basking in reflected glory: three (football) field studiesschaller/528readings/cialdini1976.pdf ·...

Journal of Personality and Social Psychology1976, Vol. 34, No. 3, 366-375

Basking in Reflected Glory: Three (Football) Field Studies

Robert B. CialdiniArizona State University

Richard J. BordenPurdue University

Avril ThorneArizona State University

Marcus Randall Walker and Stephen FreemanOhio State UniversityLloyd Reynolds Sloan

University of Notre Dame

The tendency to "bask in reflected glory" (BIRG) by publicly announcingone's associations with successful others was investigated in three field experi-ments. All three studies showed this effect to occur even though the personstriving to bask in the glory of a successful source was not involved in thecause of the source's success. Experiment 1 demonstrated the BIRG phenome-non by showing a greater tendency for university students to wear school-identifying apparel after their school's football team had been victorious thannonvictorious. Experiments 2 and 3 replicated this effect by showing that stu-dents used the pronoun we more when describing a victory than a nonvictoryof their school's football team. A model was developed asserting that the BIRGresponse represents an attempt to enhance one's public image. Experiments 2and 3 indicated, in support of this assertion, that the tendency to proclaim aconnection with a positive source was strongest when one's public image wasthreatened.

It is a common and understandable tend-ency for people who have been successful insome positive way to make others aware oftheir connection with that accomplishment.However, there also appears to be a seem-ingly less rational but perhaps more inter-esting tendency for people to publicize aconnection with another person who has beensuccessful. This latter inclination might becalled the tendency to bask in reflected glory(BIRG). That is, people appear to feel thatthey can share in the glory of a successfulother with whom they are in some way asso-ciated; one manifestation of this feeling isthe public trumpeting of the association. Sucha phenomenon is not hard to understand whenthe one wishing to share in another's successhas been instrumental to that success. How-ever, the more intriguing form of the phe-nomenon occurs when the one who basks inthe glory of another has done nothing to

Requests for reprints should be sent to Robert B.Cialdini, Department of Psychology, Arizona StateUniversity, Tempe, Arizona 85281.

bring about the other's success. Here, asimple case of affiliation or membership issufficient to stimulate a public announcementof the critical connection.

There does seem to be abundant anecdotalevidence that people try to make us cognizantof their connections with highly positive orsuccessful others. The forms of these connec-tions are varied. For example, they may im-ply similarity of residence, past or present:States and cities like to list the names offamous entertainers, statesmen, beauty con-test winners, etc., who live or were bornwithin their boundaries; the state of Indianahas even gone so far as to brag that morevice-presidents of the United States havecome from Indiana than any other state.Other such connections involve ethnic orreligious affiliation: Italians speak proudly ofthe ethnic background of Marconi, and Jewsrefer to Einstein's heritage. Still other con-nections reflect physical similarities: "Napo-leon was short, too." Sexual identity may alsogive rise to the BIRG phenomenon: At awomen's movement forum attended by oneof the authors, there was a round of feminine

366

BASKING IN REFLECTED GLORY 367

applause when it was announced that Ma-dame Curie was a woman and Lee HarveyOswald was not. Finally, connections suitablefor BIRGing may be as tenuous as an inci-dental contact: We all know people who de-light in recounting the time they were in thesame theater, airplane, or restroom with afamous movie star.

While there appears to be rich informalsupport of the sort described above for theexistence of a BIRG phenomenon, there seemto be no experimental investigations of theeffect. Thus, it was the purpose of this seriesof studies to examine this tendency to baskin the reflected glory of another or group ofothers. In so doing, it was hoped to (a) reli-ably demonstrate the existence of the phe-nomenon, (b) establish its generality overexperimental contexts and measures, (c) de-termine a mediating process for its occurrence,and (d) discover some of its limiting con-ditions and thereby gain further informationas to its nature.

One of the most obvious arenas for theworking of BIRG effects in our society is theathletic arena. Fans of championship teamsgloat over their team's accomplishments andproclaim their affiliation with buttons on theirclothes, bumper stickers on their cars, andbanners on their public buildings. Despite thefact that they have never caught a ball orthrown a block in support of their team'ssuccess, the tendency of such fans is to claimfor themselves part of the team's glory;it is perhaps informative that the chant isalways "We're number one," never "They'renumber one."

It was our view that a sports contextwould be ideal for a test of some of ournotions concerning BIRG effects. Our expec-tation was that an individual would attemptto bask in the glory of an associated, suc-cessful source by publicly announcing his orher affiliation with the source and that thiseffect would obtain even when the affiliationwas clearly irrelevant (i.e., noninstrumental)to the success of the source. In order togather data relevant to the above hypothesis,an experiment was simultaneously conductedat seven universities with powerful intercol-legiate football teams during part of the

1973 football season. It was predicted thatstudents at these schools would be more likelyto announce publicly their connection withtheir universities after the varsity footballteams had been successful than after theteams had not been successful. We decidedto measure students' tendency to announcetheir university affiliation by means of anexamination of wearing apparel. The fre-quency with which students wore apparel thatclearly identified the university that they at-tended was hoped to be a subtle yet sensitivemeasure of the willingness to declare publiclya university affiliation.

EXPERIMENT 1

Method

Procedure. From the third week of the 1973 col-legiate football season through the last week ofregular play, the apparel of students enrolled insections of introductory psychology courses at sevenlarge universities was covertly monitored. At eachschool, three types of data were recorded in thesame classes every Monday during the season: (a)the number of students present in the class, (b) thenumber of students with apparel identifying theschool of attendance, and (c) the number of stu-dents with apparel identifying a school other thanthe school of attendance. Data recorders at eachplace received the following definitions prior to datacollection:

Apparel identifying the school of attendance isidentified as apparel which unambiguously identi-fies your school through names, insignia, or em-blems. Examples would be buttons, jackets, sweat-shirts, tee shirts, etc., which display the schoolname, team nickname or mascot, or universityinsignia. Apparel which appears school-relatedsolely through the use of colors would not qualify.Also excluded are utilitarian objects which an-nounce a university affiliation such as briefcases,notebooks, or bookcovers. Apparel identifying aschool other than the school of attendance arethose which meet the same criteria for inclusionas above but which identify a school other thanyour own.

The data recorders were not members of the classesthey monitored.

Results

Over all schools and across all weeks, anaverage of 176.8 students were present in themonitored classes; an average of 8.4% ofthese students wore apparel identifying theuniversity of attendance, while 2°/Q of them

368 CIALDINI, BORDEN, THORNE, WALKER, FREEMAN, AND SLOAN

wore apparel identifying a school other thanthe university of attendance. Because of hugedifferences among the schools in absoluteamounts of these two kinds of apparel wear-ing and in order to make comparisons be-tween the universities as well as between thetypes of apparel wearing, standardized indexesof relevant apparel wearing were considerednecessary. The standard we decided on was thehighest percentage of relevant apparel wear-ing that occurred on any Monday duringthe season; this standard was simply com-puted as the number of students wearing rele-vant apparel that day divided by the numberof students in class that day. The percentagesof apparel wearing on all other Mondays ofthe season were scored as proportions of thehighest percentage. So, the Monday with thelargest percentage of relevant apparel wearingwas scored as 1.00, and any other Mondaypercentage was scored as a fraction (pro-portion) of that standard. This procedure wasperformed on the data from each school forthe two relevant categories of apparel wear-ing: school-of-attendance apparel wearing andschool-of-nonattendance apparel wearing. Amean proportion for each category was ob-tained for Mondays following a team's winsand nonwins; these are the mean proportionspresented in Table 1. As can be seen fromTable 1, these indexes showed a generallyconsistent tendency for students to wearschool-of-attendance apparel more after vic-

TABLE 1INDEXES OF RELEVANT APPAREL WEARING AT THE

SEVEN MONITORED UNIVERSITIES

School

Arizona StateLouisiana

StateOhio StateNotre DameMichiganPittsburghSouthern

CaliforniaM

School-of-attendance

apparel wearingWins

.63 (5)

.80 (5)

.69 (4)

.67 (7)

.52 (5)

.76 (4)

.36 (6)

.63

Nonwins

.61 (1)

.33 (3)

.30 (1)

.49 (1)

.83 (1)

.27 (2)

.26 (1)

.44

School-of-nonattendanceapparel wearingWins

.58 (5)

.58 (5)

.56 (4)

.62 (7)

.20 (5)•31 (4)

.17 (6)

.43

Nonwins

.68 (1)

.51 (3)

.94 (1)

.52 (1)

.00 (1)

.50 (2)

.00 (1)

.45

tories than nonvictories; but this was notthe case for school-of-nonattendance apparel.Because of the non-normality of the propor-tion data, the scores were converted to ranks,and Wilcoxon matched-pairs signed-ranks testswere performed using school as the unit ofanalysis.1 Despite the conservativeness of suchan approach (for this mode of analysis, n isonly 7), the Wilcoxon T reflected a conven-tionally significant difference on the school-of-attendance measure (T = 2, p < .05, two-tailed) . This result indicated, as predicted, thatMondays following football victories rankedsignificantly higher in school-of-attendanceapparel wearing than Mondays following non-victories.2 The mean rank for victories was3.2, while that for nonvictories was 4.9. Asimilar test for school-of-nonattendance ap-parel did not show any effect; the victoryand nonvictory mean ranks for this measurewere 3.4 and 3.7, respectively. This latterresult suggests that the obtained effect on theschool-of-attendance measure is not attribut-able to a simple tendency to wear clothing ofa certain type (e.g., athletic team jackets,sweat shirts, tee shirts, etc.) after an athleticteam victory.

Discussion

In all, we found support for our expecta-tions concerning the BIRG phenomenon.

Note. Numbers in parentheses represent the number of gamesthat fell into wins and nonwins categories for each school,

1 In any conversion of parametric data to ranks,the possibility exists that the ranked scores will notfully reflect the character of the parametric data.In order to examine such a possibility with respectto our results, a correlational analysis was performedon the standardized index scores and their derivedrank scores. A highly similar relationship (r = — .83)between the two forms of scores was found; thenegativity of this correlation is simply due to thefact that the better ranks of those of lowernumerical value.

2 It may be instructive to note that the singleexception from this pattern in Table 1 occurred atthe University of Michigan as a direct result of a10-10 tie with Ohio State University in a gamefor the Big Ten Conference Championship. Mostobservers, especially the Michigan supporters, feltthat the Michigan team had outplayed Ohio Statethat day and that the game demonstrated Michigan'ssuperiority. However, that tie game constitutedMichigan's only entry in our nonwin category,resulting in the only reversal in our data.


Students chose to display more apparel indi-cators of their academic affiliation after theiruniversity's varsity football team had re-cently been successful. It appears, then, fromthese data and from numerous anecdotal re-ports that people desire to make others awareof what seem to be their causally meaning-less associations with positive sources.3 Why?what do they intend to get from it? Perhaps,the answer has to do with Heider's balanceformulation (1958). Heider discussed twotypes of perceived relations between things:sentiment relations, which imply a feelingstate between stimuli, and unit relations,which merely imply that things are connectedin some manner. It is the unit relationshipthat seems akin to the noninstrumental con-nection that people tend to publicize betweenthemselves and a successful or otherwise posi-tive source. The results of the present experi-ment could well be seen as consistent withbalance theory. For example, if observersperceive a positive unit relationship (e.g., uni-versity affiliation) between a student and asuccessful football team and if observers gen-erally evaluate successful teams positively,then in order to keep their cognitive systemsin balance, the observers would have to evalu-ate the student positively as well. Hence, wemight expect the student to want to makethe unit connection evident to as many ob-servers as possible, in this case, through thewearing of university-identifying clothing.The process whereby one publicly seeks toassociate himself or herself with a successfulother, then, may be reinforced by the tend-ency of observers to respond in a similarfashion to associated stimuli.

Indirect evidence that tends to support thishypothesis comes from research concerningthe transmission of positive and negative in-formation. Manis, Cornell, and Moore (1974)have shown that one who transmits informa-tion that the recipient favors is liked moreby the recipient than one who transmits in-formation that the recipient disfavors andthat this liking occurred even though it wasunderstood that the transmitters did notnecessarily endorse the communicated infor-mation. Like the royal messengers of oldPersia who were feted when they brought

news of military victory but killed when theybrought news of defeat, the transmitters inthe Manis et al. (1974) study acquired thevalence of the message with which they weresimply paired. Moreover, there is evidencethat people recognize this generalization effectand tend to take actions that connect them,in the eyes of observers, with positive ratherthan negative news. For example, Rosen andTesser have repeatedly shown (e.g., Rosen &Tesser, 1970; Tesser, Rosen, & Batchelor,1972; Tesser, Rosen, & Tesser, 1971) thatpeople prefer to be connected with the com-munication of good news to another than withthe communication of bad news. Investigatingthe basic effect, Johnson, Conlee, and Tesser(1974) found their subjects reluctant to com-municate negative information not becausethey felt guilty about transmitting bad newsbut because they feared that they would benegatively evaluated by the recipient of suchnews; again, this was true even though allconcerned knew that the communicators hadin no way caused the bad news. Thus, it ap-pears from these data that: first, individualswho are merely associated with a positive ornegative stimulus (in this case, favorable orunfavorable information) will tend to share,in an observer's eyes, the affective quality ofthe stimulus; and second, at some level indi-viduals seem to understand the workings ofthis phenomenon and make use of it in theways they present themselves to others. Wewish to interpret the results of Experiment 1in terms of this formulation. Students at ourseven monitored universities chose to wearschool-of-attendance apparel after footballteam victories in order to display their con-nection with the successful team and therebyto enhance their esteem in the eyes of ob-servers to the connection. However, anotherexplanation of our findings exists as well.Perhaps the tendency to wear university-

3 It might be argued that some subjects felt thattheir presence in the stands on the day of a gamedirectly contributed to their team's success. Thisseems an unlikely explanation for the obtained re-sults, as an analysis of the data of Experiment 1showed an equally strong BIRG effect for home andaway games.


related clothing following team wins hadnothing to do with an attempt to proclaimthe favorable connection to others but onlyreflected an increased positivity toward theuniversity as a consequence of team success.That is, it is possible that a football victorycaused students to like their school more,and this heightened attraction manifested it-self in the tendency to wear school-identifiedapparel. To test these alternative explanationsand to establish the generality of the tend-ency to BIRG in a different experimentalsituation than that of Experiment 1, a secondexperiment was conducted.

EXPERIMENT 2

The major distinction between the com-peting interpretations described above is thecontention of the BIRG model that studentswore school-of-attendance clothing after vic-tories in order to publicize their universityaffiliations and hence increase their prestigein the view of others. The "heightened at-traction" formulation makes no such claim:One is simply seen to like the school morefollowing victories, and this, rather than thepossibility of increased interpersonal prestige,is said to stimulate the wearing of relevantapparel. We decided to test these explanationsby way of an examination of the pronounusage of university students describing theoutcome of one of their school's football con-tests. Earlier in this article we alluded to thetendency of athletic fans to crowd in frontof television cameras, wave their index fingershigh, and shout, "We've number one!" Thechoice of this pronoun seemed to us a verygood measure of the tendency to BIRG. Byemploying the pronoun we, one is publiclyable to associate oneself with another personor group of persons. Through the use of someother designation, for example, they, one isable to distance oneself from (i.e., to weakenthe perceived association with) another per-son or persons. It was our feeling that inorder to BIRG a successful football team,students would be more likely to describe theoutcome of a team victory using the pronounwe than they would a team nonvictory. Thus,it was our expectation that this tendency toconnect oneself with a positive source but

distance oneself from a negative source wouldinfluence subjects to use the term, "We won,"to describe a team win but use the thirdperson (e.g., "They lost") to describe a teamloss.4 Further, in line with our BIRG model,it was expected that this differential use oflanguage would be most pronounced when thesubject's esteem in the eyes of an observerhad been recently lowered. That is, if we arecorrect in proposing that one proclaims a con-nection with a positive source in an attemptto raise one's esteem in the view of others,then one should be most likely to declare sucha connection when that esteem has recentlybeen jeopardized. Thus, if we were to createexperimentally in subjects a need to bolsteresteem in the eyes of an observer, subjectsshould be most likely to announce publicly(through use of the pronoun we) a connectionwith a successful team and be least likely topublicize a connection with an unsuccessfulteam. On the other hand, subjects who haveless need to elevate an observer's evaluationof themselves should show a lesser effect. Thesimple "heightened attraction" model wouldnot make such a prediction, since one's pres-tige in the eyes of others is not a criticalvariable in that formulation.

Method

Subjects. The subjects were 173 undergraduatesat a large state university with a nationally rankedfootball team. Subjects were randomly selected fromstudent listings in the university phone directory. Thesample included approximately equal distributions ofmales and females.

Procedure. During a 3-day period midway throughthe 1974 football season, subjects were contacted onthe phone by one of 16 experimenters (eight malesand eight females) identified as an employee of a"Regional Survey Center" with headquarters in anout-of-state city. The caller explained that he (she)was conducting a survey of college students' knowl-edge of campus issues and was in town that daycalling students at the subject's university. Subjectsagreeing to participate (93%) were then asked aseries of six factually oriented forced-choice ques-tions about aspects of campus life (e.g., "What per-cent of students at your school are married? Wouldyou say it's closer to 20% or 35%?"). Following

4 It should be evident to the reader that thegeneral statement of the BIRG formulation includesnot only the tendency to bask in reflected glory butalso the tendency to distance unattractive sources.


the subject's sixth response, the caller administeredthe first manipulation. Half of the subjects weretold that they had done well on the test, and halfwere told that they had done poorly. Specifically,subjects were told:

That completes the first part of the questions.The average student gets three out of six correct.You got [five; one] out of six correct. Thatmeans you [did really well; didn't do so well]compared to the average student.

Subjects were then told that there were a fewmore questions and that the first concerned students'knowledge of campus athletic events. At this pointthe second experimental manipulation occurred. Halfof the subjects were asked to describe the outcomeof a specific football game; their school's footballteam had won this game. The other half wereasked to describe the outcome of a different game;this was a game that their team had lost. Thequestion was phrased as follows:

In the [first; third] game of the season, yourschool's football team played the University of[Houston, Missouri]. Can you tell me the outcomeof that game?

If a subject did not know the results of the game,a new subject was called. Otherwise the subject'sverbatim description of the game outcome wasrecorded. At the end of the interview, all subjectswere fully debriefed.

Independent variables. Two factors were manipu-lated; a subject's personal outcome on the surveytask (success or failure) and the affiliated footballteam's outcome in the game described (win or non-win) . These factors combined to produce a 2 X 2factorial design.

Dependent variables. Subjects' tendency to use awe or non-we response in describing a team outcomeconstituted our dependent measure. Descriptions suchas "We won," "We got beat," etc., were consideredwe responses. All other descriptions (e.g., "The scorewas 14-6, Missouri." "They lost.") were classifiedas non-we responses.

Predictions. Two predictions were made. First, itwas hypothesized that subjects would emit more weresponses in describing a team victory than a teamdefeat. Second, it was expected that the effect ofHypothesis 1 would be greatest for subjects whohad "failed" the survey test. The latter hypothesiswas based on the assumption that subjects in thepersonal failure conditions would attempt to asso-ciate themselves publicly with a positive event ordistance themselves from a negative event throughlanguage usage in order to bolster or salvage theirdamaged images in the eyes of the caller. Subjectsin the personal success conditions were not expectedto show a similar sized effect, as their prestige hadalready been ensured via their successful task per-formance. Evidence that public success and failureon an experimental task leads to differential tend-

encies for social approval has been offered bySchneider (1969). He manipulated success and failureand found failure subjects to present themselvesmore favorably to an observer who could providean evaluation. Thus, if the BIRG phenomenon isindeed an attempt to gain social approval, we shouldsee our failure subjects BIRG more than our successsubjects.

Results

Of the 173 subjects, the data of 5 werediscarded because they clearly reported thegame results incorrectly. For example, thedescription "We won" was not counted if infact the subject's team had lost the game inquestion. The percentages of we responsesemitted in the four experimental conditionsare presented in Table 2. The first prediction,that we usage would be greater in the descrip-tions of team victories than team defeats, wastested by comparing the team win conditionsagainst the team nonwin conditions. A sig-nificant effect was obtained, x 2 ( l ) = 4 - 2 0 ,p < .05, confirming Hypothesis 1. The secondprediction, that the tendency for we responsesto attend victory rather than defeat descrip-tions would be strongest after a personal fail-ure, was tested as an interaction of the twomajor independent variables. The resultantstatistic, suggested by Langer and Abelson(1972) for testing interactions within a2X2 contingency table, just missed conven-tional significance levels, Z = 1.75, p < .08,two-tailed. Tests of the simple main effects ofthe interaction strongly supported Hypoth-esis 2. The difference in we responding be-tween the team success and team failure cellsof the personal failure condition was highlysignificant, x

2 ( l ) = 6.90, p < .01. The com-parable test within the personal success con-dition did not approach significance, x2O)= .07, ns. There were no significant sex effectsin the data.

TABLE 2PERCENTAGE OF SUBJECTS USING "WE",

EXPERIMENT 2

Teamoutcome

WinNonwin

2422

% Persona

Success

(11/45)(9/41)

il outcome

Failure

40 (16/40)14 (6/42)

Mean %

32 (27/85)18 (15/83)


DISCUSSION

The data of Experiment 2 seem clearly tosupport the general BIRG formulation. Sub-jects used the pronoun we to associate them-selves more with a positive than a negativesource, and this effect was most pronouncedwhen their public prestige was in jeopardy.We interpret these results as evidence for ourcontention that people display even the mostnoninstrumental connections between them-selves and the success of others so as to re-ceive positive evaluations from the observersof those connections.

It should be evident that the observer'stendency to assign positivity to one who isassociated with positive things is crucial toour hypothesizing about the BIRG phenome-non. It follows from our previously statedassumption that if a person understood thata given observer did not value the success ofa specific source, that person would be lesslikely to try to BIRG that source to theobserver. So, if one of our subjects knew thatan observer abhored successful college ath-letic programs, we would predict that therewould be little likelihood of the subject at-tempting to make visible a connection witha winning football team. But this is a fairlyobvious example; few people would predictotherwise. A more subtle and perhaps moreinformative demonstration might be obtainedthrough a somewhat different manipulationof the observer's relationship to the connec-tion. When an observer to a highly positiveassociation can also lay claim to the associa-tion, the prestige of the connection is diffusedand, consequently, reduced for anyone at-tempting to bask in its glory. It is when one'sbond to a positive source is not shared byan audience that its prestige value is optimal.Thus, when everyone has a similar positivecharacteristic, there is no special prestige in-volved in possessing it, and the likelihoodthat any one person will boast about thatquality should be reduced. For example, aresident of California is less likely to brag tofellow Californians about the favorable cli-mate than to geographically distant others,especially those who cannot claim similarlypleasant weather. It is our hypothesis, then,that the tendency to BIRG a positive source

should occur most often when one's connec-tion with the source is stronger than theobserver's.5 A third study was conducted totest this contention.

EXPERIMENT 3

In Experiment 2, it was shown that a per-sonal failure experience increased our sub-jects' tendency to associate themselves witha positive source and decreased their tend-ency to associate themselves with a negativesource. We have argued that this result oc-curred because the failure experience loweredperceived prestige and motivated subjects totry to either bolster their images in the eyesof others or prevent them from being furtherdegraded. Central to this argument is theassumption that one's simple, noninstrumen-tal connections are seen to influence observers'personal evaluations. If so, it should be thecase that in addition to their use as dependentmeasures, such connections could be used aseffective independent variables. That is, itshould be possible to influence subjects' be-havior by publicly connecting them witheither positive or negative events. In fact, ifwe are correct in our assumption, manipu-lating one's public connections with good orbad things should have the same effect asmanipulating one's personal success or failureexperiences. For example, just as Experi-ment 2 showed that subjects who failed a taskincreased the tendency to affiliate themselveswith a winner and decreased the tendency toaffiliate themselves with a loser in the eyesof an observer to their failure, it follows fromour formulation that subjects who are merelypublicly connected with a negative eventshould emit comparable BIRG responses inthe presence of an observer to that connec-tion. Experiment 3 was designed to testthis possibility and represented a conceptualreplication and extension of Experiment 2.

5 We do not wish to suggest that the tendency toBIRG a positive source never takes place when theobserver's association to a successful other is asstrong as one's own but only that the prestige to bederived from a unique (vis-a-vis the observer)connection is relatively more desirable.


Method

Subjects. The subjects were 170 undergraduates ata large state university with a powerful footballteam. The university was not the same as that ofExperiment 2; however, subjects were selected forparticipation in a fashion identical to that ofExperiment 2.

Procedure. Following the completion of play forthe university's football team, subjects were calledon the phone by 1 of 18 experimenters (11 malesand 7 females) identified as an employee of eitherthe "university's Survey Center" located on campusor the "Regional Survey Center" located in an out-of-state city. Subjects were told that a survey wasbeing conducted of "undergraduates' knowledge ofuniversity athletic events." Those agreeing to par-ticipate (96%) were asked to describe the outcomeof first one, then another of their football team'slast games of the year. One of the games constitutedan important victory, and the other an importantnonvictory in the team's season. Half of the subjectswere first requested to describe the nonvictory gameand, having responded, to describe the victory game.The other subjects had the requests put to themin reverse order. The subjects' verbatim descriptionsof the game results were recorded.

Independent variables. Two factors were orthogo-nally varied: the strength of the subject's affiliationto the university team compared with that of theobserver (same as observer's or stronger than ob-server's) and order of presentation of the games tobe described (victory game description requestedfirst or nonvictory game description requested first).

Dependent variables. The dependent measure wasthe pattern of we and non-we usage employed bysubjects to describe the combination of the victoryand nonvictory games. Three combinations were pos-sible. A subject could have used the same we ornon-we term to describe both the victory and thenonvictory, could have used we to describe the non-victory, and non-we to describe the victory, orfinally, could have used we for a victory and non-wefor a nonvictory.

Predictions. It was predicted, first, that therewould be an overall tendency for subjects to usewe in their descriptions of a team victory andnon-we in their descriptions of a team nonvictory.Such a finding would replicate the basic BIRGeffect obtained in Studies 1 and 2. A second hypoth-esis was that the tendency to use we for victory andnon-we for nonvictory descriptions would be greaterin the nonvictory-description-requested-first condi-tions. Such a result would constitute a conceptualreplication of the second finding of Experiment 2.On the basis of the BIRG model, we expected thatthe effect of publicly describing a negative eventwith which one is connected would be equivalent innature to publicly failing on a task. Both operationswere thought to reduce subjects perceptions of theirprestige as seen by an observer and, hence, to in-crease the likelihood of subjects' attempts to ensurethe positivity of subsequent evaluations. The third

prediction was that Hypothesis 2 would hold moststrongly when the observer was identified with anoff-campus organization. This expectation was basedon the belief that felt prestige to be gained fromone's connections to a source is greater and, thusmore sought after, when one's connections to thatsource are stronger than an observer's. Confirmationof this prediction would appear as an interaction ofthe independent variables of the study.6

Results

As expected and consistent with the resultsof Experiments 1 and 2, the basic BIRG ef-fect occurred in Experiment 3 to support ourfirst hypothesis. That is, subjects used theterm we nearly twice as often to describe avictory than a nonvictory (26% vs. 13.5%).This effect is further confirmed when the dataare examined in terms of individual subjects'we/non-we usage patterns. The majority ofsubjects were constant in their pattern of re-sponding to the two requests for descriptions;they consistently used either we or non-we todescribe both game outcomes. Thus, there wasa strong tendency for our subjects to be con-sistent in their verbal usage patterns for thetwo descriptions. However, in 23 instancessubjects provided an inconsistent we/non-wepattern. In 22 of those instances, the patternsupported the BIRG model; the pronoun wewas used for the victory description and anon-we term was used for the nonvictory de-scription. Using McNemar's test for the sig-nificance of changes (Siegel, 1956, pp. 63-67)

6 The experimenters, undergraduate students in alaboratory social psychology course, were not fullyaware of these predictions. In order to test theinfluence of conscious or unconscious experimenterbias on the results of this study, the experimenterswere informed of the nature of Hypothesis 1. How-ever, they were blind to the more subtle Hypotheses2 and 3. If only Hypothesis 1 were confirmed, thedata would likely have to be interpreted as poten-tially influenced by the experimenter bias artifact.

The investigation of such a possibility was deemedan important one, since in the prior experiments,experimenters had knowledge of the experimentalhypothesis. In Experiment 1, some data recorderswere unintentionally informed of the major hypoth-esis, while others were not. An analysis of the datafrom these two groups found only a minimal dif-ference in the data patterns, with the uninformedgroup's data actually more favorable to predictionthan the informed group. However, in Experiment 2,all experimenters had knowledge of the prediction.


TABLE 3

PERCENTAGE OF SUBJECTS USING BOTH " W E " FORVICTORY DESCRIPTIONS AND " N O N - W E " FOR

NONVICTORY DESCRIPTIONS,EXPERIMENT 3

Order of request*

Strength of subject'sconnection to team

relative to observer's

Victory-descriptionrequested

first

Nonvictorydescriptionrequested

first

Stronger than observer's 3 (1/39) 21 (10/47)Same as observer's 11 (4/36) 14 ( 7/48)

Mean % 7 (5/75) 18 (17/95)

» Numbers given are percentages.

and correcting for continuity, the data arehighly significant, X

2U) = 17.39, p < .001.The tests of Hypothesis 2 and 3 were con-ducted by considering the distribution (acrossthe cells of our design) of the instances ofwe/non-we usage fitting the pattern pre-dicted by the BIRG model. Table 3 presentsthese data.

Hypothesis 2 stated that more subjectswould use we to describe the victory andnon-we to describe the nonvictory when theywere asked to describe the nonvictory first.As expected, subjects were significantly morelikely to so respond in the nonvictory-descrip-tion-requested-first conditions, x2( 1 )=4 .69 ,p < .05. Hypothesis 3 stated that Hypothe-sis 2 would hold most clearly when the sub-jects were more strongly connected with theuniversity team than was the observer. Aspredicted, Hypothesis 2 was supported to agreater extent when the observer was affili-ated with an off-campus rather than a campusagency. However, this tendency did not quitereach conventional levels of significance, Z= 1.72, p < .085. As in Experiment 2, therewere no significant effects for sex of subject.

GENERAL DISCUSSION

Overall, Experiments 1, 2, and 3 providedstrong support for the BIRG formulation. Allthree experiments showed a significant tend-ency for students to strive to associate them-selves publicly with their university's footballteam more after the team had been successful.A striking aspect of the phenomenon is thatsubjects sought to proclaim their affiliationwith a successful source even when they in

no way caused the source's success. Thiscomponent of the effect suggests a mediatorconsistent with balance theory. It is ourcontention that people make known theirnoninstrumental connections with positivesources because they understand that ob-servers to these connections tend to evaluateconnected objects similarly. It appears thatthe tendency to BIRG is an attempt to secureesteem from those who can perceive the con-nection. Studies 2 and 3 provided support forsuch an interpretation. Both showed that ex-perimental operations designed to threaten asubject's esteem in the eyes of an observercaused subjects to be more likely to try pub-licly to associate themselves with positivesources. Intriguingly, it was possible to in-crease the tendency to BIRG in these experi-ments either by initially causing the subjectto experience personal failure in an observer'seyes or by initially causing the subject to benoninstrumentally connected with a negativeevent in an observer's eyes. These manipula-tions proved functionally equivalent in modi-fying subject pronoun usage. Thus, in supportof our basic argument, being merely associ-ated with someone else's success and failurehad much the same effect as personal successand failure. Experiment 3 provided evidencein a different way as well that the desire forprestige is the mediator of the BIRG re-sponse. It demonstrated that when subjects'affiliation with a positive source was strongerthan an observer's (and therefore carried agreater amount of prestige), they were mostlikely to BIRG that source in the presenceof the observer.

These studies suggest a way to understandhow the fortunes of affiliated sports teams cancause lavish displays of civic gratitude andpride in American cities, or "sports riots" inEurope, or murders in South America ofplayers and referees whose actions had causeda home-team defeat. Through their simpleconnections with sports teams, the personalimages of fans are at stake when their teamstake the field. The team's victories and de-feats are reacted to as personal successes andfailures.

Throughout this article we have stressedan interpersonal mediator of the BIRG phe-nomenon—the perceived esteem of others. We


do not wish, however, to preclude the pos-sibility of the tendency to BIRG privately.That is, for wholly intrapersonal reasons,people may draw connections between them-selves and positive sources. For example, onemay well feel an enhancement of self-esteemthat is unrelated to the assessments of otherswhen one is associated with success or posi-tivity. Such an effect could also be interpretedin terms of a tendency to respond similarlyto associated objects. It might be that theresults of our experiments are, in some degreeat least, due to a desire to bolster or maintainone's self-concept. The tendency to employappropriate apparel or language in a way thatconnects oneself to something good may in-volve an attempt to remind oneself of suchconnections and, thereby, positively affectself-esteem. The fact that in Experiment 2we were able to influence the BIRG responsesimply by manipulating the characteristics ofthe observer suggests that the BIRG phe-nomenon is not mediated solely by intra-personal phenomena. Nonetheless, it remainspossible that the tendency to BIRG has itsbasis in a desire to affect self-image as wellas social image. In fact, since there is evi-dence that how we regard ourselves is influ-enced by how we perceive that others regardus (e.g., Harvey, Kelley, & Shapiro, 1957),these two mediators are not mutually exclusive.

REFERENCES

Harvey, 0 . J., Kelley, H. H., & Shapiro, M. M.Reactions to unfavorable evaluations of self madeby other persons. Journal of Personality, 1957,25, 393-411.

Heider, F. The psychology of interpersonal rela-tions. New York: Wiley, 1958.

Johnson, R., Conlee, M., & Tesser, A. Effects ofsimilarity of fate on bad news transmission: Areexamination. Journal of Personality and SocialPsychology, 1974, 29, 644-648.

Langer, E. J., & Abelson, R. P. The semantics ofasking a favor: How to succeed in getting helpwithout really dying. Journal of Personality andSocial Psychology, 1972, 24, 26-32.

Manis, M., Cornell, S. D., & Moore, J. C. Transmis-sion of attitude-relevant information through acommunication chain. Journal of Personality andSocial Psychology, 1974, 30, 81-94.

Rosen, S., & Tesser, A. On the reluctance to com-municate undesirable information. The MUM ef-fect. Sociometry, 1970, 33, 253-263.

Schneider, D. J . Tactical self-presentation aftersuccess and failure. Journal of Personality andSocial Psychology, 1969, 13, 262-268.

Siegel, S. Nonparametric statistics for the behavioralsciences. New York: McGraw-Hill, 1956.

Tesser, A., Rosen, S., & Batchelor, T. On the re-luctance to communicate bad news (the MUMeffect): A role play extension. Journal of Person-ality, 1972, 40, 88-103.

Tesser, A., Rosen, S., & Tesser, M. On the reluctanceto communicate undesirable messages (the MUMeffect): A field study. Psychological Reports,1971, 29, 651-654.

(Received October 14, 1975)

basking in reflected glory: three (football) field studiesschaller/528readings/cialdini1976.pdf ·...

Documents