......

 

, ”l?“ LIBRARY
.1 ”,4- Michigan §tate
” " University

This is to certify that the
thesis entitled

SORRY. YOU’RE JUST NOT MY TYPE: EXPLORING ROMANTIC
REJECTION IN COMPUTER-MEDIATED COMMUNICATION

presented by
STEPHANIE TOM TONG

has been accepted towards fulﬁllment
of the requirements for the

MA. degree in COMMUNICATION

 

 

.4 51% {3 W

/ ' Major Professor’s Signature
z

‘1’ (Stu V4“; ’2 can ’85

 

Date

MSU is an afﬁrmative-action, equal-opportunity employer

¢‘---t—.--

up.-.-.- -.-.—.—

 

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

 

JUN 2 6 73:2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 Kthroilec8Pres/CIRCIDaIeDue,indd

 

SORRY, YOU’RE JUST NOT MY TYPE: EXPLORING ROMANTIC REJECTION
IN COMPUTER-MEDIATED COMMUNICATION

By

Stephanie Tom Tong

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS
Department of Communication

2008

ABSTRACT

SORRY, YOU’RE JUST NOT MY TYPE: EXPLORING ROMANTIC REJECTION
IN COMPUTER-MEDIATED COMMUNICATION

By

Stephanie Tom Tong

Although online dating has increased in popularity, little is known about how online
daters respond to date requests, or how they handle the delivery of romantic rejection.
This research was designed to identify the types of linguistic strategies employed in
romantic date rejection. One hundred and ninety respondents were asked to compose
a message rejecting a romantic date request initiated by a member of the opposite sex.
Results showed that use of politeness strategies varied as a function of the
relationship between the requester and rejector (acquaintance vs. stranger) and the
type of media used (email vs. online dating messaging service). Additionally, it was
found that online daters take advantage of certain features provided by dating website
messaging services which allow new ways for rejectors to deliver refusals that were

not previously available in FtF communication or prior types of CMC.

Copyﬁghtby
STEPHANIE TOM TONG

2008

Acknowledgements
I am extremely grateful to my committee chair, Dr. Joe Walther, for his superior
guidance and mentorship throughout this project. Without his scholarly expertise or
extraordinary patience this research would never have achieved this level of academic
rigor or excellence; without his humor or encouragement, this project would not have
been as rewarding as it has been. Likewise, I owe thanks to my committee, Dr. Steve
McCornack and Dr. Nicole Ellison for their insight and attention to detail—everything

from theory and experimental design to misplaced commas.

I thank my family for allowing me to move away from California in pursuit of my
academic endeavors, never letting me forget that I am a daughter, a sister, and a
granddaughter. Thanks especially to my mother (a data coder, stats and SPSS consultant,
reviewer, and all-around great PT), my dad (the best ﬁnancial backer), and my brother

(best all-around goof ball).

I thank my Ohio family for their spirit, support, friendship, and love.

To Daniel: human-dictionary, grammarian, editor, chauffer, coach, cheerleader,

part-time data coder, and full-time best friend. Thank you for your support, your time,

your understanding, your love.

iv

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................... vii

CHAPTER 1

INTRODUCTION ............................................................................................. 1

CHAPTER 2

REJECTION MESSAGES ............................................................................... 4
Goals........ ........................................................................................... 4
Politeness in Rejection Messages ............................................. 7
Elements of Rejection Messages ........................................... 14
Reasons for Romantic Rejection ............................................ 16

CHAPTER 3

IMPLICATIONS FOR REJECTION IN CMC ENVIRONMENTS .............. 19
The Use of Asynchronous Media for Rejection .................................. 19
The Choice to Use CMC for Romantic Rejection ............................... 20
The Impact of CMC ............................................................................. 23
Rejection Choice in CMC and Social Distance ................................... 29

CHAPTER 4 '

METHOD... .................................................................................................... 33
Sample.................... ...................................................................... 33
Procedure.................. ..................................................................... 34
Design ........................................................................... 35
Independent Variables ......................................................................... 35
Manipulation Checks ........................................................................... 37
Content Analysis. . . . . . .. ....................................................................... 38
Dependent Variables ........................................................................... 40

CHAPTER 5

RESULTS... .. ................................................................................................. 40
Manipulation Checks” ....................................................... 40
Hypotheses... ...42
Gender and Media Use ........................................................................ 46

CHAPTER 6

DISCUSSION ............................................................................ 46

CHAPTER 7

CONCLUSION ........................................................................... 53

APPENDICES ............................................................................................ 55

REFERENCES ........................................................................ 66

LIST OF TABLES
Table 1: Politeness Strategies ..................................................... 39

Table 2: Reason Attributions ...................................................... 40

vi

Chapter 1: Introduction

Although computer-mediated communication (CMC) has a well-established
reputation for its use in a variety of task-oriented activities, researchers are coming to
understand how this medium functions in interpersonal relationships. Online dating and
online personals have become a popular practice for those trying to ﬁnd romantic
partners. An initial survey of the online public projected that as of 2006, 16 million adults
have “gone to an online dating website or other site where they can meet people online"
(Madden & Lenhart, 2006, p. 3). Searching for a potential romantic date can be a
dramatically different process when done on a computer from the comfort of one’s desk.
Increased convenience, control, and affordability are only some of the reasons individuals
utilize online matchmaking services (Ellison, Heino, & Gibbs, 2006). Lack of time to
peruse traditional settings (e.g. bars, clubs, etc.) for dates is one of the top reasons that
single adults turn to online dating (Close & Zinkhan, 2004; Whitty & Carr, 2006).

The present research investigates whether another reason online dating is
attractive to users is that it alters the process of rejections to date requests. Online dating
websites provide a venue for adults to view individual proﬁles of other daters. The
overall goal of these websites is to help users “meet” another person who is potentially
compatible in order to start a romantic relationship. The websites themselves are willing
to share many anecdotal stories documenting the “success’? of mixed-mode relationships
that began online (e.g., success.match.com) but less is known about the nature of online
romantic rejection. While researchers have focused on the self-presentation (Ellison et al.,
2006), characteristics and demographics (e. g., income, physical attractiveness, education

level, etc.), and interactions among online daters (Hitsch, Hortacsu, & Ariely, 2005;

Whitty & Carr, 2006), none have examined the process by which online daters reject
initial requests to meet. As online dating increases in popularity, it provides a new
environment in which to examine the messages people construct and their behaviors
when rejecting romantic date requests.

The process of romantic refusal can prove communicatively stressful. Oﬁen
delivering rejection means attending to several distinct, and sometimes competing, goals
at once (Johnson, 2007; Kline & Floyd, 1990). The rejector must construct a message that
conveys the refusal clearly, manage self—presentation concerns, and minimize the threat
to the target’s feelings. Furthermore, those who are acquainted with the target of their
rejection must mitigate the threats to the existing relationship. The fact that one simple
rejection may contain many a “face threatening act” (FTA) has prompted researchers to
examine how individuals prioritize and manage these FTAs when designing refusal
messages in different contexts ranging from interpersonal romantic rejection to
organizational rejection (Besson, Roloff, & Paulson, 1998; Johnson, 2007; Kunkel et al.,
2000; Saeki & O’Keefe, 1994).

Several reasons make online romantic rejection an important topic to research.
Because much of the previous research of rejection messages has been studied within FtF
contexts there may be unanticipated ways in which CMC inﬂuences the self-presentation
management of the rejector, the facework strategies used to mitigate the face threats to
the target, and the reasons given in rejection messages—particularly along the
attributional dimensions of locus, controllability, and stability. Additionally, the editing,
timing, and selective self-presentational capabilities found in different types of CMC can

help rejectors to construct their ideal refusal messages and their ideal selves. Those

rejectors who are concerned with the face needs of the target and are acquainted with the
target will be more likely to take advantage of the special features of online dating
websites.

A question raised by these communicative features is whether or not rejection
messages conveyed in dating websites contain the same elements as refusals delivered
through email. As CMC begins to expand in its application to the arena of romantic
dating, researchers must also explore how these technologies facilitate (or inhibit) one’s
ability to design a message to reject a romantic date request. Also, relational dynamics
can impact rejection message construction. Online rejectors may modify their refusals
based on their anticipation of future interaction with an acquainted target. Complicating
matters is the mixed-mode nature of such relationships and the different ways in which
offline and online contexts inﬂuence communication between individuals. However the
question of how different communication technologies and relational contexts inﬂuence
the content and delivery of romantic date refusals remains to be answered.

This paper will explore the argument that the type of relationship the requester
and rejector share, the pursuit of multiple communicative goals, and the type of CMC
they use to communicate all inﬂuence rejectors’ motivation for and construction of
messages used to refuse a date request. As background for this study, a literature review
is presented which contains an examination of the goal-driven behavior of romantic
rejectors and an analysis investigating the characteristics and themes found in refusal
messages and the conditions under which these elements can change. Theory regarding
the rejectors’ self-presentational concerns, the mitigating potential face threat of the

target, and the implication of the symbolic meaning embedded in media selection

follows. In addition, the results of an empirical study designed to examine the above
argument will be reported.
Chapter 2: Rejection Messages

Rejection message construction has been studied as compliance-
resistance and as overt refusal messages (Saeki & O’ Keefe, 1994). No matter how it is
conceptualized, there is an abundance of literature examining both the motivation for
performing refusals and their linguistic construction. Both of these elements will be
reviewed in the following sections.
Goals

As Kline and Floyd (1990) point out, a refusal message is a speech act designed to
fulﬁll multiple goals at once. Goals have been shown to be powerful motivations for
human interaction and behavior, especially in problematic or stress-inducing situations.
Goals have been deﬁned as “end states toward which individuals continually strive”
(Berger, 1995, p. 5). Research examining the pervasiveness of goals as motivations for
human communication has found that they are present in a variety of situations ranging
from compliance-gaining (Dillard & Burgoon, 1985; Hunter & Boster, 1987), to date
initiation (Berger & Bell, 1988; Mongeau, Serewicz, & Therrien, 2004), to refusal of
sexual initiation (Aﬁﬁ & Lee, 2000). In fact, goals are such important motivations for
communicative action that a lack of goal focus has been shown to lead to a chaotic or
unclear communicative episode (Berger, 1995). Goals themselves are constructs that
often compete with one another: Indeed, individuals can have multiple goals within one
communicative situation, making it difﬁcult to prioritize them and decide on a course of

action. Berger (1995) states that both research and theory pertaining to goal-oriented

behavior are based upon two assumptions, namely that “individuals pursue clearly
deﬁned goals in their interactions with others, and they frequently seek to achieve
multiple goals simultaneously” (p. 5).

Because of the multiple goal focus of many communicative situations, scholars
have tried to organize goals as having a hierarchical relationship to each other such that
goals are ordered cognitively in a step-wise fashion prioritizing the desired, overarching.
end-goal (sometimes called a “superordinate” goal). However, both Berger (1995) and
Dillard (1997) point out that goals and hierarchies are not static; they continually
ﬂuctuate in both importance and focus during a communicative episode and are
somewhat dependent upon time and context for their deﬁnition. Dillard (1997) suggests
that because individuals often seek to attain multiple goals within one situation that a
hierarchical structure of goals is more or less a simpliﬁcation of the actual cognitive
process of goal-directed communication. Dillard also notes that someone who is ending
an intimate relationship must attend to the “plausibility of lasting affection, the potential
for rejection, and the prospect of disharmony” (p. 55). Forcing a hierarchical structure
upon these goals makes for a simpliﬁcation of multiple motivations. It is useful to look at
the communicative context in which these goals are embedded to see which goal(s)
individuals eventually ascribe importance to and how that prioritization manifests itself in
their goal-driven communication (Mongeau et al., 2004).

As Dillard (1997) suggests, those who communicate refusals desire to achieve
multiple goals types: ﬁmctional and relational. The functional goal is simply conveying
the refusal clearly. The relational goals are (l) mitigating face threat of the target

(especially important for those who are acquainted with the target of rejection) after

directly opposing the target’s wants, while simultaneously maintaining one’s own desired
self-image by managing self-presentation. To accomplish the functional goal, the rejector
must ﬁrst recognize the target’s request and then convey his or her refusal to perform that
request. The substantive meaning of the rejection is based on the dyad’s shared
knowledge of the request, and the shared inferences that both parties make based on the
original request. Thus in order for the rejection to be “intelligible and legitimate” it must
attend to the issue(s) set forth in the request (Kline & Floyd, 1990). For example, the
rejection of the date request must somehow reference the previous utterance of the date
request (“I can’t go out on a date with you”) rather than some ambiguous or unresponsive
remark (“It’s cloudy today”). According to Berger (1997), the goal(s) that is considered
most important by the communicator will manifest itself in the end communicative result,
in this case message production.

Furthermore, Saeki and O’Keefe (1994) note that to best fulﬁll the functional goal
of a refusal, the message content must share the same frame of reference as the original
request by virtue of its being a complementary (or closely related) communicative action:
“An act of refusal or rejection is made and justiﬁed in relationship to the conditions that
make the original request legitimate, and so different ways of rejecting are seen as
elaborations of one or more issues raised by the request itself” (p 69). In order to obtain
the functional goal, the sender must construct his or her refusal in conjunction as the
reciprocal of the request message.

The rejector must also complete the second goal of the rejection message which is
to mitigate the face threat caused by the direct opposition to the target (Kline & Floyd,

1990). More precisely, to refuse is to disagree with the target’s desires which creates an

intrinsic threat to the target’s face wants. From the target’s perspective, a rejection is
thought of as an undesirable response, and the rejector must then handle the face threats
that such a response creates. According to Kline and Floyd (1990) the rejector ﬁrst must
recognize that the failure to comply with the requested action has created an obstacle
blocking the target’s wishes which constitutes a threat to the target’s negative face (one’s
desire to be autonomous and free from impediment). Secondly, the reﬁJsal signiﬁes to the
target that the rej ector simply does not value the target’s desires, which constitutes a
threat to positive face (the need to be accepted and appreciated by others, or the want to
be wanted). Because rejector’s communicative goals require the fulﬁllment of several
objectives, certain situational or contextual factors will impact goal prioritization, and
this impact will show itself in the features of the refusals they construct.
Politeness in Rejection Messages

Because the simultaneous pursuit of multiple goals can create greater uncertainty
in communication than the pursuit of singular goals (Berger, 1995), Brown and
Levinson’s (1987) politeness theory provides a useful theoretical framework with which
to examine the fulﬁllment of multiple goals within the refusal. Politeness theory
maintains that when the goals of facework and communicative clarity are in conﬂict,
simple message design becomes complicated. In an attempt to satisfy all communicative
goals, the speaker designs a longer, more convoluted message.

Johnson (2007) provides an instance of prior applications of politeness theory to
the study of refusal messages. She argues that politeness theory is uniquely well-suited in
its application to interactions such as refusal situations “where diverse interpersonal goals

are present” (p. 197). Johnson noted that how communicators attended to these goals

(e.g., dealing with several types of F TA, attending to one’s own and partner’s face needs,
inﬂuencing subsequent message content and perception) could potentially affect how
others viewed the appropriateness, effectiveness, or competence of such messages.
Although Johnson’s study did not examine the actual underlying differences in message
construction, she does provide rationale for the application of politeness theory to the
current research.

The desire to fulﬁll multiple goals forces communicators to compromise between
communicative efﬁciency and redressive action (or actions that “give face” to the target
in an attempt to repair the FTA present in the refusal). Such a compromise means that
individuals will vary their messages with respect to the amount and type of redressive
action they choose to employ. Politeness theory identiﬁes different strategies which
communicators can choose from when attempting to fulﬁll the relational and functional
goals of the rejection. Initially, the speaker must choose to perform or not perform the
FTA (i.e., refuse or accept a date request). If the speaker does perform the FTA, he or she
must choose to do so “on record” or “off record.” In essence, the rejector must choose
how explicit the refusal message should be. An off record refusal allows the rejector to
deny communicative intent. If one uses sarcasm or a joke (“You want to go out with Me?
Really? Ha, ha! ”) the meaning and intent are often indistinguishable, while an on record
refusal which is very clear in its communicative intent (“I can’t go out with you on
Saturday”).

The off record refusal is characterized by extreme indirectness, which allows the
speaker to deny the communicative intent behind the utterance, and forces the receiver to

make his or her own inferences from the speaker’s message. Off record messages hint at

the refusal, but are not explicit in their rejection in the same way an on record rejection
would be. While off record refusals allow the rejector to avoid committing to the speciﬁc
intent of the rejection, the extreme indirectness also renders the substantive meaning of
the refusal essentially negotiable. Forcing the requester to make the correct interpretation
based on the rejector’s utterance leaves much room for subjective interpretation and the
rejector risks failing to fulﬁll the functional goal of the refusal. As an example, the above
refusal of the joke neither conﬁrms nor disconﬁrms the rejector’s desire or ability to go
out on a date with the requester.

Prior research found that off record utterances were utilized much less oﬁen in
CMC than in FtF interaction. In his content analysis of an interactive teleconference,
Hiemstra (1982) found only one off record utterance from a transcript with a total of 71
utterances. This is an extreme difference in the politeness strategies observed by Brown
and Levinson who report off record utterances as “the most pervasive” of politeness
forms in natural, FtF conversations they examined. Hiemstra (1982) proposed that the
differences in use of off record utterances is due to “the pressure exerted by the computer
medium toward efficiency, the lack of nonverbal cues needed to interpret indirect or off-
record utterances and a task-related emphasis on efficiency” (p. 900).

Hiemstra’s (1982) claim that off record utterances are less prevalent in CMC due
to reduced nonverbal cues is certainly relevant: To convey meaning through an off record
utterance, a communicator (usually) must do so through the use of paralinguistic cues
such as vocalics, proxemics, or gestures. However, reduced cue environment of CMC
does not necessarily mean that interactants do not ever “go off record,” but rather adapt

the ambiguity of off record utterances and perform them differently than in FtF

interaction. For instance, in response to an online date request, an off record refusal could
be performed using an ambiguous statement (as in F tF encounters), but rejectors could
also choose not to respond. In this case, the absence of a response (or lack of
reciprocation) could be viewed as an ambiguous, off record rejection. Those rejectors
who choose to reject a request through silence can convey (indirectly) that they are not
interested in a date and simultaneously resist openly committing themselves to that
particular communicative intent. The lack of response also provides deniability in
communicative intent as a one could simply deny ever receiving the request in the ﬁrst
place. Thus this reconceptualization of the off record strategy to include the deliberate
ambiguity of a “no response” reﬁrsal would seem to contradict Hiemstra’s claim, and
instead assert that off record strategies could be more prevalent in CMC rather than less.
If an on record refusal is chosen, the rejector then has the choice to perform the
refusal baldly, without redress or with redress. Bald on record (BOR) refusals performed
without redressive action are as clear, concise, and direct as possible (Brown & Levinson,
1987). Because they prioritize communicative efﬁciency, BOR refusals fulﬁll only
functional goals and do not offer any redressive action to account for the threat to the
target’s face: “I can’t go out with you this weekend.” Alternatively, rejectors can choose
to perform the refusal on record, with redressive action. Depending on which aspect of
face is being emphasized, the rejector can deploy positive or negative politeness
strategies. Positive politeness appeals to the target’s desire to be socially accepted,
included, valued, and appreciated and could include such strategies as preceding the
rejection with an expression of appreciation or a compliment (exaggeration of interest in

target), using in-group language or identity markers, or asserting knowledge of the

10

requester’s wants (Brown & Levinson, 1987): “You're really sweet to ask me out, but I
can’t go out tonight. Let’s try another time, shall we?” Negative politeness includes
strategies that are applied to the requester’s threat to negative face, or the desire to be free
from impediment. An apology such as the following would be considered a negative
politeness strategy: “You want to go out tonight? Gee, I’m awfully sorry, but I don’t
really think I can.”

Brown and Levinson (1987) note that differences in power distance and social
distance may greatly impact strategy selection, message design, and delivery. Differences
in power and social distance underscore the role orientation that each individual
embodies with respect to the communicative environment in which they interact. Power
distance is described as a “vertical” continuum along which subordinates communicate
with their superiors or those with signiﬁcantly more status than themselves. However,
those individuals engaging in an interaction consisting of a date request and subsequent
refusal would be more likely to interact along the continuum of social distance which is a
“horizontal” relation. High differentials in social distance are exhibited among
individuals who are not familiar with each other, while low levels of social distance exist
between individuals who are friendly, casual, and familiar with each other (Morand &
Ocker, 2002).

It follows that those who are more familiar (or have less social distance) with the
target of their rejection message would be more concerned with fulﬁlling the relational
goals of the rejection message (i.e., maintaining the relationship with the target and
maintaining one’s desired self-image) than strangers who have no pre-existing

relationship with the target (high social distance pairs). As such, known acquaintances

11

who have a lower degree of social distance will still want to reduce the effects of their
rejection through use of politeness strategies, whereas relative strangers with high
degrees of social distance will be less motivated to do so. Consequently high social
distance communicators are more likely to produce messages BOR, which convey
“impersonality, formality, coldness” (Morand & Ocker, 2002, p. 7).

Prior research has found that social distance does impact romantic date goal
fulﬁllment. Mongeau et a1. (2004) found that relationship type (or degree of familiarity
between dating partners) had a signiﬁcant impact on ﬁrst date goals. Speciﬁcally, they
conceptualized relationship type on a ﬁrst date as a continuum with one end being no
prior acquaintanceship (as in a blind date) and the other end being a ﬁrst date between
good friends who have known each other for several years, but are just initiating a
romantic turning point in their relationship. Subjects were asked to imagine themselves
on a ﬁrst date with either a good friend or a mere acquaintance. Mongeau et a1. (2004)
found that those in the friend condition placed more importance on goals relating to
sexual activity and the investigation of the romantic potential of the relationship than
those in the acquaintance condition. It follows that relational type, or degree of social
distance, will also have an impact on other date-related communication between
acquainted or unknown individuals.

The degree of social distance that conversational partners have also raises
questions regarding how the complexities of a pre-existing acquaintanceship may affect
their expectation of anticipated future interaction. As Saeki and O’ Keefe (1994) suggest,
anticipated future interaction may impact how individuals craft their messages, but was

not considered initially by politeness theory:

12

However, their [Brown and Levinson] analysis presupposes a stable network of
acquaintance; they did not consider the consequences of being in versus not being in a
relationship with the other person. Quite apart from social distance, the anticipation of
future interaction with the target should lead to a higher priority on protecting the face
wants of the target, whereas the expectation of no future interaction should lead to a
lower priority given to face wants. (p. 72)

Within CMC contexts, research regarding anticipated future interaction has found
it to have an impact on interpersonal dynamics. Gibbs, Ellison, and Heino (2006) found
that those online daters who anticipated future interaction with others from the website
were more open and honest in their self-disclosures. as they were more likely to be “held
accountable” for such self-presentational claims. Walther (1994) found anticipated future
interaction to be a more powerful predictor of interpersonal communication behaviors
(intimacy, similarity, and trust) than media (CMC vs. FtF). Those that anticipated a long-
term relationship with group members in CMC and FtF were more affected by the
prospect of future interaction with their groups than those interacting in a “one-shot,”
short-term setting. However, while anticipation of future interaction was more varied in
CMC groups, anticipation among FtF group members remained unaffected by the
prolonged group membership. Speciﬁcally FtF groups maintained similar expectations of
anticipation regardless of short-term or long-term interaction, showing that such
anticipation is heightened among CMC interactants.

Thus for CMC dyads, anticipation of future interaction is expected to impact date
refusal messages depending on the type of relationship dyads have. Low social distance

dyads that are currently acquainted will likely have a higher expectation of future

13

interaction with the target of rejection. On the other hand, strangers will have no
expectations for future communication, and will reﬂect that in their refusals. Under these
conditions, anticipated future interaction is expected to have some impact on the way
individuals construct rejection messages. Those low social distance dyads that are
acquainted have greater cause to expect future interaction. and consequently will attend
to the relational goals of the refusal more intently than those who remain strangers.
Elements of Rejection Messages

Because managing FTAs, direct opposition, and communicative disharmony are
cognitively taxing, in order to ensure that they have attended to all the relevant goals,
rejectors often rely on standardized message designs that contain many of the same
elements. Presumably, rejectors are motivated to accomplish both functional relational
goals, but their motivation to do so is contingent upon the differences in social distance.
While the direct reﬁrsal is easy to create and deliver, the relational goals are more
difﬁcult to accomplish. Individuals will want to use a variety of facework and politeness
strategies to meet such goals, but may be hard pressed to generate them spontaneously
during a fast-paced communicative encounter.

Goffman (1967) asserted that to remedy this, senders may rely on predetermined
or ritualized discourse to reconcile the relational disruption the refusal has caused.
Besson, Roloff, and Paulson (1998) argue that “conventionalized” refusal messages
develop not because they help restore the social order but because these conventionalized
messages have proven to be effective in helping senders accomplish the necessary goals.
This habitual message framing allows individuals to anticipate the frequently occurring

clash of communicative goals mentioned above.

14

No matter the reason. several studies have examined the elements of such
ritualized rejection messages and have developed speciﬁc typologies accounting for their
shared elements. The most popular study of these typologies exists within an
organizational context such as a rejection for a promotion or job (Jablin & Krone, 1974;
Locker, 1995; Limaye, 2001; Saeki & O’Keefe, 1994). However, Besson et al. (1998)
point out several characteristics that exist in conventionalized date rejection messages
delivered in an interpersonal context. These elements serve to fulﬁll the instrumental goal
of conveying the refusal by direct opposition and elements of facework to mitigate the
threat posed to the target’s face wants:

(a) a statement that announces the sender carmot go out on the date, (b) an

apology for not granting the request, (c) an extension of appreciation for the

target’s request, (d) a reason for rejection, (e) an expression of concern for the

target’s potentially hurt feelings. (p. 186)

Besson et a1. (1998) describe (a) and (d) as elements which satisfy the
instrumental goal of conveying the refusal and identifying the reason for non-compliance,
and elements (b), (c), and (e) as facework attending to the threats of the target’s negative
and positive face wants. In addition to those elements mentioned above, rej ectors may
offer statements of encouragement to the requester (i.e., “giving face”), and suggestions
for future contact on a platonic level. While Besson et al.’s (1998) study was conducted
in a F tF context, it is expected that these elements will also be present in date refusal
messages delivered via CMC. Similarly to FtF settings, the differences in social distance

will change how these elements are utilized in online rejection message design:

15

H1: When rejectors communicate with the target of rejection using CMC, those

who are acquainted (low social distance) with the target design refusal messages

that contain more positive or negative politeness strategies (such as an apology,

an expression of appreciation for the request, an expression of concern or

sympathy for the requester’s feelings, expressions of encouragement, suggestions

for future relationship or contact, and longer length) than those rejectors who are

do not know (high social distance) the target.
With regard to refusal message design, it is the reason for rejection upon which many
targets focus their attention. Limaye (2001) has gone as far to suggest that the reason or
explanation for rejection is the single, most important part of the refusal message. A
review of the relevant literature reviewing reasons for romantic date rejection is
considered below.
Reasons for Romantic Rejection

Many targets of rejection feel the need to understand the cause of the rejection in
order to make a critical assessment of rejection episode (Folkes, 1982). Understanding
the importance of the reason for rejection, rejectors are motivated not only by their own
privately-held opinions about the target (which may have inﬂuenced their refusal in the
ﬁrst place) but also the desire to communicate the refusal, concern for threats to the
target’s face, and their own self-presentation.

Previous studies looking at the reasons that rejectors give for romantic date
refusals in FtF contexts suggest that they communicate reasons that vary along three
different attributional dimensions of locus, stability, and controllability (Weiner, 1979).

Locus refers speciﬁcally to the target of the rejection (rather than the rejector) and can be

16

internal (personal) or external (impersonal). F olkes’ (1982) analysis found that female
date rejectors rarely reported a reason with an internal locus (“You’re just not my type”)
preferring instead to use a reason with an external locus (“My cousins are in town this
weekend”). With an extemally-based rejection (e.g., undesirability of proposed activity or
unavailability) male rejectees may feel less dejected than when an intemally-based
rejection is used that points to a speciﬁc characteristic associated with the target. Folkes
(1982) reported that impersonal reasons were shown to cause less individual uncertainty
in the target than personal ones, presumably because the source of the rejection was
external rather than internal.

Stability refers to the rejection being perceived as “permanent and unchanging or
temporary and ﬂuctuating” (Folkes, 1982, p. 237). Using unstable reasons (“I have to
study for ﬁnals”) can leave the rejectee feeling a sense of opportunity since these
circumstances cannot last forever and allows the female rejector to maintain her desired
positive self-image. Contrastingly, a stable reason (“I can’t date you because I’m dating
someone else”) does not. In communicating an unstable reason, rejectors can manage
their self-presentation and mitigate the F TA to the target’s face.

Varying the controllability allows the rejector to either deny responsibility for
rejection or place that responsibility on the target of rejection. The woman may not have
control over her health (i.e. “I can’t go out with you because I’m sick”), but she does
have control over her choice of desired activities (“I can’t go outwith you because I don’t
want to miss my favorite TV show”). In the former, no blame can be ascribed to her;
however, the latter suggests that she has control over her own actions, but chose to reject

the date request. Alternatively, the woman can place the blame on the man making him

17

the cause of rejection, which decreases her responsibility. Thus if the man is presumably
in control over the reason for rejection such as sloppy appearance, then the fault is his
rather than hers. Both options allow the rejector to “avoid blame” because while
individuals can be blamed for negative events that they can control, they cannot be held
responsible for those occur outside of their control.

Folkes (1982) suggested varying these attributional dimensions allowed rejectors
to minimize the hurtful effects of the rejection while maximizing self—presentational
beneﬁts. The reasons for concealing or revealing attributional information may be
motivated by self-presentational concerns and concern for the other’s feelings. For
example, rejecting a date request because of the physical disabilities of the initiator would
damage the woman’s self-image by making her appear unfair and heartless while also
making the target of rejection feel dejected and helpless (F olkes, 1982).

Folkes also found that regardless of their privately held reason(s) for refusal,
female rejectors most often vary the locus of their reason by giving extemally-based
reasons for refusal to protect the face of the target and their own self-presentational
concerns. When surveying the male rejectees, Folkes found that these men conﬁrmed
receiving impersonal reasons for rejection most often. Thus the degree of social distance
that requester-rej ector dyad shares is expected to have an effect on the attributional
dimensions of the reason that rejectors will give in their refusals such that:

H2: When rejectors communicate with the target of rejection using CMC,

rejectors who are acquainted with the target (low social distance) will use reasons

with an external locus for rejection more than those rejectors who do not know the

18

target (high social distance), regardless of the locus of the privately-held reason

for rejection.
Although Folkes’ (1982) and Besson et al.’s (1998) analyses were conducted in F tF
contexts, they give rise to the assumption that communicative subtleties should also be
present within CMC rejections. New technologies have changed how self-presentation
occurs in CMC settings and may also change how the variations in attributional
dimensions function in the reasons that rejectors use in online rejection messages.

Chapter 3: Implications for Rejection in CMC Environments

Because rejection within an online environment has not yet been explored, it is
useful to investigate some background research to understand how the CMC context may
change or modify previous theoretical assumptions about rejection message design.
Implications of how the use of asynchronous media and the different CMC types can
affect message design are presented in the following sections.
The Use of Asynchronous Media for Rejection

Is rejection easier to deliver asynchronously? It certainly seems so given that
several institutions utilize the rejection letter as a form of refusal for job or student
applicants. The rejection letter has received much scholarly attention in organizational
literature, in both its usage and construction (Jablin & Krone, 1984; Limaye, 2001;
Locker, 1999). These studies have primarily focused on the formulaic elements of
rejection letters (such as the reason for rejection, use of buffers, greetings and closings,
perceptions of letter clarity, personalness, and appreciative tone) rather than their effects
on job applicants. However, these studies have documented the overwhelming use of

asynchronous media for interpersonal rejection tasks within organizational contexts.

19

a‘

The Choice to Use CMC for Romantic Rejection. The “cues ﬁltered out” (CFO;
Culnan & Markus, 1987) perspective suggests that reduced cue systems, increased
anonymity, and decreased social presence make CMC an undesirable medium for
interpersonal communication. However, when CMC is applied to interpersonal tasks,
some CF 0 theorists have argued that the increased anonymity present in asynchronous
media may promote a decreased amount of facework due to the lack of importance placed
on conventional maintenance of face (Hiltz & Turoff, 1978). If it is true that rejectors feel
“safer” behind a computer screen due to reduced social presence and the built-in buffer
that the medium provides, they might be less likely to “sugar coat” their message and
more likely to state their actual, privately-held reasons for rejection more explicitly and
with less redressive action (McGlone & Batchelor, 2003). As such, CMC refusals may
reﬂect rejectors’ prioritization of efﬁciency over politeness, even if the risk of performing
a F TA increases. Thus from a CFO perspective, the reduced attention to face needs
suggests that more bald, on record (BOR) rejections would be present in CMC, especially
for those dyads who share high social distance. Those who are not directly acquainted
would have less motivation to fulﬁll relational goals or perform any sort of redressive
action, and the heightened anonymity and decreased social presence in online interaction
only reduces the need to do so. This is also expected to be especially evident for those
rejectors who will deliberately choose CMC in an attempt to emphasize
depersonalization.

While the CFO perspective may apply to those communicative situations
involving high social distance dyads exploiting the impersonal nature of CMC, CFO may

be less applicable to low social distance dyads, or those using CMC in a more personal

20

manner. Those that are acquainted will be more likely to attend to the face needs of
others involved in the interaction, and will ﬁnd ways to do so, even in a reduced cue
environment. Hiemstra (1982) found that more than 75 percent of the utterances of an
interactive teleconference contained some type of linguistic facework. Additionally,
Morand and Ocker (2002) suggest that online communicators will continually adjust the
framing of their message to convey whatever degree of politeness and clarity they wish.
Thus far from being ill-suited to interpersonal communication as CFO theorists contend,
it can be argued that CMC users can fulﬁll the functional and relational goals of rejection
if they so choose. In fact, CMC may offer certain advantages over other channels, and
these advantages are expected to be important to low social distance dyads who are
concerned with relational as well as functional goals.

The hyperpersonal model of CMC (Walther, 1996; 2007) suggests that rejectors
can selectively self-present in CMC by taking advantage of editing. timing, effort, and
“the masking of involuntary cues” thereby inﬂuencing the impressions receivers make.
CMC’s heightened editing capabilities afford rejectors the ability to craft a detailed or
ambiguous, personal or impersonal message, and provide opportunities for the rejector to
frame the message according to his or her desires. In terms of message content, online
daters can minimize self-embarrassment through careful scrutiny and editing of their
online statements. CMC also provides rejectors with heightened interactional control over
their communications in cyberspace. By scrutinizing content, they can ensure that their
messages appear appreciative, polite, even apologetic, but also ﬁrm in their refusal. They

can control the timing of their messages, sending correspondence according to their

21

schedule. They can attend to the threats to target’s face by utilizing speciﬁc facework
strategies linguistically in CMC.

In recent work, many scholars have proposed new theoretical models which
depart from the CFO perspective by suggesting that CMC can certainly equal the
interpersonal involvement levels in FtF interaction, and (in some cases) can even surpass
those levels of involvement and create more emotional immediacy among interactants
(Walther, 1996). However, fewer theorists can account for those circumstances under
which interactants desire both interpersonal involvement and the communicative distance
that the CMC medium provides (see for example, Markus, 1994). Under certain
circumstances, individuals may deliberately desire to be involved and detached
simultaneously—a state that seems only to be achievable with a medium like CMC.

The above discussion sheds light on CMC’s ability to change the formulaic
construction of rejection messages and the elements which they contain. In this respect,
other theoretical frameworks that did not account for online communication may not
address all of the questions posed in this paper. It is possible that emphasizing different
aspects of CMC (e.g., control over timing, editing capabilities, etc.) elicit different effects
on the construction of the message or its impact on the target. In addition. each type of
CMC correspondence (email, online dating websites, IM, etc.) may also inﬂuence the
rejection process because of the socially constructed norms of use surrounding such
technologies. The types of CMC and their usage patterns may alter the theoretical

implications that CFO and the hyperpersonal model offer for online date request refusals.

22

The Impact of CMC

The type of CMC utilized in message delivery is expected to have a signiﬁcant
impact on message construction. 0’ Sullivan’s (2000) impression management model
points out the importance of the symbolic meaning that each type of communicative
channel conveys above and beyond its linguistic content. For example, one would expect
a colleague to fax over a ﬁnished work report, but does not expect the death of a family
member to be handled through the fax machine. Similarly, we would ﬁnd it strange if our
boss took out a television ad to remind us of a project deadline rather than sending a
reminder through email. However as O’ Sullivan notes, such norms can evolve between
interactants as to which channels to use for what purposes:

Whereas some individuals may consider electronic mail inappropriate for

anything but information exchange, others may view it as perfectly appropriate

for ﬂirting, expressions of emotion, or cybersex. The model proposes that the

interaction initiator’s perceptions of various channels’ symbolic meaning is [sic]

a. . .key consideration in channel selection. (p. 412)

It is worth noting that O’ Sullivan’s model does not take issue with transactional
communication and only considers channel selection from the sender’s point of view. As
such, he notes that the “initiator’s perceptions regarding symbolic meaning” are what
prompts media choice. Thus, if the requester initiates contact with the rejector via email,
presumably, the rejector would utilize the same medium to convey the reciprocating
speech act of the refusal. Regardless of whether or not the rejector believes another CMC
type would be more suited to the communicative task of rejection, he or she is expected

reciprocate medium choice because if it is true that the rejection is the reciprocal speech

act of the request message tied together by a common frame of reference, then the
rejector should use the same medium as the requester to complete the speech act (Saeki &
O’ Keefe, 1994). This constrains the rejector’s free choice of media, and such
restrictions are assumed to have an effect on how the message is constructed mainly
because each type of CMC has its own speciﬁc symbolic meaning and norms of use.
Although O’ Sullivan included several different types of communicative channels
in his model, this research is only examining differences in communication technologies
within CMC. Regardless, the notion of symbolic meaning remains salient among
different types of CMC technologies. Examining how users internalize the symbolic
meaning and norms of use surrounding media is one goal of this research. The social
constructivist approach to communication technology is one perspective with which to
explore how online interactants come to understand the functions and norms surrounding
CMC technologies. The several social constructivist theories have different underlying
aspects however they share a united theme: Social interaction creates shared meaning
among social actors such that the norms regarding the use of certain communication
technologies become mutually deﬁned and known (F ulk. Schmitz, & Ryu, 1995).
Following this theoretical perspective, the effects elicited by each type of CMC
depend upon norms that surround its use. Thus, contextualizing the study of the CMC
media by considering the social norms of use is an important part of understanding how
such media function in interpersonal rejection. These distinctions between personal and
impersonal asserted below are not characterizations of the media itself, but rather

distinctions as to how they are used in interpersonal communication. The media

24

themselves are neither personal nor impersonal. but they inﬂuence the processes under
which they can be conceived as such (Walther, 1996).

The purpose of online dating websites is speciﬁc: People use them to ﬁnd a
romantic connection amid the millions of participants involved in the website. As such, a
proﬁle on Match.com carries with it the expectation that several date requests may come
to the proﬁle-owner as a result of putting him or herself out on display. In this sense,
Match.com and other lntemet dating services similar to it are regarded as impersonal; one
in which relative strangers search the site and screen proﬁles for potential dates'.
Although an abundance of dating websites are available on the lntemet, the current
research focuses on the usage patterns of Match.com and other “see and screen” types of
online dating websites. The reason for this is that Match.com is one of the most widely
used dating websites. Practically speaking, it is advantageous to discover how individuals
will behave within Match.com (and other similar websites) as this information will have
implications for the over 15 million users worldwide (Match.com/howitworks).

The norms regarding email use differ with regard to communicative purposes.
While email can be used for a one-to-many distribution (as in a listserv), an individual
using email to send a date request intends to communicate with only one person. Because
email addresses are not usually included in an online dater’s proﬁle, they are not typically
used for personal correspondence between online daters. Thus if an email date request

were to be sent, this presumes that the requester must have taken the time and effort to

 

' It is important to note that not all online dating websites have this same “see and screen” type of format.
Some, such as eharmony.com make daters ﬁll out a lengthy questionnaire, and then match the participant to
another dater who ﬁts their “criteria” before either party initiates contact. These dating websites are not
considered in this research.

25

locate the email address of his or her target indicating a much more personal
correspondence than simply sending a message within the dating website itself.

Close and Zinkhan (2004) created a working deﬁnition of “dating netiquette”
based on their qualitative analysis of online daters’ perceptions of normative use and
found that the norms and etiquette regarding online dating are generally less stringent
than those found in FtF dating. Indeed, part of the appeal of online dating is the relative
anonymity associated with the medium: Users are able to “screen” or view several
proﬁles at once and send multiple request messages to those that interest them (Close &
Zinkhan, 2004). Furthermore Wang and Lu (2007) noted that this type of electronic
communication facilitates a one-to-many kind of communication “better serves
cyberdaters’ need for a greater quantity of information and enables them to multi-
communicate with several people at the same time” (p. 4).

Furthermore, online matchmaking websites have certain features that facilitate
impersonal communication. The use of “winking” in Match.com (also called “eye
contact” or a “collect call” in other dating websites) allows users to make contact with
those who interest them without extending any speciﬁc personal message. Sending a
wink generates a message sent to the recipient with the notiﬁcation that an interested
party has “winked at you” and a link to the sender’s proﬁle. The message content is pre-
written and generated by the website messaging system rather than the message sender,
and provides an impersonal ﬁrst attempt at initiating communication. A returned wink
signiﬁes mutual interest and invites continued escalation to a personal message sent
through the private channel provided by the website. Private messaging channels usually

function like email; however such correspondence is restricted to paying subscribers

(Fiore & Donath, 2004). Conversely, the use of impersonal “winking” seems to be
encouraged by Match.com as it is a free service for any person who has a proﬁle posted
on the website. Winks are therefore free to send to and receive from any other online
dater, whereas private messaging is not. Research has suggested that the relational
progression of media use in online dating starts with a “wink,” then moves on to a
personal message, and escalates from there to other types of media: “Users write each
other with the private messaging system to ﬁnd out whether they want to proceed to
contact via email, phone, or face-to-face interaction” (Fiore & Donath, 2004, p. 1397).
Thus a wink would be seen as the most impersonal, followed by a message sent within
the Match.com private messaging channel, followed by an email sent to a personal email
account (i.e. one that is not moderated or hosted by a third party), content aside.

In addition to the socially constructed usage patterns surrounding email and
online dating websites, certain mechanical features present in dating websites could also
change how receivers perceive rejection messages delivered using each medium.
Although both a messaging service and email are personal, one-to-one correspondences,
the fact that private email addresses are not typically posted content in online dating
proﬁles would make messages sent via email more personal (i.e., the sender has to search
for the receiver’s email address through some other means). If one receives a private
email from another online dater, it is perceived as more tailored, personal, or
individualistic (in comparison to website messaging) due to the increased effort needed to
locate the address. Though the use of personal email is not a communicative option
within Match.com, extending the comparison beyond the website, to consider other

messaging forms as well is not unreasonable. Because it lacks the more mechanical or

27

automated reply features of Match.com messaging systems and requires original content,
an email sent to a person’s private email account should be seen as the most personal
correspondence of all.

Thus the degree of personal ﬁt that different media afford (both alone and in
combination with the degree of social distance) should affect the reported reason for
refusal, the degree of facework, and the number and type of politeness strategies used in
the message. The increased impersonal aspects of dating websites allow rej ectors to
selectively attend to requests from those who interest them but reject those that do not.
Rejectors responding to a message in the Match.com channel may craft less polite
messages because of the high degree of anonymity and heightened aspects of
depersonalization that may be associated with the routinized and impersonal
communication that this type of CMC promotes. However, those responding to an email
message may be more likely to respond with a polite refusal to attend to the more
personal nature of the correspondence reﬂected in this medium and the increased
importance placed on face. As such a main effect for CMC type is hypothesized:

H3: Rejection messages delivered using email contain more politeness strategies

than rejection messages delivered via the Match.com messaging system.

CMC itself may change the reported reason for rejection offered in different
dyads. Given the existing research on date request rejection, it is known that rejectors will
often give a reason with an impersonal locus regardless of the fact that the actual,
privately held reason for rejection had a personal locus. Folkes (1982) suggested this
pattern was due to concerns for presentation of self and other’s feelings, but newer CMC

dynamics change how individuals manage such concerns. Additionally, using CMC may

28

provide a new reason with which to reject the requester (i.e., “this online dating proﬁle
wasn’t really my idea”, “This dating proﬁle is kind of old, my interests have changed”,
etc.) or even lying about receiving the date request altogether (i.e. “I must have deleted
your message! I’m sorry...”). Rejectors can put the blame on the use of media (an
external factor) rather than using an interally-based reason such as the requester’s
incompatability. Because of these new processes, it is expected that CMC will impact the
rejector’s disclosure of reasons under different media conditions; however, exact
predictions of effects are unknown. As such, the following research question is posed:

RQl: How does the type of CMC affect the publicly reported reason for date

rejection?
Rejection Choice in CMC and Social Distance

As online dating expands in popularity and usage, CMC users are beginning to
ﬁnd novel ways to address the communicative goals accompanying romantic rejection
that were not previously available in F tF situations. The choices that rejectors make
depend upon the degree of social distance that they share with the target of rejection and
the options that exist within the selected media of correspondence. On one hand, if
rejectors share a low degree of social distance with the target, it follows that they would
choose a rejection method that would allow the fulﬁllment of both relational and
functional goals. Alternatively, if social distance is greater and relational goals are less
important, rejectors may choose methods that focus only on functional goals. Another
option that online daters can use for refusal in CMC but not in FtF, is to remain

unresponsive. Each rejection option and the implications surrounding its use is explored

below.

29

Remain Unresponsive. When high social distance rejectors receive requests from
an unknown other, it is plausible that they may simply not respond. Requesters use online
dating websites to generate impersonal request messages (such as a “wink” or quick hello
message) have a higher chance of receiving more BOR rejections, or simply no response
to their online date requests, for two reasons. BOR rejections are expected because
attending to the competing functional and relational goals present in the refusal message
is cognitively taxing. As cognitive misers (Fiske & Taylor, 1991) communicators will
ﬁnd ways to minimize the amount of cognitive effort needed in message construction.
Thus strangers who have no relational goals to accomplish (i.e., preserve face, self-
presentation) have the opportunity to reduce their cognitive load by shifting their
attention to the functional goals of the refusal.

Secondly, the relatively high social distance coupled with the impersonality of
online dating websites suggests that increased volume of date requests from strangers
may lead some proﬁle-owners not to respond at all. In one context Limaye (2001)
suggests applicants who apply to organizations of their own volition have no right to
expect a reply since the applicants were the ones to initiate the communication: “In fact,
one common method of rejection is keeping silent; no news in such cases is bad
news. . ..if people decide to apply, they are expending their time, effort, and money
voluntarily without any external inducement and, as such, have no moral right to expect
reasons for rejection” (p. 103). While Limaye (2001) frames his analysis within an
organizational context, the idea remains the same: Those who choose to initiate contact

with a proﬁle-owner (who may expect but has not directly asked for such attentions) do

so at their own risk of being rejected.

30

The choice to disregard undesired attention becomes an enticing one to save time
and reduce cognitive load. Because such impersonal contact is so prevalent in online
dating sites, some users complain of being inundated with impersonal “winks” and
“collect calls”: “. . .some women are so deluged with collect calls that they indicate in
their proﬁles that they will not respond to them, often criticizing the stinginess of those
who do not pay for full privileges but still want a response” (F iore & Donath, 2004, p.
1397). The impersonal nature of online dating communication promotes these “winks”
and “collect calls” and (female) online daters can feel overwhelmed by advances from
strangers. In this case, they may exercise the Option to simply not respond to such
advances. Therefore the following hypothesis is proposed:

H4: Those receiving requests from a stranger (high degree of social distance)

utilizing the Match.com messaging system are less likely to construct a refusal

message with original content than those receiving requests from a stranger via
email.

One-Click Rejection. Another communicative choice for rejection in online dating
is the reply button called “Say ‘No, thanks’”. Match.com has streamlined the rejection
process into one click of the mouse, making refusals easier to deliver. By clicking on the
“say no thanks” button, an automated message is generated and sent to the requester. The
presence of this option may alter the process of online rejection and the likelihood of a
response bearing original content. Automated rejection messages would most likely be
utilized by high social distance individuals, or those who are looking for an efﬁcient way
to fulﬁll the functional goals of rejection but are not concerned with relational goals. In

contrast, low social distance rej ectors consider both goals to be important and are more

31

willing to spend a greater amount of effort to do so, thus reducing their preference to use
the “say no thanks” button. Given these assumptions, the following hypothesis is
proposed:

H5: Within the Match.com messaging system, those receiving a request from a

stranger (high social distance) utilize the “say no thanks” option more than those

receiving a request from an acquaintance (low social distance).

Tailored Rejection Messages. Although online daters have more options in how
they reﬁrse date requests compared to FtF daters, they are undoubtedly inﬂuenced by the
relationship they have with the target of rejection. Given that high social distance
rejectors in Match.com might be (a) more likely to remain unresponsive or (b) more
likely to respond using the one-click option, it is assumed that when they do choose to
respond, they will construct a refusal message with the fewest number of overall
politeness strategies when compared to low social distance rejectors in Match.com and
either type of dyad (acquainted or strangers) communicating via email.

The effect of social distance may be more subtle among email users. Email is a
much more personal medium of interaction, and a difference between the number of
strategies directed to either known or unknown targets of rejection may be minor.
However a greater number of politeness strategies may be used by low social distance
dyads who are focusing more closely on relational goal ﬁilﬁllment. The condition where
the most politeness strategies would be expected is among low social distance rejectors
using Match.com. Because other choices exist for Match.com users to (a) simply not
respond or (b) respond using the “say no thanks” option, when Match.com rejectors are

motivated to respond to the date request, they are prepared to expend cognitive effort and

32

compose a message that fulﬁlls both functional and relational goals. As such, these
messages must be tailored to the target in such a way that must not only complete
relational goals, but must also override the apparent impersonality of the Match.com
medimn. Thus the interaction of media type and social distance is expected to be much
greater in Match.com than email.

H6: Rejectors who refuse a date request from an acquaintance (low social

distance) in Match.com use more politeness strategies in their rejection messages

than rejectors using email with known targets, while strangers (high social

distance) communicating in email use more politeness strategies in their rejection

messages than rejectors using Match.com.

Chapter 4: Method

Sample

A sample of 199 undergraduate students from a large Midwestern university
voluntarily participated in this research in exchange for course credit. Only those who
had prior experience with both email and online dating were utilized in this research.
These restrictions were set in order to identify those participants who were familiar with
the technology and the social norms surrounding their use. Subjects answered standard
demographic items (age, gender), questions regarding email (number of accounts, amount
of time spent on email) and online dating use (number of dating websites visited, amount
of time spent in websites, and activity of membership). After removing those subjects
who failed to follow experimental instructions, 190 subjects remained in the sample.

Analyses revealed the sample sex (52 percent female) and age (M = 19.8, SD =

1.95, mode = 19). With regard to email, the average number of email accounts a given

33

subject subscribed to was 2.3 (SD = 0.97) (email account information is presented in
Appendix A). Time spent on email varied within the sample and is presented in
Appendix B. Online dating usage patterns showed that although only 4.2 percent of the
sample was actually comprised of active online daters at the time of data collection,
approximately 84.8 percent of subjects had previously been an active member of an
online dating website. Furthermore, an overwhelming majority of subjects (95.3 percent)
had visited at least one online dating website, (M = 1.49, SD = 1.06) (4.7 percent of the
sample failed to report a speciﬁc dating website they had visited in the past).
Approximately 30 percent of the sample reported either visiting or using Match.com
currently or in the past, making Match.com the frequently visited and used online dating
website (the reported website activity can be found in Appendix C). The amount of time
subjects reported having spent on online dating websites is presented in Appendix D.
Procedure

Participants (Ps) received a WWW address and were instructed to complete the
research on their own. Ps therefore were able to complete the tasks in a realistic
environment where they would normally use the lntemet. After providing their informed
consent, Ps clicked on a link which randomly redirected them to one of the sixteen
stimulus conditions containing a hypothetical scenario embedded with experimental
inductions (see Walther, Van Der Heide, Kim, Westerman, & Tong, 2008). They were
instructed to read the scenario and then proceed to the questionnaire by clicking another
link (examples of experimental stimuli and scenarios are presented in Appendix E).

Ps received the reason induction (the reason that they must reject the date

request). After reading the reason induction, Ps proceeded to the questionnaire section. Ps

34

indicated their likelihood of responding to the date request, and those in the Match.com
condition also indicated the likelihood of response using the automated “say no thanks”
reply option. Ps then responded to the date request by writing a rejection message:
“Please write out WORD FOR WORD exactly what you would say to respond to the date
request.” Ps were reminded that their rejection message would be communicated via the
same type of CMC the requester initially used. After composing their rejection message,
Ps reported any discrepancies between the privately-held reason for rejection given in the
scenario and the publicly-reported reason they utilized in their rejection message and any
motivations for those discrepancies. A written debrieﬁng occurred upon completion of
the experimental procedure.
Design

The experiment was a mixed groups, factorial design involving four independent
variables creating a 2 (gender of initiator: male or female) x 2 (CMC type: email or
match.com) x 2 (social distance: high or low) x 2 (reason for rejection: personal or
impersonal locus) design. Following Mongeau et al. (2004), participants were nested
within sex of initiator and were then randomly assigned the conditions of CMC type,
social distance, and the reason for rejection. They were told that the purpose of the
experiment was to explore how individuals communicate in romantic dating situations
using online technologies.
Independent Variables

Following Besson et al.’s (1998) and Mongeau et al.’s (2004) methodology,
participants received one of sixteen hypothetical scenarios in which they imagined

receiving a request for a romantic date from an individual of the opposite gender.

35

Manipulations of all four variables were embedded in the scenarios which were assigned
to subjects according to gender. The request message content was pre-written, and judged
on dimensions of realism by a group of college-aged raters before being introduced into
the current research.

Sex of rejector. Male and female participants were matched to scenarios featuring
date requests initiated by opposite-sex requestors. Previous research examining cross-sex
dating patterns found that date requests were primarily initiated by men and so
consequently focused more attention on the male requester-female rejector half of cross-
sex dyads (Berger & Bell, 1988; F olkes, 1982). However, more recent research suggests
that female date requests are becoming more common (Mongeau, Hale, Johnson, &
Hillis, 1993). Such research warrants the examination of both female and male rejectors
although no speciﬁc effects are hypothesized in this research.

CMC type. The date request message was embedded in either a Match.com or
email format. The email format was the academic email system format that all
participants were familiar with; however, future research could test whether or not
different types of email (commercial, academic, or corporate) addresses have an impact
on message construction.

Social distance. In the high social distance condition, participants were told that
the date request came from a complete stranger. Low social distance conditions describe
the request coming from an acquaintance that the participant supposedly knows as a
former classmate. In addition, the social distance induction was strengthened with the
addition of anticipated future interaction which was used to reﬂect acquaintanceship.

High social distance rejectors were told that the date requester had already graduated

36

from the university, thus the chances of seeing the requester in a common meeting place
were reduced (low anticipation of future interaction). Those in the low social distance
condition were told that the requester was still attending classes at the university, thus
increasing the probability of a chance meeting (high anticipation of future interaction).

Reason for rejection. The scenarios also provided one of two different reasons for
rejection. The personal locus condition indicated that the reason for refusal is because the
requester is a geek while the impersonal locus condition indicated the reason is because
the rejector ’s parents are coming into town. This initial reason was the privately held
reason for refusal, however participants were told that they could write any reason they
wished in the actual message, regardless of the reason given in the original scenario.
Manipulation Checks

Manipulation checks for social distance and anticipated future interaction were
conducted to assess the adequacy of experimental inductions. Both scales consisted of
three, 7-point semantic differential items. Participants indicated how familiar they felt
with the requester: “complete stranger/ well acquainted,” “best ﬁiends/ no relationship,”
and “not familiar at all/very familiar” (a = .75). Secondly, subjects’ anticipation of future
interaction with the requester was assessed using items from Walther (1994). On 7-point
scales, subjects answered the following items reﬂecting their expectations of future
interaction “anticipate/ would not anticipate future interaction,” “expectation of future
interaction whether you would want to or not,” and “to what extent would you like to
communicate with requester again” (a = .85) (Survey items can be found in Appendix F).

To check the reason for rejection manipulation, subjects were asked to (a) identify

the reason for rejection given in the scenario, (b) compare this reason to the one they used

37

in the refusal they constructed, and (c) then report any discrepancies between the scenario
reason and the reason they used in their actual message in an open ended response. The
reasons subjects used in their actual refusals were then compared to the original induction
to see if subjects were aware of the initial reason induction.

Content Analysis

A content analysis was conducted on the rejection messages created by those
participants who indicated they would respond to the request message. Following
procedures used by Roloff and Janiszewski (1989) and Besson et al. (1998), coding was
executed in a two-step process. First, two coders worked independently to unitize the
written discourse into idea units. Coders exhibited high unitizing relability (male
subjects, average agreement = .81, Guetzkow’s U = .003; female subjects, average
agreement = .85, Guetzkow’s U = .003). All disagreements between coders were resolved
by a third coder. After unitizing, two more coders (blind to hypotheses) categorized each
unit as one of ﬁve politeness strategy categories deﬁned in Table 1.

All disagreements were resolved by a third coder. Inter-coder reliability was
acceptable for female subjects (average agreement = .80, Scott’s pi = .76) and male
subjects (average agreement = .81, Scott’s pi = .78). Lastly, the reason clauses were
further coded to identify the attribution used in the rejection message. Reasons were
coded as containing either an internal, external, or ambiguous attribution. Inter-coder
reliability was acceptable (average agreement = .98, Scott’s pi = .94). Deﬁnitions of

reason attributions are listed in Table 2.

38

Table 1

 

 

Politeness Strategies
Codes Deﬁnition Exemplars
Rejection Statements directly refusing the date “I just can’t go out

Reason

Apologies

Appreciation

Concern

Encouragement

Future Contact

request

Statements that give reason for why
respondent can’t go on date

Statements indicating apology

Statements of appreciation for
(a) the message or date request
(b) complimenting the respondent

Expressions of concern for requester’s
potentially hurt feelings

Statements that bolster requester’s
self-conﬁdence in the face of rejection

Suggestions for future contact or
relationship between respondent
and requester

this weekend”

“I’m so busy with
work/ school.”

“My parents are in
town this weeken .”
“We’re just not that
compatible.”

“Sorry about this”
“I apologize.”

“Thanks for your
message.”

“I’m so ﬂattered that
you’re interested in
me.”

“Thanks for saying all
those nice things
about me.”

“This may be
too harsh, but...”

“You sound like a
great person.”

“I’m sure you can
ﬁnd someone else.”

“I hope we can
still be friends.”
“We should get to
know each other
better.”

 

Note. All exemplars are taken from actual data.

39

Dependent Variables

Overall length was computed by summing the total number of words utilized in
the entire rejection message. To determine the difference in the use of politeness
strategies, the total number of all strategies used in each rejection message was summed
and compared across conditions. Reason attributions were coded as internal, external or

ambiguous and were analyzed according to hypothesis 2.

 

 

Table 2

Reason Attributions

Code Deﬁnition Exemplars

Internal Those reasons in which the “Your interests are very
requester is responsible for the different than mine.”
rejection

External Those reasons in which certain “I’m currently seeing
circumstances prohibit the someone.”
respondent from accepting the “I’ve got lots of studying to
date request do this weekend.”

Ambiguous Reasons in which no clear “I’m really busy, but let

attribution can be inferred
from the respondent’s message

but let me check my
schedule and get back to
you.”

 

Note. All exemplars are taken from actual data. Ambiguous reason attributions were

dropped from the following analyses as they were not relevant to hypothesis testing.

Manipulation C hecks

Chapter 5: Results

A series of t-tests was used to test the adequacy of the social distance and

anticipated future interaction manipulations. Analysis revealed a signiﬁcant difference

between high and low social distance conditions, t (188) = 6.38, p < .001. Low social

40

distance rejectors (M = 2.95, SD = .88) reported feeling more acquainted with the
requester than high social distance rejectors (M = 2.12, SD =.9l). Analysis also showed a
signiﬁcant difference for anticipated and unexpected future interaction

t (188) = 8.79, p < .001. Consistent with expectations, low social distance interactants
reported a greater expectation of ﬁrture interaction with the requester (M = 4.18, SD =
1.28) than high social distance interactants (M = 2.53, SD = 1.19).

To check the reason of rejection manipulation, the initial reason given in the
scenario was compared to the actual reasons that subjects used in their messages. Results
showed that 48.9 percent of participants used the reason given in the scenario, 46.2
percent of participants used a different reason, but were able to identify and explain their
motivations for the discrepancy, and 4.9 of subjects used a different reason from the
original, but did not provide an explanation for their actions. Consistent with
expectations, 95 percent of the sample was able to recall the original reason for rejection
given in the scenario.

Due to the length and focal concerns, results from the reason of rejection
manipulation will be reported in a separate manuscript. However, to ensure that the
effects of this variable did not interact with the hypothesized factors, analysis of variance
was conducted on the total number of politeness strategies. Results showed no signiﬁcant
three-way interaction among reason type, media type, and social distance, F (1, 182) =
1.83, p = .18. A reduced model initially showed a signiﬁcant interaction between media
type and reason type F = (l, 186) = 9.98, p = .002. Post-hoe analysis determined that this
effect was due to a signiﬁcant difference within email although scores in the Match.com

conditions did not differ. These results will accompany a future manuscript.

41

Hypotheses

Hypothesis I predicted that low social distance acquaintances would use more
politeness strategies (apologies, appreciation, expressions of concern, encouraging
statements, suggestions for future contact, and longer length) in their rejection messages
than high social distance strangers. To test for differences in the length of messages, a
word count was calculated for each rejection message and an independent sample t-test
was performed to see if differences existed between high and low social distance
conditions. Analysis revealed that there were no differences in message length, t (188) =
.12, ns. Low social distance rejectors (M =66.62, SD = 33.54) did not signiﬁcantly differ
in the length of their messages from high social distance rejectors (M = 67.23, SD
=35.44).

To test overall use of politeness strategies, the total number of politeness
strategies used in each message was summed and compared across conditions. Data were
consistent with this hypothesis t (188) = 2.22, p = .01, n2 = .03. Low social distance
rejectors (M = 2.70, SD = 1.32) used signiﬁcantly more strategies than high social
distance rejectors (M = 2.11, SD = 1.29). To ﬁnd which of the politeness strategies was
most utilized by rejectors, individual chi-square tests were conducted to see if there were
differences among each politeness strategy. Results for statements of appreciation, x2 (1,
n = 175) = .02, ns, and concern for partner ’s feelings, x2 (l , n = 15) = .57, ns, indicated
that both low and high social distance rejectors used both strategies equally within the
sample of rejection messages. Surprisingly, apologies were more likely to be utilized by
high social distance rejectors, x2 (1, n = 59), = 8.32, p < .005, (p = .38. Low social

distance rejectors were more likely to make suggestions for future contact, )6 (1, n = 179)

42

= 4.92, p = .03, (p = .17. Additionally, it was found that anticipation of future interaction
was signiﬁcantly correlated with overall use of politeness strategies r (188) = .47, p <
.001. The more a rejector expected to potentially meet or interact with the target of
rejection in the ﬁrture, the more politeness strategies he or she used in the refusal
message.

Hypothesis 2 predicted that low social distance rejectors would utilize reasons
with external locus of rejection more than high social distance rejectors. Of the 187
reasons coded as either internally- or extemally-based, it was found that all subjects
overwhelmingly used external reasons for rejection more often than internal reasons, x2
(1, n = 187) = 120.2, p < .001, (p = .87, regardless of social distance conditions.
Examinations of the proportions of extemally-based reasons showed that in the low social
distance condition, 63 of the 91 reasons reported were extemally-based in this condition
(proportion = .69) while 78 of 96 reasons in the high social distance condition were
externally-based (proportion = .81). Contrary to hypothesis 2, a binomial test of
proportions revealed that high social distance rej ectors were signiﬁcantly more likely to
utilize extemally-based reasons for rejection than low social distance rejectors, z = -2.00,
p = .04.

Hypothesis 3 suggested that email rejections contain more politeness strategies
than rejections written using Match.com. An independent sample t-test showed that
Match.com (M =68.2, SD =33.93) and email (M = 65.8, SD = 35.0) users did not differ
signiﬁcantly in the length of their messages, t (188) = .49, ns. An independent sample t-
test did not reveal any differences between email and Match.com rej ectors when all

strategies were collapsed t (188) = .49, ns. However, there were differences in the use of

43

individual politeness strategies as revealed by a set of individual chi—square tests, which
were consistent with the hypothesis. Email users were signiﬁcantly more likely to utilize
apologies, x2 (1, n = 59) = 5.78, p = .02, (p = .31 and expressions of concern for the
requester ’s potentially hurt feelings, x2 (1, n = 14) = 4.57, p = .03. There were no
signiﬁcant differences between email and Match.com rejectors in the use of statements of
appreciation, x2 (1, n = 175) = .57, ns; statements of encouragement, x2 (l , n = 62) = .58,
ns; suggestions for ﬁtture contact )8 (1, n = 179) = .99, ns.

An independent sample t-test was used to test hypothesis 4 which predicted that
those individuals receiving date requests initiated by strangers would be less likely to
construct a refusal message in Match.com than in email. Data were not consistent with
this hypothesis, t (97) = .81, ns. Match.com rejectors (M = 3.70, SD = 1.82) and email
rejectors (M = 3.40, SD = 1.87) were equally likely to construct a refusal message with
original content.

Although hypothesis 4 revealed no differences in the likelihood of replies bearing
original content to unknown requesters in Match.com compared to email, results of
hypothesis 5 revealed that social distance manipulations were important predictors of
reply likelihood within the Match.com condition. Hypothesis 5 predicted that Match.com
users receiving date requests from strangers would be more likely to use the “say, no
thanks” button than those receiving date requests from acquaintances. Data were
consistent with this hypothesis t (91) = 1.97, p < .03, n2 = .04. Analysis showed that
Match.com users who were unacquainted with the date requester (M = 4.43, SD = 1.99)
were signiﬁcantly more likely to utilize the “say no thanks” button as a form of rejection

than those who were acquainted with the requester (M = 3.61 , SD = 1.94).

44

Hypothesis 6 predicted an interaction effect of media type and social distance on
the overall use of politeness strategies. As predicted, a planned contrast test revealed that
acquainted Match.com rejectors used the most politeness strategies (M = 3.00, SD = 1.16)
whereas acquainted email rejectors used more strategies (M = 2.45, SD = 1.35) when
compared to unacquainted email rejectors (M = 2.42, SD = 1.40), and lastly unacquainted
Match.com rejectors used the least amount of politeness strategies (M = 2.09, SD = 1.21),
t(186) = 2.99,p = .001.

The results for RQl showed that media type did make a difference in the reasons
rejectors used in their message. Although results showed that all rejectors were more
likely to use an external reason for rejection regardless of media type x2 (1, n = 187) =
5.36, p = .02, (p = .17, analysis of the proportions of extemally-based reasons showed that
for Match.com rejectors, 61 out of 90 reasons were externally-based (proportion = .73)
whereas for email rejectors 82 out of 97 were externally-based (proportion = .87). A
binomial test of proportions revealed this difference in proportions was signiﬁcant, 2 =
3.52, p < .001. Conversely, Match.com users had a higher proportion of internally-based
reasons for rejection (proportion = .27) than email rejectors (proportion = .13). The
binomial test of proportions also showed that this difference was signiﬁcant, 2 = 2.71, p =
.003. Thus media type did impact the publicly reported reason for rejection. A very small
percentage of the sample (5 percent) speciﬁcally used media as the reason for their
rejection, and thus was not large enough for statistical analysis (Results of hypotheses 1

through 6 and research question 1 can be found in Appendix G).

45

Gender and Media Use

Although no speciﬁc gender effects were hypothesized, analysis revealed that
men (M =4.27, SD = 1.71) were more likely to respond with rejection messages than
were women (M = 3.65, SD = 1.90), t (188) = 2.31, p = .02, 112 = .03. Additionally, some
differences were found in the use of speciﬁc politeness strategies. Individual chi-square
tests revealed that female rejectors were more likely to incorporate statements of
encouragement, x2 (1, n = 62) = 3.16, p = .08 and statements of appreciation for the date
request, 12 (l , n =174) = 11.17, p < .001 into their messages than male rejectors.
Regarding reasons for rejection, women were signiﬁcantly more likely to utilize
intemally-based reasons for rejection than men, )8 = (1, n = 34), 5.76, p =.02.

Chapter 6: Discussion

The purpose of this research was to determine how relational dynamics and media
can affect the use of politeness strategies in romantic rejection message construction. It
was assumed the degree of acquaintanceship—social distance—between requesters and
rejectors would be reﬂected in the types of politeness strategies used in romantic refusals.
It was also expected that the social norms surrounding the use of new media would alter
the rejection process when compared to other existing forms of CMC such as email.

Relational Effects. In line with prior research on the subject, low social distance
rejectors constructed refusals with a proportionately higher number of politeness
strategies overall than did high social distance rej ectors (hypothesis 1). This was largely
due to the fact that those acquainted with the target of rejection were more likely to

incorporate suggestions for ﬁxture contact into their refusals.

46

Subjects that experienced high expectations of future interaction with the target
integrated more politeness strategies into their messages. Some subjects in the low social
distance condition explained that they were motivated to be more polite by the possibility
of future contact: “First off I said I wasn’t interested because it is the truth, secondly
because i have to interact with this person in class, and i don’t want her to think im [sic]
an ass.” Social distance was also found to have an impact on the reasons rejectors used in
their messages, although not in the hypothesized direction. High social distance rejectors
used more external reasons for rejection while low social distance rejectors used more
intemally-based reasons (hypothesis 2). After analyzing subjects’ explanations regarding
their reason choice, it became clear that low social distance rejectors utilized internal
reasons more often to avoid any “false hope” or “leading on” that could occur in future
meetings with the requester:

I found his message in the ﬁrst paragraph some what creepy, saying he thought i

[sic] was “hot”. If a guy wrote that to me it would be a “no thanks” right away,

however since he wrote me a personal message and has been nice and an

acquaintance in the past I couldn't say no to everything without telling him the
truth.
Others stated that they would “let the individual down nicely” but would also “tell the
truth so there would be no misunderstandings,” or take care to “not hurt his feelings and
rejecting him so harshly after he put so much time into a nice note, but be ﬁrm—quite
ﬁrm—in my stance.” Results also revealed that all high social distance rejectors were

equally likely to respond to the target regardless of media type (hypothesis 4).

47

Media Effects. When Match.com rejectors did compose their own original
messages they didn’t differ signiﬁcantly ﬁom email users in their use of politeness
strategies (hypothesis 3). Although Match.com users reported that they would be equally
likely to respond to the target as email users, results indicate that they tend to choose
options that (a) are not extant in F tF encounters and (b) are not extant in email. These
results indicate that the social norms surrounding the use of new and existing forms of
media play a large role in the construction of polite rejection messages. It was found that
Match.com users receiving date requests from unknown others would rather respond
using the automated reply option than creating a tailored message with original content
(hypothesis 5). When given the option, online daters will take advantage of the “say, no
thanks” one-click rejection button in Match.com. In this case, the Match.com messaging
system and others like it have given online daters a new option with which to deliver
romantic rejection, one that seems to be very appealing for users as a way to reduce the
cognitive effort and stress of relational goal fulﬁllment.

Interaction of Relationship and Media. Anonymous Match.com users were less
likely to construct tailored rejection messages containing large numbers of politeness
strategies. The impersonal nature of Match.com in combination with their disregard of
relational goals prompted lower levels of politeness strategies. Contrastingly, low social
distance rejectors who were concerned with fulﬁllment of relational goals had to redouble
efforts to overcome the impersonal environment of Match.com, and produced the most
polite messages. Because email is a much more personal medium rejectors did not have

to work as hard as Match.com users to fulﬁll both relational and functional goals,

48

although low social distance rejectors were shown to use more politeness strategies
within email when compared to their high social distance counterparts (hypothesis 6).

The implications of results suggest a middle ground between prior CFO
assumptions, and more recent theory. Individuals who deliberately choose to use the one-
click rejection may have found an optimal way to balance the levels of interpersonal
involvement and relational separation. By weighing their concerns for the target’s
feelings and personal self-presentational needs against their desire to maximize cognitive
(and communicative) efﬁciency, rejectors may ﬁnd themselves attracted to one type of
media or rejection style versus another. Using the one-click rejection allows rejectors to
increase their involvement with the target (through their reciprocated response to the
request message), while simultaneously maintaining a sense of “detachment” (by virtue
of pre-generated content rather than an originally crafted message). The one-click refusal
provides a compromise between being too involved with the target (e. g., composing a
message with original content) and being too detached (simply not responding).
Rejectors may be choosing the one-click option because it allows them to execute their
primary goals without investing too much time, energy, or effort in the actual interaction
itself, creating a kind of “goldilocks refusal” that is “just right” for the fulﬁllment of
communicative, self-presentation, relational goals, and cognitive efﬁciency.

Media type was also shown to affect the publicly reported reason for rejection
(research question 1). Match.com users were more likely to use internally-based reasons,
while email users were more likely to use extemally-based reasons. This could be due to
the fact that because impersonal responses and automated messages are more widely used

within online dating messaging systems, when Match.com users do choose to compose an

49

original message, they believe it necessary to compose a message that contains a more
speciﬁc reason for rejection to allay any undesired future messages or contact from the
requester. Alternatively, the impersonal norms of use for Match.com may be reﬂected in
rejectors’ willingness to sacriﬁce the relational goals of self-presentation, relationship
concerns, or concern for other in an effort to fulﬁll the functional goals of rejection. In
such an impersonal environment of communication, users may feel that “sugar-coating”
the refusal with extemally-based reasons is unnecessary. Although it is known that online
daters take special care in the self-presentation of their online dating proﬁles to attract
attention from potential suitors, they might not be so careful in their communicative self-
presentation when rejecting or trying to avoid future contact with unwanted others.
Further research should examine the motivations and expectations behind online daters’
rejection reasons, and discover how messaging systems may impact those behaviors. It
would also be of interest to analyze the structure of both rejection and acceptance of
romantic interests and requests to see if communicative self-presentational styles differ
with respect to different functional and relational goals.

The results of this study suggest that new media provide new options for daters to
reject others, mainly (a) one-click automated rejection or (b) remain unresponsive. Online
daters are not only aware of these new communicative choices, but are using to use them
to simplify and streamline what can often be a difficult, painful, and stressful process of
romantic refusal. Automated rejection is an opportunity that many online daters take full
advantage of; indeed, it may be that part of the appeal of online dating websites is the

ease with which rejection can be delivered.

50

Although only a small number of rejectors used the actual media as their reason
for reﬁrsal, it seems that users do report the electronic media itself as a reasonable
“scapegoat” for their rejection, and remained unresponsive: “[sic] i left the box for a
response to Sam [the requester] blank, because I ...choose not to respond to her email, if
she were to bring it up in class, I would say I must have accidently [sic] deleted it.” In
this case (and others like it) rejectors would blame the medium as their reason for
refusal—something that cannot be done in FtF communication. Other users reported that
“realistically, I would not respond” or “I would not respond at all” which leads to another
option that is available in CMC but not F tF. The choice to remain unresponsive or
communicatively “unavailable” is easy to do in mediated communication, less so in F tF .

In addition, because of the selective editing offered by online communication it is
plausible that rejectors may create and use novel rejection strategies not found in FtF
rejection where communication is much faster and immediate. Future research should
examine the motivations for the reasons used by rejectors and how these refusal messages
affect the targets of rejection.

Limitations. The limitations of this study include the use of a convenience sample.
Only approximately 85 percent of the sample reported being active at some point with an
online dating service, leaving 15 percent who may be frequent visitors or “browsers” but
not active members. This means that these 30 subjects may never have actually sent or
exchanged online communications with any other daters. Although this may be true, it is
also notable that effects were found in spite of this limitation. It is unlikely that the trends
of the one-click rejection are speciﬁc only to this sample. A replication involving a more

representative sample of individuals who use online dating websites on a more regular

51

basis would most likely show the same (if not stronger) usage patterns of the one-click
rejection, rather than a different trend. This is due to the fact that as users become more
habituated to the procedures of online dating, they will ﬁnd ways to streamline the
rejection process by screening and rejecting others as quickly and efﬁciently as possible.
Although the most efﬁcient way to do this would be to simply delete the messages sent
by those they ﬁnd unattractive, it may be that as part of a community of online daters
familiar with the norms surrounding these websites, these rejectors know (a) the level of
civility expected by others and (b) how it feels to be on the receiving end of a refusal.
Lack of response would be too impolite, whereas a lengthy message with original content
would seem overbearing and odd. Thus the one-click rejection fulﬁlls the functional
goals of refusal, while also maintaining the optimal level of politeness expected by
others.

Another potential limitation of this study could be the date request message itself.
Before being introduced into the research, it was judged by a group of college-aged raters
and found to be appropriate. However, both the scenario and message had the potential to
seem overbearing or demanding, and for some, receiving such a request from a complete
stranger could seem threatening. Subjects did not seem to be concerned however, and out
of the 190 subjects, only three mentioned this in their response messages (one saying the
request was “creepy,” the other mentioning that it was “stalkerish,” and the last saying “I
guess we Spartans have a way of ﬁnding each other, but don’t worry, I’m not creeped out
or anything”). It is possible that others had similar reactions to the experimental stimuli

but did not express it. However, it did not seem to signiﬁcantly impact the results of this

study.

52

Lastly, previous research has shown that online relationships seem to follow a
very speciﬁc developmental trajectory in which dyads start with online correspondence,
followed by other media (phone, webcarn conferencing), and only then does it ever
progress to an ofﬂine meeting. Each of these stages could last an indeﬁnite amount of
time, as each couple evolves and develops at its own rate. Therefore, while it may seem
inappropriate to some for an individual to request an ofﬂine date so early in the
developmental cycle, others may ﬁnd it completely ordinary or normal. This research
focused solely on the initial stages of dyadic development, but it may be that following
the longitudinal development of an online couple would yield different rejection
processes (e. g., more concern for the target of rejection). Future research may be able to
demonstrate how rejection is delivered in the later stages of relational development.

Chapter 7: Conclusion

In conclusion, this research has uncovered some of the norms surrounding the use
of media in romantic rejection processes. This study contributes the ﬁndings that degrees
of acquaintanceship and type of media both have a substantial impact on the linguistic
composition of a romantic date rejection message. This study also showed that rejectors
use different reason types for rejection, but further research could uncover the
motivations involved in the selection of these reasons, and how media impact those
motives. This study has shown that automated rejection is a widely used (and arguably
new) practice among online daters. Because this is a new way to reﬁrse romantic
attentions, the effects of this type of refusal on the rejectee are unknown. Researchers
should examine whether receiving this type of rejection heightens or dulls the pain of

rejection from a recipient’s perspective.

53

As research continues to examine how the romantic rejection works it should also
focus on the way communication technologies alter these communication processes.
Clearly, the use of CMC for romantic relationships is on the rise and as it continues,
researchers must uncover how media change the way information is collected, conveyed,

and understood in the rejection process.

54

Appendix A

Subscribed Accounts other than Campus email

 

 

Email Account Frequency Percent of Sample
Gmail ‘ 37 19.4
Hotrnail 37 1 9.4
Yahoo! 55 28.8
MSN 2 1 .0
AOL 26 1 3.6
Comcast 6 3. 1
Other 5 2.6

 

Note. 22 individuals (11.5 percent of sample) indicated they only used their campus email

account and are not included in this table.

55

Time Spent on email Daily

Appendix B

 

 

Time Spent Frequency Percent of Sample
Less than 1 hour 85 44.5
1 hour 67 35.1
2 hours 30 15.7
3 hours 5 2.6
More than 3 hours 4 2.1

 

56

Appendix C

Dating Websites Visted by Subjects

 

 

Website Frequency Percent of Sample
Match.com 55 28.8
eharmony.com 23 1 2.0
AllMSU.com 53 27.7
Yahoo lpersonals 1 6 8 .4
chemistry.com 4 2. 1
Other 3 1 16.3

 

Note. Nine individuals (4.7 percent of the sample) did not report visiting a website.

57

Appendix D

Time Spent in Online Dating

 

 

Time Spent Frequency Percent of Sample
Less than 1 hour 94 49.2

1 hour 25 13.1

2 hours 13 6.8

3 hours 0 0.0

4 hours 0 0.0

More than 3 hours 29 15.2

 

Note. 29 (15.7 percent of the sample) individuals indicated that they were never paying
subscribers or active members of online dating websites, and had only visited an online

dating website to “browse.”

58

Appendix E

Example Scenarios

 

Sam is a student at MSU. You have a class together this semester, and you’ve talked to
him a few times—just casually about class. When he was absent from class last week,
you let him copy the midterm review sheet that your professor handed out. Although he
has a crush on you, he has never expressed it, nor has he ever asked you out. Recently,
however, he came across your proﬁle in the online dating website, Match.com, and has
decided to ask you out on a date for this weekend. Using the Match.com messaging
system, Sam sends the following date request message:

Lclick this link to see message]

Sam was a student at MSU, but graduated last year. You had a class together during his
last semester, but you never noticed him. Recently, however, Sam found your email
address in the MSU online directory. He decided to email you and ask you out on a date
for this weekend with the following date request message:

[clickthis link to see message]

Sam is a student at MSU. You have a class together this semester, and you’ve talked to
her a few times—just casually about class. When she was absent from class last week,
you let her copy the midterm review sheet that your professor handed out. Although she
has a crush on you, she has never expressed it, nor has she ever asked you out. Recently,
however, she came across your proﬁle in the online dating website, Match.com, and has
decided to ask you out on a date for this weekend. Using the Match.com messaging
system, Sam sends the following date request message:

[clig this IM) see message]

Sam was a student at MSU, but graduated last year. You had a class together during her
last semester, but you never noticed her. Recently, however, Sam joined the online dating
website, Match.com. She came across your proﬁle, and decided to ask you out on a date
this weekend. Using the Match.com messaging system, Sam sends the following date

request message:
[clicktthis link to see message]

59

Example Stimuli

W MSUGmsu.edu M65533:
> x [—3 2:3) ti; E”) MessageZof 10
‘ B

 

 

<
vets!

 

From: SamSpartan@gmail.com[Trust] [Block]
To: MSU@msu.edu
Date: 25 Jan 2008, 11:53:26 AM

Subject: Hey!

w.¢—-<.— . v... . .... .... 3....1....: ,ls.-l. n.l ._,_ —r-.,......~ -...Acr ~a._-_-“.-—.—-.‘.—-«—

 

Hi There,

I’m not sure if you remember me, but my name is Sam and we had a class together last
year. I’ve been getting ready to move out of my apartment since I graduated from MSU,
and I came across some old notes from that class we had together last year. It made me
think of you...and how I always thought you were kinda hot, but I was just too shy to say
anything! So I decided to look you up in the MSU email directory. I hope you don’t mind
me contacting you. If you don't remember me, I guess I should tell you a little about
myself.

I’m 23 years old and originally from central Michigan (Owosso specifically), but I’m still
living in Lansing even though I finished school. I have a few hobbies that I really like.
Recently I learned how to knit, and I'm into baking, especially cake decorating. But I
should probably cut back on that, since I eat too many of my own creations, which has
caused me to gain a few pounds. :) (My new year’s resolution was to lose about 10
pounds.) I have 2 cats named Farfalle and Tippy that I love a lot. I suppose I’m not really
much of a partier, and I have to admit that I’m not much of an athlete. I’m ashamed to
admit that never even went to a Spartan football game during my four years at MSU!

Since graduating, I’ve been looking for work, but haven’t found anything I like yet. But
I’m still in the East Lansing area, so I was wondering if you’d like to go out sometime?
Maybe this weekend? I'd really like to go out and maybe talk about what you’ve been up
to since last year. I look forward to hearing from you!

-Sam

 

 

 

60

 

q

My Match: Connections Winks Email Favorites Renmvedpnitiaets

 

Inbox 10 << back to Inbox

0 new ~.~
( ) Reply Cl - Say "No, thanks" Q Report a Concern , Delete 5
Sent 10
Trash 0 Forward to msu
From: samspartan mail

Don't let Date Received: January 25
your emails Subject: Hey!
get blocked

Forward

them to

your email

today! Hi There,

I'm not sure if you remember me, but my name is Sam and we had a class
together last year. I’ve been getting ready to move out of my apartment since I
graduated from MSU, and I came across some old notes from that class we had
together last year. It made me think of you...and how I always thought you were
kinda hot, but I was just too shy to say anything! Lucky for me, I just came across
your proﬁle, so I hope you don't mind me contacting you. I guess I should tell you
a little about myself.

I’m 23 years old and originally I’m from central Michigan (Owosso speciﬁcally), but
I'm still living in Lansing even though I ﬁnished school. I have a few hobbies thatI
really like. I love playing video games—mainly stuff like Worlds of Warcraft. But I
should probably cut back on that, since sitting so much in front of my computer has
caused me to gain a few pounds. :) I suppose I’m not really much of a partier, but
I like to sit in and watch movies. My favs are the Lord of the Rings trilogy, Jurassic
Park, Star Wars. I'm a bit of a sciﬁ buff—I really like Star Trek.

I have to admit that I’m not much of an athlete or a sports fan...I never even went
to a Spartan football game during my four years at MSU! Since graduating, I’ve
been looking for work, but haven't found anything I like yet. But I'm still in the East
Lansing area, so I was wondering if you’d like to go out sometime? Maybe this
weekend? I’d really like to go out and maybe talk about what you've been up to
since last year. I look forward to hearing from you!

-Sam

Reply O. , Say "No, thanks" 0 . Report a Concern ii: 312‘ . Delete Q

<< back to Inbox

61

Appendix F
Survey Items

What is the likelihood that you would reply to Sam’s message?

Not at all likely I 2 3 4 5 6 7 Extremely likely

How likely would you be to respond to Sam’s message by clicking on the “Say no
thanks” reply option?*
Not at all likely I 2 3 4 5 6 7 Extremely likely

Please write out WORD FOR WORD exactly what you would say to Sam in your
response message. You will be sending her a reply via the Match.com messaging system.

 

 

 

 

In this scenario, how well would you say you know Sam?
Complete Stranger 1 2 3 4 5 6 7 Well Acquainted
In this scenario, how would you describe your relationship with Sam?
Best Friends 1 2 3 4 5 6 7 No relationship
In this scenario, how familiar do you feel with Sam?
Not Familiar at all 1 2 3 4 5 6 7 Very familiar

To what extent would you anticipate you'll have future interaction with Sam in the near
future?

No Anticipation 1 2 3 4 5 6 7 Certain of Future Interaction
To what extent would you expect to interact with Sam again whether you want to or not?

No Anticipation 1 2 3 4 5 6 7 Certain of Future Interaction

To what extent would you want to communicate Sam again?

Do NOT want to communicate 1 2 3 4 5 6 7 Deﬁnitely want to Communicate

What would be the likelihood you could have a chance meeting with Sam somewhere?

Entirely Likely 1 2 3 4 5 6 7 Impossible

62

Think about the reason you gave to Sam as to why you couldn’t go out on a date. Was the
reason you wrote in your Match.com message the same reason we gave you (Sam is a
geek) in the original scenario?

Yes No

If you answered “no,” please explain how the reason you gave differed from the original
reason.

 

 

 

 

What was your motivation for using a different reason?

 

 

 

 

 

* Item used only in the Match.com condition

63

Appendix G

List of Hypotheses and Research Questions

 

H1: Low social distance pairs design rejection messages that contain more positive or
negative politeness strategies than high social distance pairs.

Results: H1 supported. Overall, low social distance rejectors used more politeness
strategies than high social distance rejectors t (189) = 2.22, p = .01, n2 = .03. Suggestions
for future contact were utilized more by low social distance rej ectors than high social
distance rejectors x2 (l, n = 179) = 4.92, p = .03, (p = .17. Anticipated future interaction
was positively correlated with use of politeness strategies r (188) = .47, p < .001.

H2: Low social distance rejectors will use externally-based reasons for rejection more
than those with high social distance, regardless of the locus of the privately-held reason
for rejection.

Results: H2 not supported. All rejectors were more likely to use externally-based reasons
for rejection, regardless of social distance x2 (1, n = 187) = 120.2, p < .001, (p = .87.
Binomial test of proportions revealed results opposite of the hypothesized direction: High
social distance rejectors were more likely to use extemally-based reasons for rejection
than low social distance rejectors, = -2.00, p = .04.

H3: Rejection messages delivered using email contain more politeness strategies than
rejection messages delivered via the Match.com messaging system.

Results: H3 not supported. However, individual chi-square tests were consistent with the
hypothesis showing that email rejectors were signiﬁcantly more likely to use

apologies, x2 (1, n = 59) = 5.78, p = .02, (p = .31 and expressions of concern for the

requester ’s potentially hurt feelings, x2 (1, n = 14) = 4.57, p = .03.

64

H4: Those individuals receiving date requests from strangers would be less likely to
construct a refusal message in Match.com than in email.

Results: H4 not supported, t (97) = .81, ns.

H5: Match.com users receiving date requests from strangers will use the “say no thanks”
button more frequently than those receiving date requests from acquaintances.

Results: H5 supported, t (91) = 1.97, p < .03, n2 = .04.

H6: Rejectors who refuse a date request from an acquaintance in Match.com use more
politeness strategies in their rejection messages than rejectors using email with known
targets, while strangers communicating in email use more politeness strategies in their
rejection messages than rejectors using Match.com

Results: H6 supported. Planned contrast test signiﬁcant in predicted direction, t (186) =
2.99, p = .001.

RQl: Media type will impact the locus of the reported reason for rejection.

Results: RQl showed differences in reported reasons for rejection due to media type.
Binomial test of proportions showed that signiﬁcant differences among the proportions of
Match.com rejectors’ use of intemally-based reasons over email rejectors, z = 2.71, p =

.003, and email rejectors’ use of externally based reasons over Match.com rejectors, z =

3.52, p < .001.

65

References

Aﬁﬁ, W.A & Lee, J. W. (2000). Balancing instrumental and identity goals in
relationships: The role of request directness and request persistence in the
selection of sexual resistance strategies. Communication Monographs, 67, 284-
305.

Ben-Ze’ev, A. (2004). Love online: Emotions and the Internet. Cambridge, England:
Cambridge University Press.

Berger, C. R. (1995). Inscrutable goals, uncertain plans, and the production of
communicative action. In C.R. Berger & M. Burgoon (Eds) Communication and
Social Inﬂuence Processes (pp. 1-28). East Lansing, MI: Michigan State
University Press.

Berger, C. R. (1997). Planning strategic interaction: Attaining goals through
communicative action. Hillsdale, NJ: Lawrence Erlbaum Associates.

Berger, C. R. & Bell, R. A. (1988). Plans and the initiation of social relationships.
Human Communication Research, 15, 217-235.

Besson, A. L., Roloff, M. E., & Paulson, G. D. (1998). Preserving face in refusal
situations. Communication Research, 25, 183-199.

Brown, P. & Levinson, S. C. (1987). Politeness: Some universals in language usage.
Cambridge: Cambridge University Press.

Close, A. G. & Zinkhan, G. M. (2004). Romance and the Internet: The e-mergence of
e-dating. In B. E. Kahn & M. F. Luce (Eds), Advances in Consumer Research
(pp. 153-157). Valdosta, GA: Association for Consumer Research.

Dillard, J .P. (1997). Explicating the goal construct: Tools for theorists. In J .O. Greene
(Ed.), Message production: Advances in communication theory (pp. 47-69).
Hillsdale, NJ: Erlbaum.

Ellison, N., Heino, R., & Gibbs, J. (2006). Managing impressions online: Self-
presentation process in the online dating environment. Journal of Computer-
Mediated Communication, 11 (2), article 2. Retrieved Feb 1 2006 from
http://jcmc.indiana.edu/voll 1/issue2/ellison.html

Fiore, A. T. & Donath, J. S. (2004). Online personals: An overview. Paper presented at

the 2004 Conference on Human Factors in Computing Systems, Vienna, Austria.
Retrieved October 2, 2007, from m://smg.media.mit.edu/papers/atf/chi2004

pgrsonals_ short. pdf

F iske, S. T & Taylor, S. E. (1991). Social cognition. New York: McGraw-Hill.

66

F olkes, V. S. (1982). Communicating the reasons for social rejection. Journal of
Experimental and Social Psychology, 18, 235-252.

Fulk, J ., Schmitz, J., & Ryu, D. (1995). Cognitive elements in the social construction
of communication technology. Management Communication Quarterly, 8, 259-
288.

Gibbs, J. L., Ellison, N. E., & Heino, R. D. (2006). Self-presentation in online
personals: The role of anticipated future interaction, self-disclosure, and perceived
success in lntemet dating. Communication Research, 33, 1-26.

Gofﬁnan, E. (1967). Interaction ritual: Essays on face-to-face behavior. New York:
Anchor.

Hiemstra, G. (1982). Teleconferencing, concern for face, and organizational culture.
In M. Burgoon (Ed.), Communication yearbook 6 (pp. 874-904). Newbury Park,
CA: Sage.

Hiltz, S. R. & Turoff, M. The network nation: Human communication via computer.
Reading, MA: Addison-Wesley, 1978.

Hitsch, G. J., Hortacsu, A., & Ariely, D. (2005). What makes you click: An empirical
analysis of online dating. Retrieved September 2, 2007, from
http://ﬁnwww.bc.edu/ec-i/semf2006/I-Iort.pdf

Jablin, F. M. & Krone, K. (1984). Characteristics of rejection letters and their effects
on job applicants. Written Communication, 1, 387-406.

Johnson, D. I. (2007) Politeness theory and conversational refusals: Assocations
between various types of face threat and perceived competence. Western Journal
of Communication, 71, 196-215.

Kline, S. L. & Floyd, C. H. (1990). On the art of saying no: The inﬂuence of social
cognitive development on messages of refusal. Western Journal of Speech
Communication, 54, 454-472.

Knobloch, L. K. (2006). Relational uncertainty and message production within
courtship: Features of date request messages. Human Communication Research,
32, 244-273.

Kunkel, A. D., Wilson, S. R., Owufowote, J ., & Robson, S. (2003). Identity
implications of inﬂuence goals: Initiating, intensifying, and ending romantic

relationships. Western Journal of Communication, 6 7, 382-412.

Limaye, M. R. (2001). Some reﬂections on explanation in negative messages. Journal
of Business and Technical Communication, 15, 100-110.

67

Locker, K. O. (1999). Factors in reader responses to negative letters: Experimental
evidence for changing what we teach. Journal of Business and Technical
Communication, 13, 5-48.

Madden, M. & Lenhart, A. (2006, March 5). Online dating. Pew Internet and
American Life Project. 1-27. Retrieved February 3, 2007, from
http://www.pewinternet.org/PPF/r/1 77/report_display.asp

McGlone, M. S. & Batchelor. J. A. (2003). Looking out for number one: Euphemism
and face. Journal of Communication, 53, 251-264.

Mongeau, P.A. Hale, J. L., Johnson, K. L., & Hillis, J. D. (1993). Who’s wooing
whom?: An investigation of female-imitated dating. In P. J. Kalbeﬂeisch (Ed.)
Interpersonal communication: Evolving interpersonal relationships (pp. 51-68).
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Mongeau, P.A., Serewicz, M.C., & Therrien, L.F. (2004). Goals for cross-sex ﬁrst
dates: Identiﬁcation, measurement, and the inﬂuence of contextual factors.
Communication Monographs, 71 , 121-147.

Morand, D. A. & Ocker, R. J. (2002). Politeness theory and computer-mediated
communication: A sociolingustic approach to analyzing relational messages.
Proceedings from HICSS ’03: The 3 6’“ Hawaii International Conference on
System Sciences. (HICSS-36). Los Alamitos: IEEE Press. Retrieved January 28,

2007, from
lmpzﬂcsdl.computer.org/comp/proceedings/hicss/2003/ 1 874/01/1 874toc.htm

O’ Sullivan, P. B. (2000). What you don’t know won’t hurt me: Impression
management functions of communication channels in relationships. Human
Communication Research, 26, 403-431.

Roloff, M. E. & Janiszewiki, C. A. (1989). Overcoming obstacles to interpersonal
compliance: A principle of message construction. Human Communication
Research, 16, 33-61.

Saeki, M. & O’ Keefe, B. J. (1994). Refusals and rejections: Designing messages to
serve multiple goals. Human Communication Research, 21, 67-102.

Success Stories. (n.d.). Retrieved July 2, 2007, from http://success.match.com/
Van Swol, L. M. (2007). Perceived importance of information: The effects of

mentioning information, shared information bias, ownership bias, reiteration, and
conﬁrmation bias. Group Processes & Intergroup Relations, 10, 239-256.

68

Walther, J. B. (1994). Anticipated ongoing interaction versus channel effects on
relational communication in computer-mediated interaction. Human
Communication Research, 20, 473-501.

Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal,
and hyperpersonal interaction. Communication Research, 3, 3-43.

Walther, J. B. (2007). Selective self-presentation in computer-mediated
communication: Hyperpersonal dimensions of technology, language, and
cognition. Computers in Human Behavior, 23, 2538-2557.

Weiner, B. (1979). A theory of motivation for some classroom experiences. Journal
of Educational Psychology, 71, 3-25.

Whitty, M. T. & Carr, A. N. (2006). Cyberspace romance: The psychology of online
relationships. Basingstoke: UK. Palgrave Macmillan.

69

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llllllllllllllllll/lllllllllllllllllllllllllllll

1293 02956 8411