interpersonal attraction as a communication accomplishment: …€¦ · interpersonal attraction as...

Interpersonal Attractionas a CommunicationAccomplishment: Developmentof a Measure of Affinity-Seeking Competence

ROBERT A. BELL, SHERYL W. TREMBLAY, andNANCY L. BUERKEL-ROTHFUSS*

The five studies described here lay the foundation for investigations of an importantconstraint under which individuals operate when attempting to engender liking, theiraffinity-seeking competence. Several studies were undertaken to develop and validate aself-report instrument for assessing skills of affinity-seeking. Results indicate that theAffinity-Seeking Instrument (ASI), which is comprised of two scales, has internal con-sistenc}-, a stable factor structure, and concurrent and discriminant validity. These studiesalso indicate that the instrument measures social tendencies that are observable to othersand predictive of positive social outcomes.

S OCIAL SCIENTISTS HAVE GIVEN Considerable attention to thequestion of when and why individuals are attracted and attractive

to other people (Duck, 1977; Huston, 1974). This interest is justified,for few factors influence people's social success and satisfaction in lifemore than their ability to develop bonds with others. A primary researchgoal has been the explication of personal and dyadic antecedents of in-terpersonal attraction. Investigators have given much attention to therelationship of attraction to such variables as proximity, physical at-tractiveness, and similarity. The evidence indicates that interpersonalattraction may be facilitated by propinquity (Nahemow & Lawton, 1975),physical attractiveness (Berscheid & Walster, 1974), and similarity inbackgrounds, attitudes, and values CByrne, 1971).

Dissatisfaction with this traditional focus on static characteristicsof persons has recentlj' been voiced (e.g., Gergen, 1982). In particular,the failure of many theorists to recognize that persons develop (or failto develop) attraction to others within a social context characterized by

The Western Journal of Speech Communication, 51 CWinter 1987), 1-18

2 Western Journal of Speech Communication

communicative exchanges is most problematic (Bochner, 1984). Consideras just one example the vast literature on attitude similarity as a precur-sor to attraction. This research has usually utilized BjT-ne's (1961,1971)attraction paradigm in which individuals indicate their degree of lik-ing for a bogus stranger on the basis of written descriptions of the non-existent person's attitudes toward some issue. The generalizability ofthese studies to actual social encounters is questionable. As Bochner(1984) has noted, a person's attribution of similarity or dissimilaritytypically follows rather than precedes interaction, is seldom based onwritten information, and is rarely unilateral. Recent work supportsBochner's criticisms by demonstrating that the attitude-similarity/interpersonal-attraction relationship is greatly attenuated when in-dividuals are given the opportunity to engage in normal interaction(Sunnafrank, 1983, 1985; Sunnafrank & Miller, 1981).

Not all attraction research has been guided by such a static view.A second tack examines attraction as an interactional concept. Thisperspective focuses on cultural rules to which social actors must conformwhen trying to be liked bj' others and on social skills facilitative ofattraction. Jones (1964), for instance, has explored strategies and mo-tives of ingratiation. Da\'is (1973) has identified six tasks that anindividual must perform successfully to get another person to desirefuture interactions. Harre (1977) has Eirgued for an ethogenic perspec-tive on interpersonal attraction in which the development of friendshipsand "enemyships" are considered social accomplishments. Cook (1977)has presented a social skills model of attraction, stressing its strategicand interactional nature. Argyle and Henderson (1984) have carried outa cross-cultural program of research on the rules of friendship.

Bell and Daly (1984) have attempted to synthesize and extend thethinking of these scholars by advancing a dynamic, communication-based perspective on interpersonal attraction, labeled affinity-seeking.They offer a model of the affinity-seeking process that identifiesfour sets of issues: antecedent factors, constraints, strategic actiidties,and target responses. "Antecedent factors" are characteristics of theinteraction in which affinity-seeking occurs. "Constraints" referenceindividual and situational influences on affinity-seeking choices."Strategic activities" refer to the selection, integration, sequencing, andenactment of strategies. "Target responses" encompass the effects ofaffinity-seeking, including changes in another's attraction to the affinity-seeker, reciprocity, and attributions of a motive for seeking affinity.In their initial investigations. Bell and Daly (1984) identified twenty-five strategies, clustered into seven higher-order categories, and foundperceptions of individuals' affinity-seeking behavior to be stronglyrelated to interpersonal attraction. Their studies also suggest thatstrategy selection is shaped by personality and situational param-eters.

Winter 1987 3

The purpose of the present series of investigations is to elaborate onthe constraints of affinity-seeking. Specifically, this article details thedevelopment and validation of a self-report measure of affinity-seekingcompetence. Following Cook (1977), we assume that affinity-seeking en-compasses a set of communication skills on which persons may differ.Some people have sophisticated repertoires of affinity-seeking strategies,a good understanding of the appropriateness of each in varying situa-tions and relationships, and an ability to sequence and integrate thesestrategies with considerable finesse. These people should be quite suc-cessful in their social endeavors. Conversely, some people may be defi-cient in their affinity-seeking behaviors. They may, for instance, havelimited knowledge of available strategies and a poor sense of how theirstrategic behavior should be adapted to meet situational requirements.To the extent that affinity-seeking competence fosters positive reactionsfrom others, people probably become quite aware of their abilities tofacilitate liking and should thus be able to make reports of these skills.Though numerous measures exist for assessing related constructs suchas sociability, social competence, affiliative tendencies, self-monitoring,and social anxiety, we know of no instrument for measuring people'sabilities to manage their communication to gain favor.

Of course, the development of a reliable and valid measure of affinity-seeking competence is unimportant in and of itself. Such an assessmenttool does, however, make possible the investigation of a number ofcritically important questions. For instance, what strategies andbehaviors discriminate skilled affinity-seekers from incompetent ones?What are the consequences of poor affinity-seeking? What is the impor-tance of competence relative to traditional variables such as attitudesimilarity and physical attractiveness? Answers to questions such asthese await the development of a measxire of affinity-seeking com-petence. Such a measure is advanced in this article. The first studyreports the development of a multidimensional self-report measure ofskill, dubbed the Affinity-Seeking Instrument (ASI). The second in-vestigation provides a replication of the factor structure of the instru-ment. The concurrent and discriminant validity of the ASI is evaluatedin study three. Study four examines the validity of this measure by com-paring people's scores on the ASI to others' perceptions of their affinity-seeking skills. The fifth investigation employs a multiple-act criterionto examine the relationship of ASI scores to social outcomes.

STUDY 1: INSTRUMENT DEVELOPMENT

To assess affinity-seeking competence, the authors initially generateda pool of over one hundred items, each referencing the manipulation,implicitly or explicitly, of social behaviors for the purpose of garneringthe attraction of others. Our intention was to develop a unifactorial


measure of affinity-seeking competence. Twenty-five items most consis-tent with the affinity-seeking construct were selected from the Item poolfor inclusioB in the initial collection of data. These items were ad-ministered to 466 people (208 males and 258 females) enrolled in basiccommunication courses. Of these students, 284 were from a publicuniversity in a northern state (134 males and 150 females), and 182 werefrom a large eastern university (74 males and 108 females).

Contrary to expectations, a two-factor solution was obtained for theresponses of these items. The first factor was labeled affinity-seekingcompetence (ASC). Examples of items loading on this factor include: "Iseldom know what to say or do to get others to like me," "I have troublebuilding rapport with others," and "I just can't seem to get others to likeand appreciate me." These items reference self-perceptions of a generalability to say and do what is necessary to be interpersonally attractive.The second factor was labeled strategic performance (SP), and includeditems such as: "I can put on excellent social performances to get othersto approve of me" and T can present myself as more likeable than I reallyam." These items describe a very specific kind of affinity-seeking com-petence, namely the ability to play roles, even to the point ofmisrepresenting one's self, to gain the liking and approval of others.Strategic performance in affinity-seeking is conceptually related to, butnot equivalent with, Snyder's (1974) self-monitoring construct. Snyderhas examined individual differences in the tendency to adjust one'spresentation of self to fit the requirements of social situations. High self-monitors ask, "Who does this situation want me to be and how can Ibe that person?" while low self-monitors ask "Who am I and how canI be me in this situation?" (Snyder, 1979, p. 110). Self-monitoring is abroader construct than the strategic performance dimension of affinity-seeking competence because the curr3dng of favor is just one of a numberof reasons for enacting social performances. These two categories ofafBnity-seekiBg skill parallel Argyle's (1972) model of social skills.Argyie suggests that social skills operate at the levels of routine, social-ized responses and conscious plans.

Although the results of this investigation were promising, a revisionofthe scale was deemed necessary for two reasons. First, several itemshad high loadings on both factors. Some of these items were revised tobe consistent with just one factor while others were deleted. Second, onlythree items had primary loadings on the second factor. New items werethus written to represent the strategic performance construct. The re-vised ASI, composed of twenty-six items, was administered to a groupof 168 undergraduates (96 females and 72 males) at a large easternuniversity. As before, participants indicated their level of agreementwith each item on 7-point Likert scales ranging from "very stronglydisagree" to "very strongly agree."

Winter 1987 5

A series of factor analyses (principal components with Varimax rota-tion) was carried out to test the hypothesized two-factor structure. Tobe retained, an item had to have a primary loading of at least .50 withno secondary loading exceeding .30. As expected, a two-factor solutionwas obtained. The final instrument is reported in Table 1, along withfactor loadings, item means and standard deviations, and corrected item-total correlations. The first factor was composed of eight items writtento represent the ASC scale while the second factor was composed of fiveitems for the SP measure. Composite scores were computed for the twofactors by summing across the items loading on each (Ms: 41.31 for ASCand 21.39 for SP; SDs\ 6.83 for ASC and 5.49 for SP; alpha reliabilities:.86 for ASC and .80 for SP). The correlation between these two scaleswas .31. There was no sex difference for affinity-seeking competence orstrategic performance.

To assess the concurrent validity of the ASI, participants alsoresponded to six single-item measures of personal and social function-ing: "I am a shy person," "I am a lonely person," "I feel I am a good com-municator in one-to-one situations," "All in all, I am a very happy per-son," "I feel I am in control of my life," and "I often wish I had morefriends." Participants' scores on the ASC correlated negatively andsignificantly (p<.05) with shyness (r= -.48) and loneliness (r= -.63),and positively with reports of communicator image (r= .52), happiness(r=.44), feelings of control (r=.38), and satisfaction with number offriends (r = .39). Strategic performance had a significant, negative cor-relation with shyness (r = - .25); all other correlations were nonsignifi-cant. Thus, individuals' self-perceived competence in seeking affinitywas strongly related to various aspects of personal and social well-being.The failure of the SP scale to correlate highly with these items mayreflect a lack of congruity. The strategic performance measure may tapa set of skills more relevant to situations of an instrumental than a rela-tional nature (e.g., the job interview). This possibility awaits furthertesting.

STUDY 2: REPLICATION OF FACTOR STRUCTURE

Participants and Procedures

A second study was conducted to assess the stability of the factorstructure of the Affinity-Seeking Instrument in a more diverse sampleof individuals. Participants were 295 elementary and secondary schoolteachers (38 males and 257 females) enrolled in summer extensioncourses offered through a state university in the eastern part of the U.S.Participants indicated their level of agreement with each item of theASI on 7-point Likert scales ranging from "very strongly disagree" to"very strongly agree."

Western Journal of Speech Communication

_O 000!N05LO OCO'r^. f- . %^^ fc*"\ 1 ^ ^ \ ^ \ t*^ t ^S I

lO

1.47

oin

1.32

jro cLO U

1.23

0 CO? CQ~. LO

1.10

LO

o

LO O>

1.05

1.19

1.57

to o •*C^ C3 O5• * CO CO

1.48

1.35

1.55

onCO

1.27

It£

<

as

E a.

o c 2- 03 =

5 ?

>-, oCC tJ

II i

5c» 3 — "5

o

c: •

SS 3

CO C8O >• to

1 22 £

M:^ ":5

Bt 2

c a 2 »^.

;,-. g S-^ J3 3

OI ini.O->3'(M t O O Otc to to to C- CO .-1 00

£a,a.

I-CtJ H-( H-t H-4 )>-(

d 1-H oi CO

II

^ Q QJ -« 3

> c > c o.

tJ

00

to

co"

I 1I t£ c

Winter 1987 7

Results and Discussion

The ASC had a mean, standard deviation, and alpha reliability of46.16, 7.27, and .90, respectively. The SP scale had a mean, standarddeviation, and reliability of 27.52, 4.88, and .81, respectively. Theteachers' responses vi'ere factor analyzed using principal componentsanalysis with Varimax rotation. A two-factor solution, accounting forsixty percent of the variance in the data, was indicated by both theKaiser (1960) criterion, which advocates retaining only those factors witheigenvalues greater than 1.0, and by Cattell's (1966) scree test. Itemsfrom the ASC scale loaded on the first factor, while items from the SPmeasure loaded on the second factor. Every item on the ASC scale hada loading of at least .54 on Factor I (average loading: .72); the highestloading for the second factor was .29 (average ASC loading on FactorII: .12). Every item on the SP scale had a loading of at least .51 on thesecond factor (average loading: .68); the highest loading for the first fac-tor was .22 (average SP loading on Factor I: .16). These results providestrong evidence that the factor structure identified in the precedingstudy is stable.

STUDY 3: CONCURRENT AND DISCRIMINANT VALIDITY

A third study was carried out to assess more thoroughly the concur-rent and discriminant validity of the ASI. Affinity-seeking skill was ex-pected to correlate with other constructs indicative of social activity,social awareness, and interpersonal competence. The constructs assessedwere sociability, shyness, assertiveness, self-monitoring, self-esteem, in-teraction involvement, loneliness, and communication apprehension.Discriminant validity analyses were conducted to determine if theaffinity-seeking competence and strategic performance constructs, asoperationalized by the ASI, are distinct from these related variables.

Procedures

Participants were 64 females and 50 males (A^=114) enrolled in basiccommunication courses at a large eastern universitj'. The students com-pleted the ASI and other measures of social-communicative individualdifferences in randomized order. Shyness and sociability were assessedwith measures developed by Cheek and Buss (1981). Shyness was ex-pected to correlate negatively with both ASI scales, while positive cor-relations were expected for sociability.

Assertiveness was assessed with Lon- and More's (1980) multidimen-sional measure, which includes scales for assessing social assertiveness,directiveness, defense of rights and interests, and independence. "Socialassertiveness" refers to the ability to project oneself into social situa-tions. As such, it is defined by a set of skills necessary, though not suffi-cient, for effective affinity-seeking (e.g., introducing oneself and


carrying on conversations). We thus expected positive relationships be-tween social assertiveness and the two ASI scales. "Directiveness"describes the ability to take control of problematic interpersonal situa-tions via the demonstration of leadership. One would expect a positivecorrelation between directiveness and affinity-seeking competence, asleadership qualitj' should be enhanced by such skills. "Defense of Rightsand Interests" measures the willingness to defend one's rights and toreject the unjust demands of others, ^ndependence" is the ability to resistsocial pressures to conform to expectations that are personally objec-tionable. These last two dimensions of assertiveness were consideredconceptually unrelated to affinitj*-seeking comf)etence and were includedto assess discriminant validity.

Interaction involvement was measured with Cegala's (1981) Interac-tion Involvement Scale (IIS), an 18-item measure purportedl}' composedof responsiveness, perceptiveness, and attentiveness dimensions. Sincethe IIS proved to be unifactorial in this study, a single interaction in-volvement score was assigned to each participant by summing acrossall items. Interactionally-involved people are highly participative com-municators. Interaction involvement is presumably a critical prere-quisite of effective affinity-seeking and should thus correlate positivelywith both dimensions of the ASI.

Self-monitoring was assessed with Snyder's (1974) Self-MonitoringScale. As noted earlier, high self-monitors adapt their behaviors to meetthe requirements of social situations and others' expectations. Briggs,Cheek, and Buss (1980) have found that Snyder's instrument is three-dimensional, including acting, extraversion, and other-directedness fac-tors. Though all three scales should correlate positively with bothmeasures of affinity-seeking, an especially strong relationship betweenstrategic performance and the acting dimension of the self-monitoringscale was anticipated.

Self-esteem was measured with Rosenberg's (1965) ten-item scale.Since affinity-seeking skills should enhance the probability that one willbe accepted by others, affinity-seeking competence and strategic per-formance skills were expected to correlate positively with self-esteem.Communication apprehension was measured with McCroskey's (1981)PRCA-24. This questionnaire assesses the tendency to feel apprehen-sive about communicating within dyadic encounters, groups, meetings,and public speaking situations. Both scales of the ASI were expectedto correlate inversely with communication apprehension within eachsituation. From the standpoint of discriminant validity, the correlationswere expected to be strongest for dyadic communication apprehensionand weakest for public speaking apprehension, given the interpersonalorientation of the ASI.

Loneliness was measured with the twenty-item Revised UCLALoneliness scale (Russell, Peplau, & Cutrona, 1980). Individuals with

Winter 1987 9

finely-tuned affmity-seeking skills should be more successful in meetingtheir relational needs than people with a skills deficiency. For thisreason, inverse relationships were predicted for loneliness and the twoASI scales. Finally, subjects completed Crowne and Marlowe's (1964)social desirability scale to assess the vulnerability of the ASI torespondents' needs for social approval.


Table 2 reports the alpha reliabilities, inter-correlations, means, andstandard deviations for all measures. The ASC and SP scales had ac-ceptable reliabilities (.85 and .87, respectively); their inter-correlationwas .48. As Table 2 documents, scores on the ASC and SP scales werenot significantly related to social desirability. A factor analysis of theASI replicated the two-factor structure found in the first study with theone exception that item 5 had a loading of .32 on the affinity-seekingcompetence factor.

Concurrent Validity. The pattern of correlations in Table 2 suggeststhat individuals reporting high affinity-seeking competence also believethey are assertive, communicatively nonapprehensive, interaction in-volved, nonlonely, high in self-esteem, good social actors, extraverted,nonshy, and somewhat sociable. As expected, social assertiveness proveda better predictor of the ASC measure than the other forms of asser-tiveness. Using a test of the significance of differences between depen-dent correlations (Cohen & Cohen, 1975), it was determined that thecorrelation between ASC and social assertiveness was significantlygreater (p<.05) than the correlations between the ASC scale and theother three assertiveness measures. Likewise, dyadic communicationapprehension was more strongly related to ASC than was apprehensionin group, meeting, or public situations (p< .05). Contraj-y to prediction,the other-directedness scale of the self-monitoring questionnaire did notcorrelate significantly with affinity-seeking competence.

The strategic performance measure did not correlate as highly withthe criterion measures as did the ASC measure. Nevertheless, in-dividuals reporting competence in strategic performance of affinity-seeking had a tendency to perceive themselves as assertive, nonap-prehensive in dyadic and meeting situations, interaction involved, goodsocial actors, extraverted, other-directed, and nonshy. The moderate rela-tionships between the strategic performance measure and the actingfactor (r=.4O) ofthe self-monitoring scale is especially worth noting,given the performative nature of each.

Discriminant Validity. The bivariate correlations reported in Table2 indicate that the affinity-seeking competence and strategic perfor-mance constructs may be discriminated from measures of otherconceptually-related constructs. Even after correcting for attenuation,no variable accounted for more than 36 percent of the vai-iance in


a:

£•aX

a

sX

•5

>a;

lard

1:

c

iN ^ ??

r" X cca

X

rrel

atii

-

s5

ribi

lit

0;

CO

X

10T-l

Tft—<

r:

CJ

t—1

>—'

X

tc

___

—•

-.

c

XCO

0

I

CO

Oi

!

j

CO1

1

•t

CM

-

X

CO•*}*

lOX

ASC

•-'

CO

1

C"CM1

CO

0

0Tf

c

-

CMCM

1

X

1

CO

1

COCOi

CM

0

CO

X

&cc

CM

10

6.3

i

1

c

C^

_ i

0!

'v£CO

CO

1

cCO

COC]

1

gCO' • "

X

CO

iven

ess

Ass

ert

CO

2

Sa;a:

(Def

en

CO

oi

IC0

laCO1

0.—(

1

01CO

CO

IM

IM

COCM

CMCO

1

XCOj

XCO1

CO

CO

X

iven

ess

Ass

ert ;iv

enes

s)(D

irec

l18

6.

0

d

CO

CO01

COCO1

XCM1

XCM

0

CO.—'

1

01CM

1

COCO1

XCMi

in

01CD

iven

ess

Ass

ert

in

ende

ncel

dapuj)

CO

01

COCM

CMCM

0UO1

in01

to• « •

to' -

01

COCO1

0

toCO1

1

-46

l-O10

1

01X

CO

iven

ess

Ass

ert

CD

15

4.1

CMCO1

CO1"

CD

CM.—<

toLO

1

TTCO1

X

1

CM

to1

CO

10CD

COCO

l-OX

B

sio

2"oJI.

a.

B

unic

atio

c)C

omm

(Dya

di

CO,

"1

CM1

01

CM

CD

1

CO

!

i

CJ

0UO

1

iCCO

(Gro

up!

B

sio

B

%

rel

0.a

B

unic

atio

Com

m

X

CM

UO

1

CM

CM

CO

[

tD'

1

1

toCO

1

eg

0

(Mee

B

SIO

a>r.Pa.

B

unic

atio

Com

m

O)

'

ir

c

CJ

j

c

1

X1

0

XCJi

XX

u

(Pub

l

B0

rel

unic

atio

Com

m10

.53

9.

2

CO

10

1

CMCM

COCM

CM1

t-X

ci

si

"p

ctio

n In

iIn

tera

i11

.

CM

id

cq

CO

01CO

CMC

C^CO

1

OlX

ness

0

c

12.

gd

CM

_

....CMi

[

X

stee

mSe

lf-E.

'13

.

X

IC

CM

1

c

CO1

.....

CO

CM

0

onit

orin

Self-

M14

.

CO

CO

cCDi

CMC

cX

CS

X

s

onit

orin

Self-

M15

.12

5.

2

01

i

010

XCO

Ia;

-eel

5Ci

0

onit

orin

Self-

M16

.

Oi

CM

CO

"?i

CD

Xm

Shyn

e;

oi

01

XCO

ilit

ySo

ciab

18.

X

0

>

iii

Xia

cC

Soci

al19

Winter 1987 11

eitber ASI scale. A more rigorous test of discriminant validity was car-ried out via regression analyses by assessing tbe amount of variancein tbe ASC and SP measures left unaccounted for by optimal linear com-binations of tbe set of criterion measures. Tbese variables were enteredinto two independent stepwise regression analyses witb tbe ASC andSP scales serving as dependent measures. At eacb step in tbe analyses,tbe independent variable witb tbe smallest F-to-enter probability wasintroduced into tbe equation, provided tbat its associated probabilityvalue did not exceed tbe .05 level. Tbe results of tbese analyses arereported in Table 3.

Table 3Stepwise Regression Analysis Predicting ASI Scores

From Conceptually-Related Measures(Study 2)

DiDendentVariable

Affinity-SeekingCompetence l.'iSC)

StrategicPerformance (SP)

Step

123

1234

Predictor

Dyadic Comm. ApprehensionLonelinessSocial Assertiveness

ActingOther -DirectednessDyadic Comm. ApprehensionIndependence

R

.507

.579

.603

.396

.471

.535

.558

.257

.335

.364

.157

.222

.287

.312

R2Change

.257

.078

.029

.157

.065

.065

.025

Std.Beta

-.320-.253

.207

.179

.368-.258

.188

OverallF

38.77*27.95*20.95*

20.88*15.86*14.73*12.35*

•p<.001

For affinity-seeking competence, tbe significant predictors weredyadic communication apprebension, loneliness, and social asser-tiveness. Tbese tbree variables accounted for 36 percent of tbe variancein tbe ASC scale. Tbus, 64 percent of tbe variance in tbis variable couldnot be accounted for by tbe set of construct-related measures. For tbestrategic performance mesisure, tbe significant predictors were acting,otber-directedness, dyadic communication apprebension, and in-dependence. Tbese measures accounted for 31 percent of tbe variancein tbe SP scale. Tbus, 69 percent of tbe variance in tbis measure couldnot be accounted for by tbis set of variables. Tbe results of tbis studyindicate tbat tbe ASI bas concurrent and discriminant validity.

Tbe next section reports an investigation of an equally importantquestion; Do people bave sufficient awEireness of tbeir affinity-seekingskills to report accurately on tbese skills? Earlier we speculated tbat


the social rewards people receive are directly related to their ability tofacilitate positive regard. As such, we posited that individuals probablydevelop a sense of their affinity-seeking competence and could thus of-fer valid reports of these skills. This proposition was put to an empiricaltest in a fourth investigation. The issue here is whether people's percep-tions of their competence, as assessed by the ASI, are consistent withothers' perceptions of their abilities. This question was addressed by com-paring people's self-reports of their proficiency at affinity-seeking to theirfriends' perceptions of their competence.

STUDY 4: FRIEND EVALUATIONS


Students enrolled in communication courses at a large easternuniversity were asked to participate in a study of liking. Each broughta close friend to the investigation, someone who knew him or her well.Participants were 43 pairs of friends (20 male-male pairs, 15 female-female pairs, and 8 mixed-sex pairs). After each pair arrived to adesignated room at a specified time, they were separated and then givena questionnaire consisting of two copies ofthe ASI. Initially, participantswere asked to complete the ASI with their own communication tenden-cies in mind (hereafter referred to as their self-rating). Next, each par-ticipant was told to pretend that he or she was the friend and to com-plete the ASI as the friend should if the friend was being completelyhonest in his/her self-assessment (hereafter referred to as thefriend-rating).


Each participant's self-rating of affinity-seeking skill was matchedwith the friend-rating of his/her comrade (A^=86). The alpha reliabilitiesof the scales for the self-ratings were .89 for ASC and .84 for SP; theinter-correlation for these two self-ratings was .29. The means (and stan-dard deviations) were 43.20 (7.61) for the ASC scale and 21.98 (5.79) forthe SP measure. The reliabilities for the friend-ratings on the ASC andSP measures were .84 for both. If self-reported affinity-seeking skillreflects actual abilities, we should expect substantial correlations be-tween the self- and friend-ratings on the ASC and SP scales. The cor-relation between self-ratings and friend-ratings of ASC was .44 (p< .001);after correcting for attenuation, the correlation was .51. The correla-tion between self-ratings and friend ratings on the SP measure was .37(p< .001); the unattenuated correlation was .44. Thus, there was a cor-respondence between people's reports of their affinity-seeking skills andtheir friends' perceptions of these abilities.

Winter 1987 13

Although these correlations are only moderate, their magnitude isprobably attenuated by a number of factors. First, even our closest frienddoes not have the opportunity to observe our tendencies in every typeof social situation. Thus the self-ratings reference a wider range of ex-periences and situations than friends' evaluations. Second, the ASI maymeasure social abilities most relevant to interactions among strangersand acquaintances. There may be only an occasional need to utilize theseskills when with friends. Bell and Daly (1984), for instance, speculatethat the skills of afTinity-see^m^ and afTmity-maintenance may bequalitatively different.

STUDY 5: AFFINITY-SEEKING COMPETENCE AND SOCIALOUTCOMES

A fifth study was undertaken to assess the relationship of affinity-seeking skill to social outcomes. When attempting to demonstrate a linkbetween behavior and a self-report measure of individual differences,investigators have at least three research alternatives from which toselect. They may attempt to demonstrate an association by relating peo-ple's scores on the measure to an observation of a single behavior ona single occasion, a single behavior on repeated occasions, or multiplebeha%aoral acts over time. This study adopts the third approach. As Daly(1978) has noted, a multiple act criterion is preferable since it isunreasonable to expect all individuals to reflect some disposition in ex-actly the same way. We hypothesized that both scales of the ASI wouldhave considerable correlations with a sum of social outcomes intuitivelyassociated with affinity-seeking.

Selection of Outcomes

Students enrolled in communication courses generated the outcomesused in this study. These students were given descriptions of prototypicalcompetent and incompetent affinity-seekers and asked to generate a listof outcomes more likely to be obtained by high-skilled than low-skilledaffinity-seekers. A final list of fourteen outcomes was selected on thebasis of nonredundancy and consensus (examples: making a new friend,getting someone to go with you on a date, being asked by another per-son to go out on a date, and receiving a gift). While one might objectto the use of self-reports of behavior in lieu of direct observations,previous investigations have found such reports to be accurate whenthe meaning of the behaviors in question are relatively clear (Dean,1958; Fishbein & Ajzen 1974, 1975; Pomazal & Jaccard, 1976). Morepractically, reliance on self-reports of behaviors is necessitated by theimpossible task of observing 14 naturally-occurring events for a groupof research participants over an extended time period.



Seventy-four students (32 males and 42 females) were given, in ran-domized order, the two ASI scales [alpha reliabilities: .88 for ASC and.86 for SP) and a checklist of the outcome variables. Participants wereinstructed to indicate those outcomes they had obtained in the paist sevendaj'S. Each participant was given one point for each outcome he or shechecked off. The multiple act measure was constructed bj' summingacross the fourteen outcomes {alpha reliability: .56).


The means (and standard deviations) for the ASC, SP, and outcomevariables were, respectively, 39.52 (9.86), 22.12 (6.07), and 7.10 (3.21).The correlation between the ASC and SP scales was .51 (p<.001). TheASC had a correlation of .43 (p<.001) with the sum of the outcomes,and an unattenuated correlation of .61. The correlation between thestrategic performance scale and the outcome measure was .36 (p < .001);the unattenuated correlation was .52. Although these results are onlj'correlational, they suggest that affinity-seeking skills facilitate the ob-tainment of positive social outcomes.

GENERAL DISCUSSION

The theoretical explication of affinity-seeking requires anunderstanding of factors that shape persons' attempts to generate lik-ing (Bell & Daly, 1984). We have sought to lay the groundwork for in-vestigations of what is perhaps the most fundamental constraint onaffinity-seeking, personal competence. The notion that individuals dif-fer in their abilities to gamer positive regard jibes with experience. Mostof us are acquainted with people who seem to know precisely what mustbe done to make themselves attractive to others. For these people, foster-ing good impressions and making friends appears to come naturally.On the other hand, most of us are familiar with individuals who findit terribly difficult to get others to like them. For these folks, it seemsthat the more they persevere, the less they are successful.

Research on the etiology, correlates, and consequences of affinity-seeking abilities will not progress until assessment procedures aredeveloped and validated. As Bell and Daly (1984) note, it is necessaryto devise measures of competence at two levels, knowledge and perfor-mance. The former pertains to individuals' awareness of strategic op-tions, while the latter relates to the utilization of this knowledge in on-going interaction. Knowledge is necessaiy but not sufficient for compe-tent performance. Some people might be quite knowledgeable of affinity-seeking but lack the motivation to put this knowledge into effect. Evenindividuals with knowledge and motivation may be unable to perform

Winter 1987 15

adequately for otber reasons, sucb as communication anxiety or a lackof experience witb tbe situations in wbicb tbey find tbemselves.

Bell and Daly (1984) bave reported a metbodologj' for measuringthe knowledge component of competence. Tbey asked researcb partici-pants to describe everytbing tbat could be said or done to get a targetperson in a bypotbetical interpersonal situation to like tbem. Tbeseresponses were initially unitized to create for eacb participant a list ofnonredundant affinity-seeking bebaviors, and tben counted to derivea measure of knowledge depth. Next, tbese bebaviors were coded intoa typolog5' of affinity-seeking strategies to construct a measure ofknowledge breadth, representing tbe number of strategies in eacb par-ticipant's responses. Altbougb tbis procedure bas not been used exten-sively, preliminary results demonstrate its potential. Tbe presentresearch complements and extends tbis work by providing a brief andeasily administered metbod for assessing people's abilities to performcompetently wben seeking affinity. Altbougb these studies can only pro-vide tentative evidence of reliability and validity, tbe results are prom-ising. First, botb ASI scales were consistently reliable. Second, tbe fac-tor structure of tbe instrument seems to be stable. Tbird. tbe ASI ap-pears to be a valid measure of competence. Tbe instrument bas facevalidity in that eacb item references individuals' abilities to managetheir social bebavior to enbance attraction. Tbe pattern of correlationsbetween tbe two .ASI measures and otber social dispositions sucb asshyness, social assertiveness, and dyadic communication apprebension,as well as tbeir relationsbips to self-evaluations, including lonelinessand self-esteem, suggests concurrent validity. Yet, on tbe basis of tbediscriminant validity analyses, it cannot be concluded tbat tbe .• SI issimply anotber measure of extanc constructs.

A basic requirement of any self-report of individual differences inovert communication bebavior is tbat it reflect actual variations inbebavior across people. Two studies were undertaken to determinewhether self-perceptions of affinity-seeking competence are rooted inbehavioral tendencies. The first study related participants' .^Sl scoresto friends' perceptions of tbeir competence. We assumed that friends'evaluations were based, for tbe most part, on past observations of par-ticipants' social accomplisbments in varied settings, involving a widerange of people, over an extended period of time. Tbe second investiga-tion related ASI scores to a multiple-act criterion measure consistingof a wide variety of social outcomes determined to be indicative ofaffinity-seeking competence. Botb studies were designed to be consis-tent witb tbe assumption tbat tbe fairest test of a measure of socialpredispositions is its ability to predict tbe enactment of a set of beba\aorsover time. Tbe results of eacb indicate tbat respondents' reports reflectsocial abilities.

A final determination of tbe usefulness of tbe ASI must await addi-tional studies of its psycbometric properties. Tbe refinement of tbis


instrument will necessitate at least five kinds of investigations. First,additional studies of the relationship of the ASI to conversationalbehavior are needed. Second, people presumably differ in theirawareness of their own affinity-seeking skills and their motivation toseek affinity. These individual differences should be explored. Third,the relationship of affinity-seeking competence to interpersonal evalua-tions should be assessed. Fourth, it may prove valuable to compare thismeasure of performance skills to Bell and Dalj''s (1984) procedure forassessing atffinity-seeking knowledge to determine if the two approachesconverge and diverge in expected ways. Finally, it will be necessary toexamine the relative importance of the affinity-seeking competence andstrategic performance dimensions in varying situations. An attemptshould be made to identify the characteristics of those situations andinterpersonal relationships in which each takes on significance.

Although the probability that one individual will be attracted toanother may be enhanced by relatively static factors such as attitudesimilairity and physical attractiveness, the case for conceiving of interper-sonal attraction as a communication accomplishment is compelling. Thisperspective requires close examination of affinity-seeking competence.We have sought to facilitate such examination by providing a workablemethodologj' for assessing skills in this important realm of social life.

ENDNOTE

*The authors express appreciation to Professors John A. Daly and Margaret L. McLaughlinfor their comments on an earlier draft of this report. Professor Daly is also thanked forhis assistance in the collection of data for the second study.

REFERENCES

Argj'le, M. (1972). The psychology of interpersonal behavior (2nd ed.). Harmondsworth;Penguin.

Argyle, M., & Henderson, M. (1984). The rules of friendship. Journal of Social and Per-sonal Relationships, 1, 211-237.

Bell. R. A., & Daly, J. A. (1984). The afTinity-seeking function of communication. Com-munication Monographs, 51, 91-115.

Berscheid, E., & Walster, E. (1974). Physical attractiveness. In L. Berkowitz (Ed.), Ad-vances in experimental social psychology (Vol. 7, pp. 158-215). New York: AcademicPress.

Bochner, A. P. (1984). The functions of human communication in interpersonal bonding.In C. C. Arnold and J. W. Bowers (Eds.), Handbook of rhetorical and communicationtheory (pp. 544-621). Newton, Mass.: Allyn and Bacon, Inc.

Briggs, S. R., Cheek, J. M., & Buss. A. H. (1980). An analysis of the self-monitoring scale.Journal of Personality and Social Psychology, 38, 679-686.

Byrne, D. E. (19611. Interpersonal attraction and attitude similarity. Journal of Abnor-mal and Social Psychology, 62, 713-715.

Byrne, D. E. (1971). The attraction paradigm. New York: Academic Press.Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral

Research, I, 245-276.

Winter 1987 17

Cegala, D. J. (1981). Interaction involvement: A cognitive dimension of communicativecompetence. Communication Education, 30, 109-121.

Cheek, J. M., & Buss, A. H. (19811. Shyness and sociability. Journal of Personality andSocial Psychology, 41, 330-339.

Cohen. J., & Cohen, P. (1975). Applied multiple regre.^siorvcorrelation analyst'! for thebehavioral sciences. Hillsdale, N.J.: John Wiley and Sons.

Cook, M. (1977). The social skill model and interpersonal attraction. In S. Duck (Ed.), Theoryand practice, in interpersonal attraction (pp. 319-338). New York: Academic Press.

Crowne, D. P., & Marlow, D. (1964). The approval motive: Studies in evaluative dependenceNew York: Wiley.

Daly, J. A. (1978). Communication apprehension and behavior: Applying a multiple actcriteria. Human Communication Research, 4, 208-216.

Davis, M. S. (1973). Intimate relations. New York: Free Press.Dean, L. R. (1958). Interaction, reported and observed: The case of one local union. Human

Organization, 17, 36-44.Duck, S. (Ed.). (1977). Theory and practice m interpersonal attraction.. New York: Academic

Press.Fishbein, M., & Ajzen, I. (1974). .'Vttitudes towards objects as predictors of single and multi-

ple behavioral criteria. Psychological Review, SI, 59-64.Fishbein. M.. &. Ajzen, I. (1975). Belief attitude, intention and behavior An introduction

to theory and research. Reading, Mass.: Addison-Wesley.Gergen, K. J. (1982). Toward transformation in social knowledge. New York:

Springer-Verlag.Harre. R. (1977). Friendship as an accomplishment: An ethogenic approach to social rela-

tionships. In S. Duck (Ed.), Theory and practice in interpersonal attraction (pp. 339.354).New York: Academic Press.

Huston, T. L. (Ed.). (1974). Foundations of interpersonal attraction. New- York: .•VcademicPress.

Jones, E. E. (1964). Ingratiation: A social psychological analysis. New York:AppletonCenturv'Crofts.

Kaiser. H. F. (1960). The application of electronic computers to factor analysis. Educa-tional and Psychological Measurement, 20, 141.151.

Lorr, M., & More, W. W. (1980). Four dimensions of assertiveness. Multivariate BehavioralResearch, 15, 127138.

McCroskey, J. C. (1981). Oral communication apprehensioru Reconceptualization and anew look at measurement Paper presented at the meeting of the Central States SpeechAssociation, Chicago.

Nahemow, L., & Lawton. M. P. (1975). Similarity and propinquity in friendship forma-tion. Journal of Personality and Social Psychology, 32, 205-213.

Pomazal, R. J. & Jaccard, J. (1976). An informational approach to altruistic behavior.Journal of Personality and Social Psychology. 35, 317-326.

Rosenberg, M. (1965). Society and the adolescent self image. Princeton, N.J.: PrincetonUniversity Press.

Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The Revised UCLA Loneliness Scale:Concurrent and discriminant validity evidence. Journal of Personality and SocialPsychology, 39. 472-480.

Snyder, M. (1974). Self-monitoring of expressive behavior. Journal of Personality and SocialPsychology, 30, 526-537.

Snyder, M. (1979). Self-monitoring processes. In L. Berkowitz CEd.), Advances in experimen-tal social psychology (Vo\. 12, pp. 85-128). New York: Academic Press.

Sunnafrank, M. J. (1983). Attitude similarity and interpersonal attraction in communica-tion processes: In pursuit of an ephemeral influence. Communication Monographs, 50,273-284.


Sunnafrank, M. J. (19851. Attitude similEirity and interpersonal attraction during earlycommunicative relationships: A research note on the generalizability of findings toopposite-sex relationships. The Western Journal of Speech Communication, 49, 73-80.

Sunnafrank, M. J., & Miller, G. R. (1981). The role of initial conversations in determin-ing attraction to similar and dissimilar strangers. Human Communication Research,S, 16-25.

interpersonal attraction as a communication accomplishment: …€¦ · interpersonal attraction as...

Documents