HI

1
I

IHHHIHI

I

 

141
480
THS_

20M

This is to certify that the
thesis entitled

The Effects of Liking and Obligation on Compliance

presented by

Allison 8. Shaw

has been accepted towards fulﬁllment
of the requirements for the

Master degree in Communication

 

 

j’ML‘M / 791. it;

Maj? Proféssor’ 3 Signature

1

12/12/08

 

Date

MSU is an Affirmative Acricn/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE ‘ DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/Dale0ue indd

THE EFFECTS OF LIKING AND OBLIGATION ON COMPLIANCE
By

Allison S. Shaw

A THESIS

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

MASTER OF ARTS
Communication

2008

ABSTRACT
THE EFFECTS OF LIKING ON OBLIGATION ON COMPLIANCE
By
Allison S. Shaw

In the compliance gaining literature multiple studies have established a relationship
between pregiving favors and compliance. Despite many attempts to investigate
intervening variables the mediating mechanism for this relationship remains unclear. In
an attempt to separate the two most commonly proposed mediators of the pregiving-
compliance relationship, liking and obligation, three experiments were designed. The
results indicated evidence of a liking X obligation interaction effect, suggesting that if an
inﬂuence target does not like the person providing the favor, no amount of obligation will
make them comply (r = .32). Implications of the ﬁndings for future compliance gaining

research are discussed.

ACKNOWLEDGEMENTS

First, and foremost, I would like to thank my advisor, Dr. Franklin J. Boster, and
my committee members, Dr. Timothy Levine and Dr. Toby Ten Eyck, for their
encouragement, and continued support during this rather lengthy process.

Secondly, I would like to thank Dr. Michael Kotowski without whose help this
project would have never gotten off the ground and Kyle Andrews without whose help I
would have no procedure! I would also like to thank all of my research assistants: Erin
Roach, Jenna Bender, Samantha Sharp, Shannon Morey, Michelle Cosier, Brandon
Strassberg, Scott Prajzner, Jessica McFadden, Jessica Denbow, Jonelle Pariseau, Becky
Elliot, Case Johnson, Lindsey Diggleman, Sarah Haskins, and Ron Christiansen without
whose help I would have no data!

Finally, I would like to thank my father, Douglas Shaw, whose support
throughout my lifetime has given me the ability to draft the following pages. I appreciate

everything you have done for me and only hope that I can one return in kind!

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... vi
INTRODUCTION .................................................................................... l
EXPERIMET 1 ........................................................................................ 4
Method ........................................................................................ 4
Subjects ............................................................................... 4
Design ................................................................................ 4
Procedure ............................................................................ 5
Instrumentation ...................................................................... 6
Pilot Study .................................................................................... 6
RESULTS ............................................................................................. 7
DISCUSSION ........................................................................................ 8
EXPERIMET 2 ........................................................................................ 9
Method ........................................................................................ 9
Subjects ............................................................................... 9
Design/Procedure ................................................................... 9
Instrumentation .................................................................... 10
Pilot Study ................................................................................... 10
RESULTS ............................................................................................ 11
DISCUSSION ................................ 12
EXPERIMET 3 ....................................................................................... l 3
Method ....................................................................................... 14
Subjects ............................................................................. 14
Design .............................................................................. 14
Procedure ........................................................................... l4
Instrumentation .................................................................... l 5
RESULTS ............................................................................................ 15
GENERAL DISCUSSION ........................................................................ 15
REFERENCES ...................................................................................... 25

iv

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

LIST OF TABLES

Descriptive Statistics of Pilot Study for Experiment I: Liking ......................... 19
Compliance Rates for Experiment I: Liking ............................................... 20
Descriptive Statistics of Pilot Study for Experiment 2: Obligation ..................... 2]
Compliance Rates for Experiment 2: Obligation .............................. 22
Compliance Rates for Experiment 3 .................................................... 23
Contrast Coefﬁcients for Experiment 3 ......................................... 24

INTRODUCTION

The extant literature demonstrates consistently that providing a favor to inﬂuence
target before soliciting their compliance increases the likelihood of the target’s
compliance with the solicitation (e.g., Greenberg & Frisch, 1972; Regan, 1971; Whatley,
Webster, Smith. & Rhodes, 1999; Goei, Lindsey, Boster, Skalski, & Bowman, 2003).
Why the relationship between pregiving favors and compliance exists remains unclear.
One plausible mediator of the relationship is that the target develops liking for those who
provide favors. A second, equally plausible mediator of the relationship is the target’s
development Of feelings of obligation toward those who give favors. Research has
provided conﬂicting evidence regarding which, if either, mediator has the stronger causal
force. For example, Goei et al. (2003) attempted a test Of the two mediators in question
and concluded liking to be the stronger mediator. Conversely, Regan (1971) performed a
similar test and found that Obligation was the stronger of the two.

Neither Of these two studies controlled liking and obligation independently within
a pregiving compliance-gaining interaction. As a result, little is known of the necessity of
either Obligation or liking in the success of pregiving as a compliance-gaining technique.
A clearer understanding of the factors mediating the relationship between pregiving
favors and compliance would allow for the design of compliance—gaining requests that
are adapted to speciﬁc compliance-gaining situations. Effective message adaptation in
turn increases the likelihood of Obtaining compliance in the given situation and makes for
a more efﬁcient social interchange. TO that end, this study controls obligation and liking
independently across three experiments so that the intermediary relationship between

pregiving and subsequent compliance can be observed.

Liking is characterized as positive affect that attracts persons toward objects. By
providing favors, an agent portrays themselves as generous and kind, which in turn
stimulates the target’s liking for the agent (Regan, 1971; Whatley et al., 1999). Goei et a1.
(2003) proposed that increased liking increases the likelihood of gaining compliance
because people are motivated to maintain positive relationships with those who are
generous and kind. This claim is consistent with other research demonstrating that people
tend to comply more with someone with whom they have positive relationships (Baron,
1971; Burger, Soroka, Gonzago, Murphy, & Somervell, 2001). Thus, compliance with an
agent’s request may in part be due to the target’s desire to maintain a positive relationship
with the agent.

Obligation, on the other hand, can be characterized as a motivation to perform a
task arising from an uncomfortable feeling of indebtedness. The provision of a favor
creates a sense of inequity in the relationship between the target and agent. Because the
perception of indebtedness is not pleasant, the target is motivated to act in a way that
restores equity (Walster, Walster, & Bershcheid. 1978). An opportunity to restore equity
in the compliance-gaining scenario is provided by the agent’s compliance-gaining
request. Gouldner (1960) explicated this rationale as the norm of reciprocity, positing that
in most situations people should, and do, help those who have helped them. Therefore,
compliance with an agent’s request following receiving a favor may result, at least in
part, from feelings of obligation in the target.

If the presentation of a favor produces both liking and obligation in the target, it is
important to assess the impact of each on compliance. The ability to separate liking and

obligation from one another and then examine compliance rates makes it possible to

determine if both liking and obligation are necessary to impact compliance with the
agent’s request, or if only one Of the two is necessary. For instance, if an agent performs a
favor for a target, the target may experience both liking and Obligation toward the
inﬂuencing agent. If, however, the agent performs a favor for the target but feelings of
Obligation are inhibited from developing, the compliant response will arise from the
target’s liking for the agent. One could imagine this circumstance occurring when the
agent performs a favor for the target but it is a favor that the agent would have done for
anybody, a favor performed by accident, or the agent does not believe they are providing
a favor, thereby not giving rise to feelings of indebtedness (Greenberg et al.. 1972).
Alternatively, if the agent performs a favor for the target and feelings of liking are
inhibited, compliance will arise from the target’s feelings of obligation toward the agent.
This situation could arise if the agent displays undesirable personality characteristics. On
the other hand, inducing both obligation and liking in the target might be necessary to
elicit a compliant response. To disentangle these possibilities, the impact of liking and
obligation on compliance must be observed by comparing compliance rates arising in
situations where Obligation and liking are controlled systematically against situations in
which Obligation and liking are left to vary naturally as a result of the pregiving favor.
TO this end, this study controls obligation by varying if the target perceive that the
agent feels they have performed a favor or, alternatively, by making the favor rather
large. In one instance the favor is performed for the subjects but the agent does not
believe that they are performing a favor, whereas in the other condition, the favor has
inconvenienced the agent and therefore is perceived as much larger. Concurrently, liking

is controlled by varying the agent’s behavior toward a third party after the presentation of

the favor. Speciﬁcally, the agent acts rudely or in a likeable manner toward the third
party. The rude behavior is designed to create the impression that the agent is a surly and
insensitive person, and hence, not likeable. On the other hand the likeable behavior
presents the impression of a caring and considerate person. These scenarios are compared
to a control condition in which the same favor is performed, but with obligation and
liking left to vary freely.
Experiment 1

Method

Subjects. The sample consisted of 60 peOple enrolled in undergraduate
communication classes at a large Midwestern university. Subjects (53) were offered extra
credit in their class in exchange for participation in the study. Each S was assigned
randomly to one of three experimental conditions with the constraint that an equal
number of 83 were assigned to each condition. The sample consisted of 23 males and 37
females ranging in age from 18 to 27 years old with a mean age Of 19.42 years old.

Design. The experiment was a single factor independent groups design with
control, no liking and liking conditions. Compliant behavior was examined after the
presentation ofa pregiving favor. In the control condition, Ss arrived at a room where
they were told they could ﬁnd an experimenter who would assist them in completing the
ﬁnal component of a study in which they were participating. Upon arriving Ss found a
person arranging a collection of books on a book shelf. The person. a research
confederate (C 1 ), notiﬁed the S that the experiment had been relocated and that they
would help the S ﬁnd the new location, saying, “I ’m sorry, but the experiment has

changed rooms. [t is downstairs in the...actually I will just Show you were it is. it is kind

of confusing. Follow me. " As the S and C1 arrived at the new location, the S delivered a
direct request, saying, “I need to get those books put away by my next class and it is
taking longer than I had expected. You don’t have to answer now, but when you are done
here can you help me organize those books for a little while? If you can, just come back
up when you are done here.”

The no liking and liking conditions replicated the same basic design of the control
condition with each including one additional comment made by the Cl. In all three
conditions, upon arriving with the S at the new location, a second research confederate
(C2) exited the room just as the C l and S arrived. In the control condition the C1 collided
with C2, but the bump was only acknowledged nonverbally. In the no liking condition,
the C1 collided with the C2 and said, in a hostile manner, “Watch where you are going,
bitch.” In the liking condition, the same collision occurred but the C1 said, in an
apologetic and considerate manner, “I ’m so sorry. Are you okay?"

Procedure. 33 signed up to participate in a study described as an investigation of
how information travels within social networks. 83 arrived at a small auxiliary
communication library located on the fourth ﬂoor of a campus academic building where
they were expecting to ﬁnd an experimenter. Instead, the C 1 was in the library sorting
through a pile of books and placing the books onto shelves. Because the S was instructed
to arrive at this room, it was typically assumed that the Cl was the experimenter. Hence,
the S usually made initial contact by announcing they were there for the experiment. If
the S did not say anything upon entering the room, the Cl solicited information by
asking, “Can I help you? ’ to which most 83 replied, “I am here for the experiment.” It

was at this point that the Cl informed the S that the experiment had been moved and was

now in a new location. The C1 then lead the S to the new location, ceasing all interaction
between themselves and the S until arrival at the new location.

The new location was a communication research laboratory located in the
basement Of the same building. While walking down the hall to the laboratory the C1
delivered the request for compliance. By the time the request was completed, the C1 and
S had arrived at the laboratory, just as the C2 was walking out. The C 1 and C2 collided
into one another, upon which the C1 delivered the control, no liking, or liking induction
before leaving to return to the fourth ﬂoor library.

Waiting in the laboratory was an experimenter (E). The E welcomed 83 to the lab
and asked them to take a seat. The E administered a consent form and then provided the S
with a packet Of two self-report measures. The ﬁrst measure consisted of demographic
questions for use in the present experiment. The second measure was composed of ﬁller
items irrelevant to this study. After 85 completed the two measures they were given a
debrieﬁng form, after which the E solicited questions and then dismissed the S.

Meanwhile, the C1 waited in the fourth ﬂoor library for the S to return. 83 were
given 10 minutes after completing the ﬁller items to return to the library before they were
considered non-compliant. All 85 were debriefed according to IRB protocol.

Instrumentation. Compliance was assessed by recording whether or not the S
returned to the library to help the C1 after completing the ﬁller items. 85 were also
debriefed in accordance with IRB protocol once the study was completed.

Pilot Study
Prior to the start of the study, a pilot study was conducted to examine the efﬁcacy

of the liking inductions. The pilot study was composed of hypothetical scenarios

mirroring the conditions Of the experiment. Pilot study 85, who were not permitted to
participate in the actual experiment, read a scenario and responded to several items
comprising liking and Obligation indices. A one-way omnibus ANOVA demonstrated a
statistically signiﬁcant difference between the control, liking, and no liking conditions for
the liking dependent measure (F(2, 49) = 11.96, p < .01, 132 = .33). Trend analysis
revealed substantially more liking was generated by the liking induction than by the
control, which generated more liking than the no liking induction (t(l , 49) = 4.69, p <
.01, r = .30) (see Table l for means and standard deviations). Because previous research
has also identiﬁed Obligation as a mediating mechanism in the pregiving and compliance
relationship, 85 also completed an index of obligation items. Another one-way ANOVA
was conducted in order to ensure that obligation was not activated by the liking
induction. Results of this analysis found no signiﬁcant differences in obligation across
conditions (F(2, 49) = 2.19, p = 0.12, 132 = .08). See Table l for means and standard
deviations.
Results

A preliminary analysis found that there was no statistically signiﬁcant effect for
sex on compliance (x20 , N = 60) = 0.03, ,p = 0.86, r = .02). In addition, analyses were
run in order to investigate if there were confederate effects. NO statistically signiﬁcant
differences were found for the ﬁrst confederate (12(5, N = 60) = 3.54, p = 0.62), the
second confederate (,1/(4, N = 60) = 4.14, p = 0.39), or the experimenter (78(3, N = 60) =
4.37, p = .22).

NO statistically signiﬁcant difference in compliance was found when comparing

the three conditions (78(2, N = 60) = 5.25, p = 0.07). Compliance in the liking condition.

\J

however, was substantially higher than in the pregiving control condition (980, N = 40) =
5.02, p < .05, r = .35, Odds Ratio (OR) = 4.50), but not the no like condition 060, N =
40) = 2.66, p = 0.10, r = .26, OR = 3.00). Furthermore, the no liking condition was not
different from the pregiving control condition (380, N = 40) = 0.40, p = 0.53, r = .03, OR
= .75). In addition, a statistically signiﬁcant difference was found when the pregiving
control and the no liking condition were pooled and compared to the liking condition
(x20, N = 60) = 4.85, p < .05, r = .28, OR = 3.67) Frequencies used for the chi-square
analysis can be found in Table 2.
Discussion

This study demonstrated that inducing liking can lead to a greater rate of
compliance when people are presented with a pregiving favor than when presented with
the favor alone. This effect is so strong that if one were to use a liking-inducing message,
it would increase compliance approximately 4.5 times that Of performing a pregiving
favor alone. However, the effect of disliking was not as predicted. It is unclear exactly
why there was increased compliance in the disliking condition, when it was expected to
be lower than the liking condition or the pregiving control. It may be the case that a
different mechanism (e. g., obligation) works to increase compliance when liking is not
present. It is unclear from these ﬁndings whether obligation could also be at work in this
causal model. Therefore, a second experiment was performed in order to investigate how
obligation may affect the pregiving-compliance relationship

In addition, although not formally documented, it seemed to be the case that those
in the no liking condition, in which the C1 was rude toward the C2, Often laughed after

the delivery of the condition message. It is plausible that $3 in this condition found the

insult to be humorous, as opposed to being rude. Cohen, Nisbett, Bowdle. and Schwarz

(1996) did ﬁnd that persons from Northern states found a similar insult amusing. It is

possible that subjects in this study also found the insult to be humorous and laughed.
Experiment 2

Experiment I produced some evidence of an effect Of liking on compliance rates.
The effect obligation has on compliance rates is still unclear. however. Therefore,
Experiment 2 was designed in order to examine this relationship. The procedures and
design Of the study were carried out in a fashion similar to Experiment 1 in order to
eliminate any methodological explanations for differences between the two experiments.
Method

Subjects. Similar to Experiment 1, there were 60 Ss recruited from undergraduate
communication courses at a large Midwestern university who participated in exchange
for extra credit. Each S was assigned randomly without replacement to one of three
conditions. The sample consisted of 24 males and 36 females; other demographic
information, such as age or race, was not obtained because there were no differences
found in Experiment 1. Undergraduates at the university as a whole are predominately
Caucasian, with 8% African American, 5% Asian/Paciﬁc Islander, 3% Hispanic, and 1%
Native American.

Design/Procedure. The design and procedure Of this experiment was similar in all
aspects to Experiment 1 except for the messages in each condition. A single factor
independent groups design with control, no Obligation, and obligation messages was
used. In the reduced Obligation condition, the Cl walked the S down the hall to the

laboratory and delivered the message. “I am glad you came when you did. I really needed

a break.” Conversely, in the induced Obligation condition the C1 said, “I ’ve had to show
people where the room has been changed all day and I haven’t been able to get any of my
work done.” In both conditions, the initial message was followed by the request, “I need
to get those books put away by my next class and it is taking longer than I had expected.
You don ’t have to answer now, but when you are done here can you help me organize
those books for a little while? If you can, just come back up when you are done here.”
This request was the only message delivered in the control condition. All other
procedures of Experiment 2 are the same as Experiment 1.

Instrumentation. Similar to Experiment I, compliance was assessed by recording
whether or not the S returned to the library to help the C1 after completing the ﬁller
items. The only demographic information recorded in this study was the S’s sex. Similar
to Experiment 1, all 33 were debriefed according to IRB protocol.

Pilot Study

A pilot study was performed prior to beginning data collection for Experiment 2
in order to determine the effectiveness of the Obligation inductions. A set of 83 who had
not participated in any of the other portions of this study were used for this study. These
Ss were presented with written scenarios that paralleled the experimental inductions. A
similar analysis was performed for this pilot study as was done in Experiment 1. A one-
way AN OVA demonstrated that the mean obligation scores are within sampling error of
zero (F (2, 45) = 0.67, p = 0.52, 132 = .08). A linear contrast was performed in order to
examine if hypothesized conditions were in fact different from one another; however, no
statistically signiﬁcant differences were found (t(1, 42) = 1.13. p = 0.26). Although there

was not a signiﬁcant difference between these three conditions, the means were in the

10

expected direction and there was a relatively large effect size, r = .16, suggesting that
with a larger sample differences may be more apparent (see Table 3 for means and
standard deviations).

Because liking is known to mediate the pregiving and compliance relationship,
levels of liking were measured in addition to levels of Obligation. A one-way ANOVA
was conducted in order to test if liking varied as a function of the Obligation induction.
The results indicated no statistically signiﬁcant differences for liking between the
Obligation conditions (F(2, 45) = 0.90, p = 0.41. 132 = .04). See Table 3 for means and
standard deviations.

Results

Preliminary analysis found that there was no statistically signiﬁcant sex effect on
compliance (x20 , N = 60) = 0.95, p = 0.33, r = .12). Additional analyses examined
confederate effects and found no statistically Signiﬁcant confederate effects for
confederate two (x2(4, N = 60) = 4.14, p = 0.38) or experimenter effect on compliance
(x2(3, N = 60) = 4.37, p = 0.22). A statistically signiﬁcant effect for confederate one on
compliance was found, however (750, N = 60) = 6.16, p < .05). This effect may be due to
an unequal number of subjects and conditions collected by each of the confederates. One
confederate collected more than twice as many subjects and due to the conditions being
randomly assigned to each confederate this particular confederate also collected more of
some conditions rather than others. Furthermore, there was a rather modest effect size, r
= .05, indicating that there was little effect of confederate one. In addition, further
analyses found that neither of the confederates nor the experimenter signiﬁcantly

interacted with any of the conditions.

11

Chi-square analysis found no statistically signiﬁcant differences among the three
experimental conditions (780, N = 60) = 3.53, p = 0.17). Further analysis found no
statistically signiﬁcant differences between the no Obligation and the control (78(1, N =
40) = 2.56, p = 0.11, r = 0.25, OR = 2.85) or the Obligation and the control condition
(380, N = 40) = 2.56,p = 0.11. r = 0.25, OR = 2.85). Null effects outcomes were also
Obtained comparing each experimental condition with the pooled data from the other
experimental condition and control group. A comparison between obligation and the
pooled ﬁndings of the no obligation and control condition yielded no statistically
signiﬁcant effects (780, N = 60) = 0.88, p = 0.35. r = 0.12, OR = 1.73). In addition, a
comparison between no obligation and the pooled ﬁndings of the obligation and control
conditions also did not yield statistically signiﬁcant ﬁndings (x2( 1 , N = 60) = 0.88, p =
0.35, r = 0.12, OR = 1.73). Table 4 presents the pertinent frequency data.

Discussion

The ﬁndings of Experiment 2 were mixed. Although the induced obligation
message did not increase compliance from the control at conventionally accepted levels
Of statistical signiﬁcance, it did yield a modest effect, OR = 2.85. This result suggests that
if obligation messages were employed, they would increase the compliance rate by
almost three times that of a control message. Although these ﬁndings are consistent with
the hypothesis that Obligation could increase compliance, the ﬁndings of the no
obligation message make it difﬁcult to draw any deﬁnitive conclusions. Similar to the no
liking message, the no Obligation condition seemed to have a reverse effect than
hypothesized and actually increased compliance over the control, although the effect was

insufﬁciently strong to reject the null hypothesis at conventional levels of statistical

signiﬁcance. Thus, it was comparable to the obligation message in rate of compliance. It
is unclear why in the no obligation condition subjects complied at a rate comparable to
the induced obligation condition. It may be the case that there is no effect for obligation.
In order to investigate these results further and in order to investigate if obligation and
liking interact with one another, a third and ﬁnal study was conducted.

Experiment 3

The Experiment 1 results suggest that there may be an effect for liking, and the
Experiment 2 results indicate no effect for Obligation. These ﬁndings seem to support
previous research that has demonstrated that there is an effect for liking, while there is no
effect for obligation (Goei et al., 2003). Nevertheless, previous theory and research claim
that there is an obligation effect on compliance after a presentation of a favor (Gouldner,
1960; Regan, 1971). It is possible that the effect of obligation is only apparent when it
interacts with liking. Therefore, a third experiment was designed that varied both liking
and obligation simultaneously.

In addition, the ﬁrst two experiments differ in an important way, which may
explain the difference in compliance rates between the two studies. In Experiment 1 the
liking induction occurred after the request for compliance; whereas, in Experiment 2 the
obligation induction occurred before the request for compliance. This order confound was
eliminated in the third study by having the request follow both the liking and Obligation
inductions.

This experiment has the added advantage of allowing the non-additive effect Of

liking and obligation on compliance, if it exists, to be estimated. Put another way, it

13

allows the investigator to examine the possibility that the effect of liking on compliance
is contingent on obligation, or the reverse.
Method

Subjects. The sample consisted of 44 people enrolled in undergraduate
communication classes at a large Midwestem university. Ss were offered extra credit in
their class in exchange for participation. Each S was assigned randomly to one of four
experimental conditions with the constraint that an equal number of Ss were assigned to
each condition. The sample consisted of 18 males and 26 females.

Design. This experiment was a 2 (liking, no liking) x 2 (obligation, no obligation)
independent groups factorial design. Compliant behavior was examined after the
presentation of a pregiving favor. The liking induction was worded, “I’m so sorry! Are
you okay?” The no liking condition message was modiﬁed slightly to, “Watch where you
are going!” Obligation was induced with the following message, “I’ve had to show
people where the room has been changed all day and I haven 't been able to get any of my
work done.” The no obligation message was, “I am glad you came when you did, I
needed a break.”

Procedure. The procedure was similar to that in Experiments 1 and 2, the only
exception being that the bump between the two Cs occurred just after the C1 and S exited
the elevator to the basement (where the second room was located). As the C1 rounded the
comer the C2 was waiting and ran into the C1. The liking messages were then delivered
as in Experiment 1. This delivery was then followed a few seconds later by the obligation
message and then ﬁnally the request, “I need to get those books put away by this evening

and it is taking longer than I anticipated. You don ’t have to answer now, but when you

14

are done here can you help me organize those books for a little while? If you can, just
come back up when you are done here.”

Instrumentation. Compliance was measured by recording whether or not the S
returned to the library to help the C1 after completing the ﬁller items as in the ﬁrst two
experiments. Similar to Experiments 1 and 2, Ss were debriefed according to IRB
protocol.

Results

Preliminary analyses investigated whether there were effects for confederates or
experimenters. Although all of the confederates and experimenters differed from each
other, these factors did not interact with the liking or Obligation inductions.

An ANOVA produced no evidence of a main effect for obligation (F ( 1 , 40) = .47,
p = 0.50, r = .12). However, a statistically signiﬁcant main effect for liking was found
(F (1 , 40) = 4.19, p < .05, r = .30) such that compliance was higher in the liking condition
than in the no liking condition (see Table 5). Furthermore, there was a statistically
signiﬁcant interaction between liking and obligation (F(], 40) = 4.19, p < .05, r = .30).

Upon examining the pattern of results in table 5 it appears that there is an effect
for liking when obligation is present, but no effect for liking when obligation is absent. In
order to test this effect a contrast analysis was performed, and the relevant contrast
coefﬁcients are presented in Table 6. An examination of condition means indicated that
this contrast model ﬁt the data well (F(], 40) = 4.53, p < .05. r = .32).

General Discussion
Previous research has debated which mechanism, liking or obligation, mediates

the pregiving favor-compliance relationship. In this paper three experiments were

15

designed to investigate the mediating mechanisms Of a pregiving favor on compliance.
Experiment 1 and 2 both found mixed ﬁndings for the effect of liking and obligation on
compliance. In Experiment I the inconclusive ﬁndings were due to problems with the no
liking treatment, in which some $3 laughed at the no liking induction. Mixed ﬁndings
were found in Experiment 2 as well; it was unclear why there was an equal compliance
rate in the no Obligation condition and‘the obligation condition, when it was hypothesized
that the no obligation condition would have the lowest rate of compliance. Therefore, a
third experiment was designed to remedy the problems of the previous experiments and
to examine if, and how, liking and obligation interact. The ﬁndings of Experiment 3
indicate a main effect for liking.

The interaction performed in the omnibus ANOVA did yield a signiﬁcant ﬁnding.
A contrast was performed in which the no liking, obligation condition was compared to
the liking, obligation condition, which found a signiﬁcant difference with a strong effect
of r = .32. This result indicates that the effect for liking is contingent on feelings of
obligation. Speciﬁcally, the effect for liking is strongest when there are accompanying
feelings of obligation. This supports Gouldner’s (1960) claim that states that the norm of
reciprocity is strong and prevalent, which has not been demonstrated satisfactorily in
previous work in the area of pregiving favors and compliance (Goei et al., 2003).
Speciﬁcally, this study demonstrated that obligation is important because of its impact on
increasing compliance when coupled with liking.

Another consistency among many pregiving studies is that the request for
compliance is a prosocial one. It may be that the norm of reciprocity extends to non-

prosocial requests as well. Obligation with a pregiving favor may function very

16

differently with an anti-social request than a prosocial request. To date, only one study
(Boster, Fediuk, & Kotowski, 2001) has examined how a pregiving request would
function when the request for compliance was anti-social (i.e., not reporting a confederate
who was cheating on an exam), and the authors report that participants do not feel a sense
of Obligation to comply with such a request. They suggest that the feelings of obligation
that arise from a pregiving favor are eliminated when the request for compliance is anti-
social, or that the pregiving favor is reinterpreted as a bribe and not a favor at all, which
would negate any effect a favor might induce. It is worthwhile for future research to
examine how pregiving favors and their mechanisms are affected by different types of
requests. Examining anti—social requests further and the cognitive processing of pregiving
favors may elucidate the norm of reciprocity further and give a richer understanding of
social interaction.

Although Gouldner (1960) states that the norm Of reciprocity is both strong and
prevalent, the present study notwithstanding, there has been considerable difﬁculty in
documenting this claim. The pregiving and compliance literature would beneﬁt from a
meta-analysis in order to examine the effect of obligation across studies and possible
moderators of the effect. It could be that the norm of reciprocity is not nearly as strong, or
prevalent, as Gouldner (1960) suggests. It may also be the case that human cognitive
development has implications for the norm of reciprocity. A large majority of pregiving
studies are performed on student samples, and it is possible that those in this age range
are not yet highly susceptible to normative pressure such as the norm of reciprocity.

Finally, mediators other than liking and Obligation have yet to be examined

carefully. Goei and Boster (2005) suggest that participants may comply with a request not

17

because of feelings of Obligation, but instead due to feelings of gratitude toward the favor
doer. Goei and Boster (2005) do in fact ﬁnd that participants have increased feelings Of
gratitude toward the favor doer, but do not have increased feelings of obligation toward
the favor doer. In addition, they found that gratitude has a positive effect on compliance.
whereas obligation did not. Additionally, Barlett and DeSteno (2006) found that gratitude
has an impact on prosocial compliance behavior that is distinctly different than general
positive affect. To date only these two studies have examined gratitude as a potential
mediator. There appears to be great potential for gratitude as a possible explanatory.
mechanism for pregiving and compliance that future research should continue to
examine.

Yet another possible mediator is reputation. It is possible that participants comply
with a pregiving request because not to comply would threaten their reputation. If one is
concerned about reputation one might engage in impression management strategies in
order to maintain a positive reputation. Not complying with a request after someone
performs a favor may lead to a negative evaluation by others. Therefore, in order to avoid
the negative evaluation, and possible negative social sanctions, one complies with such
requests. Taking an impression management approach to understanding how the
pregiving and compliance relationship is mediated might give new insight into human
interaction. Future research would beneﬁt greatly from examining the possibility and

effect of such a mediator.

l8

Table 1

Descriptive Statistics of Pilot Study for Experiment I .' Liking

 

 

 

Condition
Control No Liking Liking

Liking

Mean 1.75 0.39 2.17

SD 1 .05 l .40 0.75
Obligation

Mean 0.01 -0.44 0.54

SD 1.38 l .47 1 . 18

 

19

Table 2

Compliance Rates for Experiment 1: Liking

 

 

 

Condition
Compliance Control No Liking Liking
Complied 8 (40%) 10 (50%) 15 (75%)
Did Not Comply 12 (60%) 10 (50%) 5 (25%)

 

Table 3

Descriptive Statistics of Pilot Study for Experiment 2: Obligation

 

 

 

Condition
Control NO Obligation Obligation

Liking

Mean 0.12 -0.33 0.32

SD 1 .23 1 .87 1.75
Obligation

Mean 2.01 1.69 1.48

SD 1 .16 0.91 l .22

 

21

Table 4

Compliance Rates for Experiment 2: Obligation

 

 

 

Condition
Compliance Control No Obligation Obligation
Complied 9 (45%) 14 (70%) 14 (70%)
Did Not Comply ll (55%) 6 (30%) 6 (30%)

 

IQ
ls)

Table 5

Compliance Rates for Experiment 3

 

 

 

Message-First Message-
Second
Compliance NO Liking Liking NO Liking Liking
Obligation 0 (.00) 6 (.54) 5 (.45) 4 (.36)
NO Obligation 4 (.36) 4 (.36) 1 (.09) 5 (.45)

 

23

Table 6

Contrast C oefﬁcients for Experiment 3

 

Message-First (only)

 

 

Compliance NO Liking Liking
Obligation -l 1
No Obligation 0 0

 

24

References

Baron, RA. (1971). Behavioral effects of interpersonal attraction: Compliance with
requests from liked and disliked others. Psychonomic Science, 25, 325-326.

Boster, F .J, F ediuk, T.A., & Kotowski, MR. (2001). The effectiveness of an altruistic
appeal in the presence and absence Of favors. Communication Monographs, 68,
340-346.

Boster, F., Rodriguez, J ., Cruz, M., & Marshall, L. (1995). The relative effectiveness of a
direct request message and a pregiving message on friends and strangers.
Communication Research, 22, 475-484.

Burger, J .M., Soroka, S., Gonzago, K., Murphy, E., & Somervell, E. (2001). The effect of
ﬂeeting attraction on compliance to requests. Personality and Social Psychology
Bulletin, 27, 1578-1586.

Cohen, D., Nisbett, R.E., Bowdle, B.F., & Schwarz, N. (1996). Insult, aggression, and the
southern culture Of honor: An “experimental ethnography”. Journal of
Personality and Social Psychology, 70, 945-960.

Goei, R., Massi-Lindsey, L.L., Boster, F.J., Skalski, P.D., & Bowman, J .M. (2003). The
mediating roles of liking and Obligation on the relationship between favors and
compliance. Communication Research, 39, 178-197.

Gouldner, A.W. (1960). The norm of reciprocity: A preliminary statement. American
Sociological Review, 25, 161-1 78.

Greenberg, M.S., & Frisch, D.M. (1972). Effect of intentionality on willingness to
reciprocate a favor. Journal of Experimental Social Psychology, 8, 99-111.

Marwell, G., & Schmidt, DR. (1967). Dimensions of compliance-gaining behavior: An
empirical analysis. Sociometry, 30, 350-364.

Regan, D.T. (1971). Effects of a favor and liking on compliance. Journal of Experimental
Social Psychology, 7, 627-639.

Walster, E., Walster, G.W., & Bershcheid, E. (1978). Equity: Theory and Research.
Allyn and Bacon, Inc.

Whatley, M.A., Webster, J .M., Smith, R.H., & Rhodes, A. (1999). The effect of a favor
on public and private compliance: How internalized is the norm of reciprocity?
Basic and Applied Social Psychology, 21, 251-259.

25

               

     
 

TATE UN

melllllallll 1| ll 111111111111llllllm
3 0 6 2 5 I7 6

3 1293 0