obesity: Parameters of reactivity. Behavior Romanczyk, R. G., Kent, R. N., Diament, C., & O'Leary, K. D. Measuring the reliability of observational data: A reactive process. Journal of Applied Behavior Analyses, 1973, 5, I75-184. Romancyzk, R. G., Tracey, D. A., Wilson, G. T., & Thorpe, G. A. Behavioral techniques in the treatment of obesity: A comparative analysis. Behavior Research and Therapy, 1973, 55, 629-640. 202 Rozensky, R. H. The effect of timing of self-monitoring behavior on reducing cigarette consumption. Journal of Behavior Therspy and Experimental Psychiatry, I974, 5, 30I-303. Rozensky, R. H., & Bellack, A. S. Individual differences in self-reinforcement styles and performance in self- and therapist-controlled weight reduction programs. Behavior Research and Therapy, 1976, 55, 357-364. Rutner, I. T., & Bugle, C. An experimental procedure for the modification of psychotic behavior. Journal of Consulting and Clinical Psyshology, 1969, 55, 651-653. Simkins, L. A rejoinder to Nelson and McReynolds on the self-recording of behavior. Behavior Therapy, 1971, 5, 598-601. (a) Simkins, L. The reliability of self-recorded behaviors. Behavior Therapy, 1971, 5, 83-87. (b) Skinner, B. F. Science and human behavior. New York: MacMillan, 1953. Skinner, B. F. Beyond freedom and dignity. New York: Bantum Books, 1971. Speidel, G. E. Motivating effect of contingent self- reward. Journal of Esperimental Psychology, 1974, 102, 528-550. Stollak, G. E. Weight loss obtained under different experimental procedures. Psychotherapy: Theory Research and Practice, 1967, 5, 61-64. Stuart, R. B. Behavioral control of overeating. Behavior Research and Therapy, 1967, 5, 357-365. Stuart, R. B. Situational versus self-control. In R. D. Rubin, H. Fensterheim, J. D. Henderson, & L. P. Ullmann (Eds.), Advances in behavior therapy. New York: Academic Press, 1972. Stuart, R. B., & Davis, B. Slim chance in a fat world: 5ehavioral control of obesity. Champaign, III}: Research Press, 1972. 203 Thomas, E. J., Abrams, K. S., & Johnson, J. B. Self- monitoring and reciprocal inhibition in the modification of multiple tics of Gilles De La Tourette's syndrome. Journal of Behavior Therapy and Experimental Psychiatry, I97I, 2, 159-171. ‘ Thoreson, C. E., & Mahoney, M. J. Behavioral self-control. New York: Holt, Rinehart, and Winston, 1974. Thoreson, C. E., & Wilbur, C. S. Some encouraging thoughts about self-reinforcement. Journal of Applied Behavior Analysis, 1976, 5, SIB-550. Todd, F. J. Coverant control of self-evaluative responses in the treatment of depression: A new use for an old principle. Behavior Therapy, 1972, 5, 91-94. Turkewitz, H., O'Leary, K. D., & Ironsmith, M. Generali- zation and maintenance of appropriate behavior through self-control. Journal of Consulting and Clinical Psychology, 1975, 55, 577-583. Tyler, V. 0., & Straughan, J. H. Coverant control and breath holding as techniques for the treatment of obesity. Psychological Record, 1970, 20, 473-478. "' Wade, T. C. Relative effects on performance and moti- vation of self-monitoring correct and incorrect responses. Journal of Experimental Psychology, 1974, 555, 245-245. Wasserman, H., Brown, D., & Reschly, D. Application of self-management procedures for the modification of academic and classroom behaviors of two hyper- active children. SALT: School Application of Learning Theory, 1974, Z, I7-24. Weingartner, A. H. Self-administered aversive stimu- lation with hallucinating hospitalized schizo- phrenics. Journal of Consulting and Clinical Psychology, 1971, 55, 422-429. Whitman, T., & Dussault, P. Self-control through the use of a token economy. Journal of Behavior Therapy and Experimental PsycHiatsy, 1976, Z, 16I-I66. Van Zoost, B. L., & Jackson, B. T. Effects of self- monitoring and self-administered reinforcement on study behaviors. Journal of Educational Research, 1974, 55, 2I5-518. 204 General References Aronfreed, J. The origin of self-criticism. Psychologi- cal Review, 1964, 55, 193-217. Bandura, A. Self-reinforcement processes. In M. J. Mahoney & C. E. Thoreson, Self-control: Power to the person. Monterey, CaI}: Brooks7Cole, 1974. Bandura, A., Mahoney, M. J., & Dirks, S. J. Discrimi- native activation and maintenance of contingent self-reinforcement. Behavior Research and Therapy, 1976, 53, 1-6. Bandura, A., & Walters, R. H. Social learning and per- sonality development. New York: Holt, RineHart and’Winston, 1963. Barber, T. X., DiCara, L. V., Kamiya, J., Miller, N. E., Shapiro, D., & Stoyva, J. (Eds.), Biofeedback and self-control. Chicago: Aldine-Atherton, 1970. Barber, T. X., DiCara, L. V. Kamiya, J. Miller, N. E., Shapiro, D., & Stoyva, J. Biofeedback and self- control. Chicago: Aldine-Atherton, 1975-76. Bass, B. A., Ollendick, T. H., & Vuchinich, R. E. Study habits as a factor in the locus of control- academic achievement relationship. Psychological Reports, 1974, 55, 906. Bernstein, D. A. Modification of smoking behavior: An evaluative review. Psychological Bulletin, 1969, 15, 418-440. Blackwood, R. O. The operant conditioning of verbally mediated self-control in the classroom. Journal of School Psychology, 1970, 5, 251-258. Bowersock, R. B. Contingency self-management in counsel- ing. Journal of Counseling Psychology, 1974, 55, 265-268. 205 Brown, S. D. Self-control skills training. Professional Psychology, 1975, 5, 319-330. Budd, W. C. Behavior modification: The scientific way to self-control. Roslyn Heights, N.Y.: Libra, 1973. Cautela, J. R. Behavior therapy and self-control. In C. M. Franks (Ed.), Behavior therapy: Appraisal and status. New York: McGraw-Hill, 1969. Cautela, J. R. Covert negative reinforcement. Journal of Behavior Therapy and Experimental Psychiatry, I970, 5, 273-278. Champlin, S. M., & Karoly, P. Rate of contract negoti- ation in self-management of study time: A pre- liminary investigation. Psychological Reports, 1975, 51, 724-726. Chapman, R. F., Smith, J. W., & Layden, T. A. Elimi- nation of cigarette smoking by punishment and self-management training. Behavior Research and Therapy, 1971, 5, 255-264. Ferster, C. B. Classification of behavioral pathology. In L. Krasner & L. P. Ullmann (Eds.), Research in behavior modification. New York: Holt, Rinehart, & Winston, 1955. Ferster, C., Nurnberger, J., & Levitt, D. The control of eating. Journal of Mathetics, 1962, 5, 81-109. Foster, C. Developing self-control. Kalamazoo, Mich.: Behaviordelia, 1974. Goldiamond, I. A diary of self-modification. Psychology Today, 1973, 1, 95-102. Goldfried, M. R., & Merbaum, M. How to control yourself. Psychology Today, 1973, 5, 102-104. Gray, F., Graubard, P. S., & Rosenberg, H. "Little brother is changing you.” Psychology Today, 1974, 5, 42-46. Homme, L., Csanyi, A. P., Gonzales, M. A., & Recks, J. R. How to use contingency contracting in the class- room. Champaign, 111.: Research Press, 1969. Jones, Kamiya, Kanfer, Kanfer, Kanfer, Kanfer, Kanfer, Kanfer, Kanfer, Kanfer, 206 G. B. Improving study behaviors. In J. D. Krum- boltz & C. E. Thoreson (Eds.), Behavioral coun- selin . New York: Holt, Rinehart, & Winston, I969. J., Barber, T. X., DiCara, L. V., Miller, N. E., Shapiro, D., & Stoyva, J. (Eds.), Biofeedback and self-control. Chicago: Aldine-Atherton, 1971. F. H. Self-regulation: Research, issues, and speculations. In C. Neuringer & J. L. Michael (Eds.), Behavior modification in clinical psy- cholog . New York: Appleton-Century-Crofts, I970. F. H. Maintenance of behavior by self-generated stimuli and reinforcement. In A. Jacobs & L. Sachs (Eds.), The psychology of private events: Perspectives on covert response systems. New York: Academic Press, I971. F. H., & Goldfoot, D. A. Self-control and tol- erance of noxious stimulation. Psychological Reports, 1966, 55, 79-85. F. H., & Karoly, P. Self-regulation and its clinical application: Some additional consider- ations. In R. C. Johnson, P. R. Dokecki, & O. H. Mowrer (Eds.), Conscience, contract, and social reality. New York} Holt, Rinehart & Winston, I972. F. H., & Phillips, J. S. Behavior therapy: A panacea for all ills or a passing fancy? Archives of General Psychiatsy, 1966, 55, 114-128. F. H., & Phillips, J. S. Learning foundations of behavior therapy. New York: Wiley, 1970. F. H., & Phillips, J. 8. Contract psychology: An operational approach to the problem of con- science and self-control. In R. C. Johnson, P. R. Dokecki, & O. H. Mowrer (Eds.), Conscience, contract, and social reality. New York: Holt, Rinehart, & Winston, 1971. F. H., & Seidner, M. L. Self-control: Factors enhancing tolerance of noxious stimulation. Journal of Personality and Social Psychology, I973, 55, 381-389. 207 Klein, R. D., Hapkiewicz, W. G., & Roden, A. H. (Eds.), Behavior modification in educational settings. SpringfieId, 111.: Charles C. Thomas, 1973. Kolb, D. A., Winter, S. K., & Berlew, D. E. Self-directed change: Two studies. Journal of Applied Behavioral Science, 1968, 5, 453-47I. Kozma, A., & Easterbrook, P. Effects of baseline se1f- reinforcement behavior and training level on post- training self-reinforcement behavior. Journal of Experimental Psychology, 1974, 555, 256-259. Kurtz, P. D., & Neisworth, J. T. Self-control possibili- ties for exceptional children. Exceptional Chil- dren, 1976, 55, 212-217. Layne, C. C., Rickard, H. C., Jones, M. T., & Lyman, R. D. Accuracy of self-monitoring on a variable ratio schedule of observer verification. Behavior Therapy, 1976, 5, 481-488. Locke, E. A., Cartedge, N., & Koeppel, J. Motivational effects of knowledge of results: A goal-setting phenomenon? Psychological Bulletin, 1968, 55, 474-485. Lopatto, D., & Williams, J. L. Self-control: A critical review and an alternative interpretation. Psycho- logical Record, 1976, 55, 3-12. Mahoney, M. J. Toward an experimental analysis of coverant control. Behavior Therapy, 1970, 5, 510-521. Mahoney, M. J. Research issues in self-management. In M. J. Mahoney & C. E. Thoreson, Self-control: Power to the person. Monterey, Cal.:firooks/ CoIe, I974. Mahoney, M. J., & Bandura, A. Self-reinforcement in pigeons. Learning and Motivation, 1972, 5, 293-303. Mahoney, M. J., & Mahoney, K. Self-control techniques with the mentally retarded. Exceptional Children, 1976, 55, 338-339. Mahoney, M. J., & Thoreson, C. E. Self-control: Power to the person. Monterey, Cal.: Brooks/Cole, I974. 208 Marston, A. R. Variables affecting incidence of self- reinforcement. Psychological Reports, 1964, 55, 879-884. Marston, A. R., & Feldman, S. E. Toward the use of self- control in behavior modification. Journal of Consulting and Clinical Psychology, 1972, 39, 559-433. -—- McCausland, D. F., & Stewart, N. E. Academic aptitude, study skills, and attitudes and college GPA. Journal of Educational Research, 1974, 51, 354-357. McLaughlin, T. F. Self-control procedures in the manage- ment of classroom behavior problems. Education, 1976, 55, 379-383. Miller, L. K. A note on the control of study behavior. Journal of Experimental Child Psychology, 1964, 5, 198-110. Miller, A., & Gimpl, M. P. Operant verbal self-control of studying. Psychological Reports, 1972, 55, 495-498. Nolan, J. D. Self-control procedures in the modification of smoking behavior. Journal of Consulting and Clinical Psychology, 1968, 55, 92-93. O'Leary, K. D. The effects of self-instruction on immoral behavior. Journal of Experimental Child Psychology, 1968, 5,4597-301. Pawlicki, R. E., & Morey, T. M. A low cost instrument for thinning self-directed schedules of reinforcement. Behavior Therapy, 1976, 1, 120-122. Phillips, R. E., Johnson, G. D., & Geyer, A. Self- administered systematic desensitization. Behavior Research and Therapy, 1972, 55, 93-96. Premack, D. Mechanism of self-control. In W. Hunt (Ed.), Learning and mechanisms of control in smoking. Chicago: Aldine, 1970. Premack, D., & Anglin, B. On the possibilities of self- control in man and animals. Journal of Abnormal Psychology, 1973, 55, 137-151. 209 Richards, C. S., Perri, M. G., Gortney, C. Increasing the maintenance of self-control treatments through faded counselor contact and high information feedback. Journal of CounselingyPsychology, 1976, 55, 495-406. Roberts, A. H. Self-control procedures in modification of smoking behavior: Replication. Psychological Reports, 1969, 55, 675-676. Rodin, M. A further look at social desirability. Psy- chonomic Science, 1972, 55, 103—105. Rubin, R. D., Merbaum, M., & Fried, R. Self-imposed punishment versus desensitization. In R. D. Rubin, H. Fensterheim, A. A. Lazarus, & C. M. Franks (Eds.), Advances in behavior therapy. (Vol. 3). New York: Academic Press, 1971. Schwartz, G. E. Biofeedback, self-regulation and the patterning of physiological processes. American Scientist, 1975, 55, 314-324. Seidner, M. L. Behavior change contract: Prior infor- mation about study habits, treatment and state- ments of intention as related to initial effort in treatment. Dissertation Abstracts Inter- national, 1974, 55, 3509b. Seligman, M. E. Learned helplessness. Annual Review of Medicine, 1972, 55, 407-412. Shapiro, D., Barber, T. X., DiCara, L. V., Kamiya, J., Miller, N. E., & Stoyva, J. (Eds.), Biofeedback and self-control. Chicago: Aldine-Atherton, 1973. Sirota, A. D., & Mahoney, M. J. Relaxing on cue: The self-regulation of asthma. Journal of Behavior Therapy and Experimental PsyEHiatry, 1974, 5, 6—5- 66'. Staub, E., Tursky, B., & Schwartz, G. E. Self-control and predictability: Their effects on reactions to aversive stimulation. Journal of Personality and Social Psychology, 197IT 55, 157-I65: Stoyva, J., Barber, T. X., DiCara, L. V., Kamiya, J., Miller, N. E., & Shapiro, D. (Eds.), Biofeedback and self-control. Chicago: Aldine-AEHerton, 1972. Stuart, 210 R. B. Trick or treatment: How and when psycho- therapy faiIs. Cfiampaign, IIII: Research Press, I970} Theophanous, A. C. In defense of self-determination: Thomas, Watson, Yates, A critique of B. F. Skinner. Behaviorism, 1975, 5, 97-115. D. R. Preliminary findings on self-monitoring for modifying teaching behaviors. In B. H. Hopkins (Ed.), A new direction for education: Behavior and anaIysis. (Vol. 1). University of Kansas, 1971. D. L., & Tharp, R. G. Self-directed behavior: Self-modification for personal adjustment. Belmont, Cal.: Brooks/Cole, 1972. A. J. Behavior therapy. New York: John Wiley & Sons, 1975. "11111111111151“