R01 - Festinger 1959 - Cognitive Consequences of Forced Compliance

7/29/2019 R01 - Festinger 1959 - Cognitive Consequences of Forced Compliance

1/5

LEON FESTINGER AND JAMES M. CARLSMITH1Stanford UniversityWHA T happens to a person's privateopinion if he is forced to do or saysomething contrary to that opin-ion? Only recently has there been any experi-mental work related to this question. Two stud-

ies reported by Janis and King (1954; 1956)clearly showed that, at least under some condi-tions, the private opinion changes so as to bringit into closer correspondence with the overtbehavior the person was forced to perform.Specifically, they showed that if a person isforced to improvise a speech supporting a pointof view with which he disagrees, his privateopinion moves toward the position advocatedin the speech. The observed opinion change isgreater than for persons who only hear thespeech or for persons who read a preparedspeech with emphasis solely on elocution andmanner of delivery. The authors of these twostudies explain their results mainly in terms ofmental rehearsal and thinking up new argu-ments. Inthisway, theypropose, the person whois forced to improvise a speech convinceshimself. They present some evidence, which isnot altogether conclusive, in support of thisexplanation. We will have more to say con-cerning this explanation in discussing theresults of our experiment.Kelman (1953) tried to pursue the matterfurther. He reasoned that if the person isinduced to make an overt statement contraryto his private opinion by the offer of somereward, then the greater the reward offered,the greater should be the subsequent opinionchange. His data, however, did not supportthis idea. He found, rather, that a large rewardproduced less subsequent opinion change thandid a smaller reward. Actually, this finding byKelman is consistent with the theory we willoutline below but, for a number of reasons, is

1 The experiment reported here was done as part ofa program of research supported by a grant from theNational Science Foundation to the senior author. Wewish to thank Leonard Hommel , Judson Mills, andRobert Terwilliger for their help in designing andcarrying out the experiment. We would also like toacknowledge the help of Ruth Smith and Marilyn M.Miller.

not conclusive.One of the major weaknesses ofthe data is that not all subjects in the experi-ment made an overt statement contrary totheir private opinion in order to obtain theoffered reward. What is more, as one mightexpect, the percentage of subjects who com-plied increased as the size of the offered rewardincreased. Thus, with self-selection of whodid and who did not make the required overtstatement and with varying percentages ofsubjects in the different conditions who didmake the required statement, no interpreta-tion of the data can be unequivocal.

Recently, Festinger (1957) proposed a theoryconcerning cognitive dissonance from whichcome a number of derivations about opinionchange following forced compliance. Sincethese derivations are stated in detail by Fest-inger (1957, Ch. 4), we will here give only abrief outline of the reasoning.Let us consider a person who privately holdsopinion "X" but has, as a result of pressurebrought to bear on him, publicly stated thathe believes "not X."1. This person has two cognitions which,psychologically, do not fit together: one ofthese is the knowledge that he believes "X,"the other the knowledge that he has publiclystated that he believes "not X." If no factorsother than hisprivate opinionare considered, itwould follow, at least in our culture, that ifhe believes "X" he would publicly state "X."Hence, his cognition of his private belief isdissonant with his cognition concerning hisactual public statement.2. Similarly, the knowledge that he has said"not X" is consonant with (does fit togetherwith) those cognitive elements correspondingto the reasons, pressures, promises of rewardsand/or threats of punishment which inducedhim to say "not X."3. In evaluating the total magnitude ofdissonance, one must take account of bothdissonances and consonances. Let us think ofthe sum of all the dissonances involving someparticular cognition as "D" and the sum ofall the consonances as "C." Then we might

203


2/5

204 LEON FESTINGER A ND JAMES M. CARLSMITHthink of the total magnitude of dissonance asbeing a function of "D" divided by "D" plus"C."Let us then see what can be said about thetotal magnitude of dissonance in a personcreated by the knowledge that he said "not X"and really believes "X." With everything elseheld constant, this total magnitude of disso-nance would decrease as the number andimportance of the pressures which inducedhim to say "not X" increased.Thus, if the overt behavior was broughtabout by, say, offers of reward or threats ofpunishment, the magnitude of dissonance ismaximal if these promised rewards or threat-ened punishments were just barely sufficientto induce the person to say "not X." Fromthis point on, as the promised rewards orthreatened punishment become larger, themagnitude of dissonance becomes smaller.4. One way in which the dissonance can bereduced is for the person to changehis privateopinion so as to bring it into correspondencewith what he has said. One would conse-quently expect to observe such opinion changeafter a personhas been forced or induced to saysomething contrary to his private opinion.Furthermore, since the pressure to reducedissonance will be a function of the magnitudeof the dissonance, the observed opinion changeshould be greatest when the pressure used toelicit the overt behavior is just sufficient todoit.The present experiment was designed totest this derivation under controlled, labora-tory conditions. In the experiment we variedthe amount of reward used to force persons tomake a statement contrary to their privateviews. The prediction [from 3 and 4 above] isthat the larger the reward given to the subject,the smaller will be the subsequent opinionchange. PROCEDURE

Seventy-one male students in the introduc-tory psychology course at Stanford Universitywere used in the experiment. In this course,students are required to spend a certain num-ber of hours as subjects (5s) in experiments.They choose among the available experimentsbysigning theirnames on a sheet posted on thebulletin board which states the nature of theexperiment. The present experiment waslisted

as a two-hour experiment dealing with "Meas-ures of Performance."During the first week of the course,when therequirement of serving in experiments wasannounced and explained to the students, theinstructor also told them about a study thatthe psychology department was conducting.He explained that, since they were required toserve in experiments, the department was con-ducting a study to evaluate these experimentsin order to be able to improve them in thefuture. They were told that a sample ofstudents would be interviewed after havingserved as 5s. They were urged to cooperate inthese interviews by being completely frankand honest. The importance of this announce-ment will become clear shortly. It enabled us tomeasure the opinions of our 5s in a context notdirectly connected with our experiment and inwhich we could reasonably expect frank andhonest expressions of opinion.When the 5 arrived for the experiment on"Measures of Performance" he had to wait fora few minutes in the secretary's office. Theexperimenter (E) then came in, introducedhimself to the 5 and, together, they walkedinto the laboratory room where the E said:

This experiment usually takes a little over an hourbut, of course, we had to schedule it for two hours.Since we have that extra time, the introductory psy-chology people asked if they could interview some ofour subjects. [Offhand and conversationally.] D id theyannounce that in class? I gather that they're interview-ing some people who have been in experiments. I don'tknow much about it. Anyhow, they may want to inter-view you when you're through here.With no further introduction or explanationthe 5 was shown the first task, which involvedputting 12 spools onto a tray, emptying thetray, refilling it with spools, and so on. He wastold to use one hand and to work at his ownspeed. He did this for one-half hour. The Ethen removed the tray and spools and placedin front of the 5 a board containing 48- squarepegs. His task was to turn each peg a quarterturn clockwise, then another quarter turn, andso on. He was told again to use one hand andto work at his own speed. The 5 worked at thistask for another half hour.While the 5 wasworkingon these tasks, the

E sat, with a stop watch in his hand, busilymaking notations on a sheet of paper. He didso in order to make it convincing hat this was

COGNITIVE C O N S E Q U E N C E S or F O R C E D C O M P L IA N C E 205what the E was interested in and that thesetasks, and how the 5 worked on them, was thetotal experiment. From our point of view theexperiment had hardly started. The hour whichthe 5 spent working on the repetitive, monot-onous tasks wa s intended to provide, for each5 uniformly, an experience about which hewould have a somewhat negative opinion.A fter the half hour on the second task wasover, the E conspicuouslyset the stop watchback to zero, put it away, pushed his chairback, lit a cigarette, and said:

O.K. Well, that's all we have in the experimentitself. I'd like to explain what this has been all aboutso you'll have some idea of why you were doing this.[E pauses.] Well, the way the experiment is set up is this.There are actually two groups in the experiment. Inone, the group you were in, we bring the subject inand give him essentially no introduction to the experi-ment. That is, all we tell him is what he needs to knowin order to do the tasks, and he has no idea of what theexperiment is all about, or what it's going to be like,or anything like that. But in the other group, we havea student that we've hired that works for us regularly,and what I do is take him into the next room where thesubject is waitingthe same room you were waiting hibeforeand I introduce him as if he had just finishedbeing a subject in the experiment. That is, I say: "Thisis so-and-so, who's just finished the experiment, andI've asked him to tell you a little of what it's aboutbefore you start." The fellow who works for us then,in conversation with the next subject, makes thesepoints: [The E then produced a sheet headed "ForGroup B" which had written on it: It was very enjoy-able, I had a lot of fun, I enjoyed myself, it was veryinteresting, it was intriguing, it was exciting. The Eshowed this to the 5 and then proceeded with his falseexplanation of the purpose of the experiment.] Now, ofcourse, we have this student do this, because if theexperimenter does it, it doesn't look as realistic, andwhat we're interested in doing is comparing how thesetwo groups do on the experimentth e one with thisprevious expectation about the experiment, and theother, like yourself, with essentially none.

Up to this point the procedure was identicalfor 5s in all conditions. From this point on theydiverged somewhat. Three conditions wererun, Control, One Dollar, and Twenty Dollars,as follows:Cont r o l Cond i t i on

The continued:Is that fairly clear? [Pause.] Look, that fellow [looksat watch] I was telling you about from the introductorypsychology class said he would get here a couple ofminutes from now. Would you mind waiting to see ifhe wants to talk to you? Fine. Why don't we go into

the other room to wait? [The E left the 5 in the secre-tary's office for four minutes. He then returned andsaid:] O.K. Let's check and see if he does want to talkto you.One and Twe n ty D o l l a r Cond i t i on s

The E continued:Is that fairly clear how it is set up and what we'retrying to do? [Pause.] Now, I also have a sort of strange

thing to ask you. The thing is this. [Long pause, someconfusion and uncertainty in the following, with a de-gree of embarrassment on the part of the E. Themanner of the E contrasted strongly with the precedingunhesitant and assured false explanation of the experi-ment. The point was o make it seem to the S that thiswas the first time the E had done this and that he feltunsure of himself.] The fellow who normally does thisfor us couldn't do it todayhe just phoned in, andsomething or other came up for himso we've beenlooking around for someone that wecould hire to do itfor us. You see, we've got another subject waiting[looks at watch] who is supposed to be in that othercondition. Now Professor , who is in charge ofthis experiment, suggested that perhaps we could takea chance on your doing it for us. I'll tell you what wehad in mind : the thing is, if you could do it for us now,then of course you would know how to do it, and ifsomething like this should ever come up again, that is,the regular fellow couldn't make it, and we had a sub-ject scheduled, it would be very reassuring to us toknow that we had somebody else we could call on whoknew how to do it. So, if you would be willing to dothis for us, we'd like to hire you to do it now and thenbe on call in the future, if something like this shouldever happen again. We can pay you a dollar (twentydollars) for doing this for us, that is, for doing it nowand then being on call. Do you th ink you could do thatfor us?

If the 5 hesitated, the E said things like, "Itwill only take a few minutes," "The regularperson is pretty reliable; this is the first timehe has missed," or "If we needed you wecouldphone you a day or two in advance; if youcouldn't make it, of course, we wouldn't expectyou to come." Af te r the 5 agreed to do it, theE gave him the previously mentioned sheetof paper headed "For Group B" and askedhim to read it through again. The E then paidthe 5 onedollar (twenty dollars), made out ahand-written receipt form, and asked the 5to sign it. He then said:

O.K., the way we'll do it is this. As I said, the nextsubject should be here by now. I think the next one isa girl. I'll take you into the next room and introduceyou to her, saying that you've just finished the experi-ment and that we've asked you to tell her a littleabout it. And what wewant you to do is just sit downand get into a conversation with her and try to get


3/5

206 LEON FESTINGER A ND JAMES M. C A R L S M I T Hacross the points on that sheet of paper. I'll leave youalone and come back after a couple of minutes. O.K.?

The E then took the S into the secretary'soffice where he had previously waited andwhere the next S was waiting. (The secretaryhad left the office.) He introduced the girl andthe S to one another saying that the 5 hadjust finished the experiment and would tell hersomething about it. He then left saying hewould return in a couple of minutes. T he girl,an undergraduate hired for this role, said littleuntil the 5 made some positive remarks aboutthe experiment and then said that she wassurprised because a friend of hers had takenthe experiment the week before and had toldher that it was boring and that she ought totry to get out of it. Most 5s responded bysaying something like "Oh, no, it's really veryinteresting. I'm sure you'll enjoy it." The girl,after this listened quietly, accepting andagreeing to everything the S told her. Thediscussion between the S and the girl was re-corded on a hidden tape recorder.Afte r two minutes the E returned, asked thegirl to go into the experimental room, thankedthe S for talking to the girl, wrote down hisphone number to continue the fiction that wemight call on him again in the future and thensaid: "Look, could we check and see if thatfellow from introductory psychology wants totalk to you?"From this point on, the procedure for allthree conditions wa s once more identical. A sthe E and the 5 started to walk to the officewhere the interviewer was, the E said: "Thanksvery much fo r working on those tasks for us.I hope you did enjoy it. Most of our subjectstell us afterward that they found it quiteinteresting. You get a chance to see how youreact to the tasks and so forth." This shortpersuasive communication was made in allconditions in exactly the same way. The reasonfor doing it, theoretically, was to make iteasier fo r anyone wh o wanted to persuade him-self that the tasks had been, indeed, enjoyable.When they arrived at the interviewer'soffice, the E asked the interviewer whether ornot he wanted to talk to the S. The interviewersaid yes, the E shook hands with the S, saidgood-bye, and left. T he interviewer, of course,wa s always kept in complete ignorance ofwhich condition the 5 was in. The interview

consisted of four questions, on each of whichthe 5 was first encouraged to talk about thematter and was then asked to rate his opinionor reaction on an 11-point scale. The questionsare as follows:1. Were the tasks interesting and enjoyable? In whatway? In what wa y were they not? Would yo u rate ho wyo u feel about them on a scale from S to +5 where5 means they were extremely dull and boring, +5means they were extremely interesting and enjoyable,and zero means they were neutral, neither interestingno r uninteresting.2. Did the experiment give you an opportunity tolearn about your ow n ability to perform these tasks?In what way? In what way not? Would you rate howyou feel about this on a scale from 0 to 10 where 0means you learned nothing and 10means you learned agreat deal.3. From what you know about the experiment andthe tasks involved in it, would you say the experimentwa s measuring anything important? That is, do youthink the results may have scientific value? In whatway? In what wa y not? Would yo u rate your opinionon this matter on a scale from 0 to 10 where 0 meansthe results have no scientific value or importance and10 means they have a great deal of value and im-portance.4. Would yo u have any desire to participate inanother similar experiment? Why? Wh y not? Would

yo u rate your desire to participate in a similar experi-ment again on a scale from 5 o +5 , where 5 meansyou would definitely dislike to participate, +5 meansyo u would definitely like to participate, and 0 meansyo u have no particular feeling about it one way or theother.A s may be seen, the questions varied in howdirectly relevant they were to what the 5 hadtold the girl. This point willbe discussed further

in connection with the results.A t the close of the interview the S wa s askedwhat he thought the experiment wasabout and,following this, wa s asked directly whether ornot he was suspicious of anything and, if so,what he was suspicious of . When the interviewwa s over, the interviewer brought the S backto the experimental room where the E waswaiting together with the girl who had posedas the waiting S. (In the control condition,ofcourse, the girl was not there.) The true pur-pose of the experiment wa s then explained tothe 5 in detail, and the reasons for each of thevarious steps in the experiment were explainedcarefully in relation to the true purpose. Allexperimental 5s in both OneDollar and Twen-ty Dollar conditions were asked, after thisexplanation, to return the money they had

been given. All 5s, without exception, werequite willingto return the money.The data from 11 of the 71 5s in the experi-ment had to be discarded for the followingreasons:1. Five 5s (three in the One Dollar and two in theTwenty Dollar condition) indicated in the interviewthat they were suspicious about having been paid totell the girl the experiment was fun and suspected thatthat was the real purpose of the experiment.2. Two 5s (both in the One Dollar condition) toldthe girl that they had been hired, that the experimentwa s really boring but they were supposed to say itwas fun.3. Three 5s (one in the One Dollar and two in theTwenty Dollar condition) refused to take the moneyand refused to be hired.4. One 5 (in the One Dollar condition), immediatelyafter having talked to the girl, demanded her phonenumber saying he would call her and explain things,and also told the E he wanted to wait until she wasfinished so he could tell her about it.These 11 5s were, of course, run through thetotal experiment anyhow and the experimentwa s explained to them afterwards. Their data,however, are not included in the analysis.

Su m m ar y of DesignThere remain, for analysis, 20 5s in each ofthe three conditions. Let us review thesebriefly: 1. Contro l condition. These 5s weretreated identically in all respects to the 5sin the experimental conditions, except thatthey were never asked to, and never did, tellthe waiting girl that the experimental taskswere enjoyable and lots of fun. 2. On e Dol larcondi t ion. These 5s were hired for one dollar totell a waiting 5 that tasks, which were reallyrather dull and boring, were interesting, en-joyable, and lots of fun. 3. Twenty Dol lar condi -t ion . These 5s were hired fo r twenty dollars todo the same thing.

RESULTST he major results of the experiment aresummarized in Table 1 which lists, separatelyfor each of the three experimental conditions,the average rating which the 5s gave at theend of each question on the interview. We willdiscuss each of the questions on the interviewseparately, because they were intended tomeasure different things. On e other point be -fore we proceed to examine the data. In all thecomparisons, the Control condition should be

207TABLE 1

A V E R A G E RATINGS O N INTERVIEW QUESTIONS P O R

Question on Interview

H ow enjoyable tasks were(rated from 5 to +5 )How much they learned(rated from 0 to 10)Scientific importance (ratedfrom 0 to 10)Participate in similar exp.(rated from 5 to +5)

Experimental ConditionControl(N -20)-.453.085.60.

-.62

OneDollar(N = 20)+1.35

2.806.45

+1.20

TwentyDollars(N -20)-.053.155.18'-.25

regarded as a baseline from which to evaluatethe results in the other two conditions. TheControl condition gives us, essentially, thereactions of 5s to the tasks and their opinionsabout the experiment as falsely explained tothem, without the experimental introductionof dissonance. T he data from the other condi-tions may be viewed, in a sense, as changesfrom this baseline.How Enjoyable the Task s Were

The average ratings on this question, pre-sented in the first row of figures in Table 1,are the results most important to the experi-ment. These results are the ones most directlyrelevant to the specific dissonance which wasexperimentally created. It will be recalled thatthe tasks were purposely arranged to berather boring and monotonous. And, indeed,in the Control condition the average ratingwas .45, somewhat on the negative side ofthe neutral point.In the other two conditions, however, the5s told someone that these tasks were interest-ing and enjoyable. The resulting dissonancecould, of course, most directly be reduced bypersuading themselves that the tasks were,indeed, interesting and enjoyable. In the OneDollar condition, since the magnitude ofdissonance was high, the pressure to reducethis dissonance would also be high. In thiscondition, the average rating was +1-35,considerably on the positive side and signifi-cantly different from the Control condition atthe .02 level2 (t = 2.48).

2 A ll statistical tests referred to in this paper aretwo-tailed.


4/5

208 LEON FESTINGER A ND JAMES M. CARLSMITHIn the Twenty Dollar condition, where lessdissonance was created experimentally becauseof the greater importance of the consonantrelations, there is correspondingly less evidenceof dissonance reduction. T he average rating inthis condition is only .05 , slightly and notsignificantly higher than the Control condition.The difference between the One Dollar andTwenty Dollar conditions is significant at the

.03 level (t = 2.22). In short, when an 5 wasinduced, by offer of reward, to say somethingcontrary to his private opinion, this privateopinion tended to change so as to correspondmore closely with what he had said. T he greaterthe reward offered (beyond what wa s necessaryto elicit the behavior) the smaller was theeffect.Desire to Part ic ipa te in a Similar Experiment

T he results from this question are shown inthe last row of Table 1. This question is lessdirectly related to the dissonance that wasexperimentally created for the Ss. Certainly,the more interesting and enjoyable they feltthe tasks were, the greater would be their de-sire to participate in a similar experiment. Butother factors would enter also. Hence, onewould expect the results on this question tobe very similar to the results on "how enjoy-able the tasks were" but weaker. Actually, theresult, as may be seen in the table, are inexactly the same direction, and the magnitudeof the mean differences is fully as large as onthe first question. The variability is greater,however, and the differences do not yield highlevels of statistical significance. T he differencebetween the One Dollar condition (+1.20)and the Control condition (.62) is significantat the .08 level (t = 1.78). The differencebetween the One Dollar condition and theTwenty Dollar condition ( 25) reaches onlythe .15level of significance (t = 1.46).The Scientific Importance of the Experiment

This question was included because therewa s a chance that differences might emerge.There are, after all, other ways in which theexperimentally created dissonance could bereduced. F or example, one way would be forthe S to magnify fo r himself the value of thereward he obtained. This, however, was un-

likely in this experiment because money wasused for the reward and it is undoubtedlydifficult to convince oneself that one dollar ismore than it really is. There is another pos-sible way, however. The 5s were given a verygood reason, in addition to being paid, forsaying what they did to the waiting girl. The5s were told it was necessary for the experi-ment. T he dissonance could, consequently, bereduced by magnifying the importance of thiscognition. The more scientifically importantthey considered the experiment to be, the lesswas the total magnitude of dissonance. It ispossible, then, that the results on this ques-tion, shown hi the third row of figures hi Table1,might reflect dissonance reduction.The results are weakly in line with what onewould expect if the dissonance were somewhatreduced in this manner. The One Dollar condi-tion is higher than the other two. T he differ-ence between the One and Twenty Dollarconditions reaches the .08 level of significanceon a two-tailed test (t = 1.79). T he differencebetween the One Dollar and Control conditionsis not impressive at all (t = 1.21). The resultthat the Twenty Dollar condition is actuallylower than the Control condition is un-doubtedly a matter of chance (t = 0.58).Ho w Much They Learned From the Experiment

The results on this question are shownin thesecond row of figures in Table 1. The questionwas included because, as far as we could see, ithad nothing to do with the dissonance thatwa s experimentally created and could not beused for dissonance reduction. One would thenexpect no differences at all among the threeconditions. We felt it was important to showthat the effect was not a completely generalone but was specific to the content of the dis-sonance which wa s created. A s can be readilyseen hi Table 1, there are only negligible differ-ences among conditions. T he highest t value forany of these differences is only 0.48.DISCUSSION OF A POSSIBLE ALTERNATIVEEXPLANATION

We mentioned in the introduction thatJams and King (1954; 1956) hi explaining then*findings, proposed an explanation in terms ofthe self-convincing effect of mental rehearsal

COGNITIVE CONSEQUENCES OP FORCED COMPLIANCE 209and thinkingup new arguments by the personwho had to improvise a speech. Kelman (1953),in the previously mentioned study, in at-tempting to explain the unexpected findingthat the personswho complied in the moderatereward condition changed then: opinion morethan in the high reward condition, also pro-posed the same kind of explanation. If theresults of our experiment are to be taken asstrong corroboration of the theory of cogni-tive dissonance, this possible alternativeexplanation must be dealt with.Specifically, as applied to our results, thisalternative explanation would maintain thatperhaps, for some reason, the 5s in the OneDollar condition worked harder at telling thewaiting girl that the tasks were fun and en-joyable. That is, in the One Dollar conditionthey may have rehearsed it more mentally,thought up more ways of saying it, may havesaid it more convincingly, and so on. Why thismight have been the case is, of course, notimmediately apparent. Onemight expect that,in the Twenty Dollar condition, having beenpaid more, they would try to do a better jobof it than in the One Dollar condition. Butnevertheless, the possibility exists that the 5shi the One Dollar condition may have impro-vised more.Because of the desirability of investigatingthis possible alternative explanation, werecorded on a tape recorder the conversationbetween each 5 and the girl. These recordingswere transcribed and then rated, by twoindependent raters, on five dimensions. Theratings were, of course done hi ignorance ofwhich condition each 5 was in. The relia-bilities of these ratings, that is, the correlationsbetween the two independent raters, rangedfrom .61 to .88, with an average reliability of

.71. The five ratings were:1. The content of what the 5 said beforethe girl made the remark that her friend toldher it was boring. The stronger the 5's positivestatements about the tasks, and the more wayshi which he said they were interesting andenjoyable, the higher the rating.2. The content of what the 5 said after thegirl made the above-mentioned remark. Thiswa s rated in the same way as for the contentbefore th e remark.

3. A similar rating of the over-all content ofwhat the 5 said.4. A rating of how persuasive and convincingthe 5 was in what he said and the way inwhichhe said it.5. A rating of the amount of time in the dis-cussion that the 5 spent discussing the tasksas opposed to going off into irrelevant things.T he mean ratings for the One Dollar andTwenty Dollar conditions, averaging the rat-ings of the two independent raters, are pre-sented in Table 2. It is clear from examing thetable that, in all cases, the Twenty Dollarcondition is slightly higher. T he differencesare small, however, and only on the rating of"amount of time" does the difference betweenthe two conditions even approach significance.We are certainly justified in concluding thatthe 5s in the One Dollar condition did notimprovise more nor act more convincingly.Hence, the alternative explanation discussedabove cannot account for the findings.S U M M A R Y

Recently, Festinger (1957) has proposed atheory concerning cognitive dissonance. Twoderivations from this theory are tested here.These,are:1. If a person is induced to do or say some-thing which is contrary to his private opinion,there will be a tendency for him to change hisopinion so as to bring it into correspondencewith what he has done or said.2. The larger the pressure used to elicit the

T A B L E 2A V E R A G E R A T I N G S O F DISCUSSION BETWEEN SUBJECTAN D GIRL

Conditionjuimension jtatea

Content before remark by girl(rated from 0 to 5)Content after remark by girl(rated from 0 to 5)Over-all content (rated from 0to 5)Persuasiveness and conviction(rated from 0 to 10)Time spent on topic (rated from0 to 10)

OneDollar2.261.631.894.796.74

TwentyDollars2.621.752.195.508.19

Valueo ft1.080.111.080.991.80


5/5

210 LEON FESXINGER A ND JAMES M. C A R L S M I T Hovert behavior (beyond the minimum neededto elicit it) the weaker will be the above-mentioned tendency.A laboratory experiment was designed totest these derivations. Subjects were subjectedto a boring experience and then paid to tellsomeone that the experience had been interest-ing and enjoyable. The am ount o f money paidthe subject wa s varied. T he private opinionsof the subjects concerning the experienceswere then determined.The results strongly corroborate the theorythat was tested.

REFERENCESFESTTNGEB, L. A theory of cognitive dissonance. Evans-ton, 111: Row Peterson, 1957.JANIS, I. L., & K ING, E. T . The influence of role-playingon opinion change. /. abnorm. s oc . Psychol ., 1954,49, 211-218.KELMAN, H . Attitude change as a function of responserestriction. Hum. Relat . , 1953, 6, 185-214.KING, 6. T., & JANIS, I. L. Com parison of the effective-

ness of improvised versus non-improvised role-playing in producing opinion changes. H u m .Relat., 1956, 9, 177-186.Received November 18, 1957.

R01 - Festinger 1959 - Cognitive Consequences of Forced Compliance

Documents