—
——
fi—
_.
—

 

This is to certify that the

thesis entitled
Sex and Gender Differences:
Perceptions of Acceptability, Competence ,
and Honesty of Deceptive Messages

presented by

Anne P . Hubbell

has been accepted towards fulfillment

of the requirements fOl'

-. A Masters Commication
degree in

 

 

~\1W VJJLW/K
W

8380!"

Date —--; "" ”2— 50C)"

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

T

militiiiiiiiii

 
  

LIBRARY
Michigan State
University

   
    

PLACE IN RETURN BOX
to remove this checkout from your record.
return on or before date due.

 

 

T0 AVO ES
DATE DU DATE DUE
1A . .

_ «mum

DATE DUE

 

 

I r I

Jawzooq

 

(mm

g.»
(
(

 

ӣ3 6002

 

W21; 2»?

HEW-2””

 

 

m

 

 

 

 

 

 

 

me animal-mu

SEX AND GENDER DIFFERENCES: PERCEPTIONS
OF ACCEPTABILITY, COMPETENCE, AND HONESTY OF
DECEPTIVE MESSAGES

By

Anne P. Hubbell

A THESIS

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

MASTER OF ARTS

Communication Department

1997

ABSTRACT
Men and women are socialized to communicate differently. They also think about
interactions and the information shared in conversations in distinct ways. Women believe
that conversations with an intimate are important and relevant to the relationship. Men
do not see everyday conversations as relationally important. Information Manipulation
Theory (McCornack, 1992) postulated that deceptive mesfie vary in their distortion of
w
relationally important and relevant information. It was hypothesized that men and
women would perceive “white lies” as equally competent, honest, and acceptable. Also,
women would perceive ambiguous messages as more competent, honest, and acceptable
than men. Men were hypothesized to perceive complete distortions of the truth as more
competent, acceptable, and honest than women. As expected, men and women found
“white lies” to be equally appropriate (combined acceptability and competence ratings).
Also, men rated completely distorted messages as more appropriate than did women.
Unexpectedly, women and men rated ambiguous messages as equally appropriate. In
terms of honesty, the rank order of messages found by McCornack et. a1. (1992) was
supported. Some messages were perceived as more honest than others. Effects from
gender orientation - traditional and untraditional individuals - were also assessed.
Differing effects were found between traditional and untraditional individuals of both
sexes in terms of the appropriateness of messages. However, the rank-order for honesty

assessments of deceptive messages remained consistent with McCornack et. al. (1992) -

regardless of biological sex, gender orientation, or context.

DEDICATION

This thesis is dedicated to my great-grandmother (Busia), Julia Ptak-Karpanty; my
mother, Patricia Ann Ptak-Hubbell; and my grandmother, Dorothy Browne-Hubbell.
These three loving, strong, and beautiful women taught me that education is extremely
important and that women can be anything they want. I carry their dreams in my heart.

iii

ACKNOWLEDGMENTS
I have many people to thank and much thanks to give to them. First of all, my initial
editors: my father, Richard Hubbell, and my love and best friend, Max Leuenberger.
Thank you for your thorough work. Next, my friends Liz and Rachel for their
understanding when I did not have time for them and for their constant support and
advise. There were also the numerous people in the Communication Department who
gave me advise on the analysis: Dr. Frank Boster, Dr. Kim Witte, Dr. Gwen
Wittenbaum, Monique Mitchell, Anne Marie Hodges, Kenzie Cameron, and Marie
Carpenter. Their suggestions and support helped me to analyze data that I thought
insurmountable. In addition, much appreciation goes out to Gretchen for inputting a large
part of the data. Also, I need to thank my advisor, Dr. Steve McCornack for the research
that I have used as a basis for this investigation and for his theoretical and statistical
advice and support. Dr. Kelly Morrison, a committee member who along with Dr.
McCornack, read numerous manuscripts and gave valuable suggestions. I am very
appreciative for her time and effort. Also, Dr. Robert Griffore, a committee member
from the Family and Child Ecology Department, gave his time and patience to this
project. His enthusiasm about the findings was greatly appreciated. But, most of all, I
need to thank Max for helping me format this manuscript, create beautiful graphs and
figures apd a table of contents. I am also grateful for his constant love and support. He
listened (and listened) and gave (and gave) and made me laugh. I am thankful just to

know him.

iv

TABLE OF CONTENTS

 

 

 

LIST OF TABLES vii
LIST OF FIGURES viii
INTRODUCTION 1
CHAPTER 1: DEFINING GENDER 7

 

CHAPTER 2: SOCIALIZATION OF THE GENDERS AND ITS INFLUENCE ON

 

LANGUAGE USAGE AND PERCEPTIONS OF LANGUAGE 9

CHAPTER 3: INFORMATION MANIPULATION THEORY l7

 

CHAPTER 4: SEX/GENDER AND PERCEPTIONS OF DECEPTIVE

MESSAGES 23

 

CHAPTER 5: THE DEFINITION, EXPLICATION, AND

OPERATIONALIZATION OF HONESTY, COMPETENCE, AND

 

 

ACCEPTABILITY 27
CHAPTER 6: METHOD 14
CHAPTER 7: RESULTS 17

 

CHAPTER 8: DISCUSSION 49

 

APPENDIX A

APPENDIX B

<4

<5

 

APPENDIX C

APPENDIX D

APPENDIX E

‘6

<7

<8

 

APPENDIX F

<9

 

REFERENCES

60

 

vi

LIST OF TABLES

Table 1: Lie Appropriateness of Messages by Sex ........................................................... 40
Table 2: Lie Appropriateness of Message Type by Gender Orientation (Females) ......... 43

Table 3: Lie Appropriateness for Message Type by Gender Orientation (Males) ............ 45

vii

LIST OF FIGURES

Figure 1: Comparison of McCornack et.al. to Current Research ...................................... 39
Figure 2: Lie Appropriateness of Messages by Sex ........................................................... 41
Figure 3: Rank Order of Appropriateness of Messages (Men and Women) ..................... 41
Figure 4: Lie Appropriateness of Message Type by Gender Orientation (Females) ........ 43

Figure 5: Rank Order of Appropriateness of Messages (Traditional and Untraditional
Females) ...................................................................................................................... 44

Figure 6: Lie Appropriateness for Message Type by Gender Orientation (Males) .......... 46

Figure 7: Rank Order of Appropriateness of Messages (Traditional and Untraditional

Males) .......................................................................................................................... 47

viii

INTRODUCTION

The key thing about acting is honesty, when you can fake that, you’re in.
- Sam Goldwyn

The one truth in the language of humans is that people lie. Other creatures use
deception as a means of perpetuating their species, they need to blend in with their
surroundings in order to deceive predators. Humans, too, may have learned deception as
an adaptive response (Bond, Kahler, & Paolicelli, 1985). Humans learned that through
the deception of another individual, they could fithher their own goals (Bond, Kahler, &
Paolicelli, 1985). The individual who is fooled loses, perhaps without even knowing it,
and the successful deceiver gains. The key difference between humans and other animals
is that humans use a complex combinations of symbols, i.e., language, to lie.

Language is a primary means of communicating between individuals. With the
now mediated conversations of email and “chat rooms,” language could be considered the
primary mode of expression for the future. Much research on language focuses on its
acquisition and development and not on how individuals view its use. This research
supports the belief that language is rule, or norm, driven and that women and men are
socialized to use it in different ways. A lie is the use of language to deceive, which,
according to Grice (1989), is a violation of the maxims of conversation. Individuals are
expected to create truthful messages. Yet, there is little research which examines the

effects of violations of conversational norms on the person receiving such a message.

2
Also, there is no research on how the biological sex or the gender orientation of an
individual can influence these perceptions. This equation appears unbalanced, i.e.,
although it is believed that the violation of maxims of conversations is inappropriate, the
effects that this can have on the individual who receives such a message is not included.
It is important to not only how language is developed but also to look at the power it has
on influencing individuals, particularly in terms of deception.

In terms of language, deception is the changing or leaving out of information in
messages (McComack, 1992). There are many different ways to alter messages in order
to make them deceptive. Individuals could keep some relevant information within a
message, or, s/he could be ambiguous or, even change the subject. S/he could also
completely change all of the information. These all are illustrations of deceptive
messages described in the Information Manipulation Theory (IMT) (McComack, 1992).
IMT postulated that the manipulation of relevant information within a message could
make a message deceptive (McComack, 1992) The four types of deceptive messages
varied from each other by how that information was changed or distorted. In a test of the
IMT, of the four deceptive message types examined, one type, Quantity violations, was
found to be more honest and competent than the others (McComack, et. a1, 1992). This
type of violation could be considered a “white lie,” in that some but not all of the
information was given. McCornack, et. a1. (1992) also found that one message type,
Quality violations, was considered to be the leis; competent and honest. Messages of this

type contained complete distortions of the truth. The two other message types, Manner

3
and Relevance violations were described as ambiguous messages or an attempt to change
the subject away from the relevant information. Manner violations were found to be less
competent and honest than Quantity violations but more competent and honest than
Relevance violations (McComack, et. al., 1992). Both of these were perceived as more
honest and competent than the Quality violations (McComack, et. al., 1992). Although
much information was found in this test of the IMT, differences between women and
men were not examined.

In terms of how different individuals view deceptive messages, one difference
between individuals which could be assessed is their biological sex and their resulting
perceptions of language - particularly deceptive language. But, why should women and
men view deceptive messages differently in terms of competence, honesty or even
acceptability? One explanation is that they are socialized differently. For example,
women and men are socialized to use language differently. Even at early ages, boys and
girls can discriminate between what language usage is appropriate for a man compared to
a woman in the same situation (Edelsky, 1976). Further, women are socialized to use
more polite language than men in every situation, and men are expected to speak
differently depending on the sex of the recipient and the context of the interaction
(Kemper, 1984).

Women and men also see communication within intimate relationships differently
(Edwards, Honeycutt, & Zagacki, 1989). Women view everyday communication with

intimates as more significant than do men (Duck, Rutt, Hurst, & Strejc, 1991) and they

4
even imagine potential interactions differently (Edwards, Honeycutt, & Zagacki, 1989).
In terms of information disclosed in conversations, men and women feel differently about
what is necessary to disclose (Levine, McCornack, & Avery, 1992) Also, upon
discovering deception, women exhibit stronger emotional reactions than do men (Levine,
McCornack, & Avery, 1992).

But what about those individuals who exhibit modes of being within a
relationship which would commonly be attributed to the opposite sex, i.e., a masculine
female or feminine male? However great the differences in the socialization of women
and men, the lines between the sexes are beginning to blur. Gender is more than the sex
‘ of an individual but instead describes a “female or male social identity” (Worell, 1993, p.
205). This “identity” prescribes how s/he will interact with another individual within a
relationship (lckes, 1993). For example, untraditionally feminine males may exhibit
traditionally feminine traits such as “being kind, affectionate, and caring” (lckes, 1993, p.
76). Whereas, untraditionally masculine females may be more like traditional males in
that they are more “assertive, forceful, and decisive” (lckes, 1993, p. 76). Does this blend
of masculine with feminine or feminine with masculine traits in the untraditional person
influence the way they perceive deceptive messages?

This research is important for several reasons. First of all, it is important to take a
closer look at the differences between men and women. If men and women are socialized
to perceive communication differently, they may perceive deceptive messages differently

as well. Secondly, although sex differences may be important, perhaps in a society based

5
on the freedom of choice, individuals may orient themselves to a particular behavior,
regardless of their socialization. This is referring to the gender orientation of an
individual - masculine or feminine. Regardless of the biological sex of an individual,
s/he may exhibit behaviors which are typical of the other sex. This research includes
these self-described gender orientations. Perhaps important differences will be found
based on the masculine or feminine or traditional or untraditional outlooks of individuals
and their perceptions of deceptive messages.

Thirdly, this research is important because IMT has been criticized for its lack of
explanatory power (Stiff, 1996). This criticism stems from the fact that IMT has only
been tested using deceptive messages in response to infidelity situations (McComack, et.
al., 1992). By creating a different type of situation with which to compare the four types
of deceptive messages, perhaps the same rank-order of competence and honesty will be
found between the message types.

Fourth, and finally, a new construct, acceptance, will be added to the competence
and honesty assessments in order to determine if another dimension exists by which
individuals evaluate deceptive messages. Competence evaluations are aimed at the
skillfulness of the messages and honesty refers to the truthfulness of the messages.

This new construct, acceptability, has been added to try to get at the “right” or “wrong”
perceptions of deceptive messages, i.e., the value-laden perceptions of individuals.

This research will examine the influence of sex and gender on perceptions of

deceptive messages. Specifically, differences between both women and men as well as

6
differences between untraditional and traditional gender individuals will be examined as
they relate to perceptions of acceptability, competence, and honesty of deceptive
messages.
First literature on the socialization of males and females will be reviewed,
followed by a discussion of deception research and the IMT. Lastly, specific hypothesis

and research questions will be postulated based on the information examined.

Chapter 1: DEFINING GENDER

When examining the differences between women and men, the lines are not
always as clear as one would believe. The sex of an individual is defined by their
physical attributes whereas the gender is defined by their socialization (Aronfreed, 1968;
Ickes, 1993; Worell, 1993). Gender is more than if a person is born with female or male
sexual organs but instead describes a “female or male social identity” (Worell, 1993, p.
205). This “social identity” (Worell, 1993, p. 205) is used by an individual to aid them in
interactions so that they will act in socially appropriate ways (Ickes, 1993; Kemper, 1984)
and it helps others to know what to expect (Ickes, 1993; Kemper, 1984; Worell, 1993 ).

Much evidence supports the existence of androgynous individuals, those who
blend the stereotypical aspects of femininity and masculinity (Ickes, 1993). These
“aspects” refer to traits that an individual exhibits. They are defined as follows:

the traditional feminine gender role is a social orientation that emphasizes

closeness and solidarity, whereas the traditional masculine gender role is a

social orientation that emphasizes power and status (Tannen, 1987, Taken
from: Ickes, 1993, p. 76).

In contrast, an untraditional individual takes on traits which would be expected
from a person of the opposite sex (Ickes, 1993). For example, an untraditional male may
be more feminine in his interactions with others. Further, an untraditional female may be
more masculine.

One example of defining and measuring gender is Bem’s Sex Role Inventory

(BSRI) (1974). Bem created the BRSI to determine how established individuals are in

8
their sex roles, and as a measurement of androgyny. Androgynous individuals fall in
between highly traditional males and highly traditional females, thus, these individuals
represent a blend of the two. Alternatively, untraditional individuals take on the reverse
sex role; men are more feminine and women are more masculine. This is further
described as:

the greater the absolute value of the Androgyny score, the more the

person is sex typed or sex reversed with high positive scores indicating

feminine and high negative scores indicating masculinity. A ‘masculine’

sex role thus represented not only the endorsement of masculine attributes

but the simultaneous rejection of feminine attributes (Bern, 1974, p. 158).

For this scale, the reverse is true for feminine scores. The higher they are, the
greater the endorsement for feminine attributes.

Androgynous individuals are not the subject of this research attempt, although
further analysis and hypotheses can be generated for these intriguing individuals.
However, traditional and untraditional females and males were selected for analysis. This
was done so that comparisons could be made between individuals of the same sex who
described their sex-role orientations as completely opposite. If these polar orientations
produce distinct perceptions of deceptive messages, then further, in-depth analysis of the
blended individuals will be warranted.

The gender-orientation of an individual may influence her/his perception of
deceptive messages, however, the sex of an individual may influence these perceptions as

well. The next chapter will review the literature relevant to the expected differences

between women and men.

Chapter 2: SOCIALIZATION OF THE GENDERS AND ITS INFLUENCE ON

LANGUAGE USAGE AND PERCEPTIONS OF LANGUAGE

There are two primary areas of research related to gender and language, usage and
perception. Women and men are socialized to use and to perceive language differently.
Although the focus here is on the perception of language, the socialization of how

individuals are expected to use language can help explain why perceptions of language

differ.

Men and Women and the Socialization of their Language

 

The ancient Greeks were one of the earliest examples of appropriate language
usage between the sexes. Women “swore by (the two) goddesses (Demeter and
Persephone)” (Gregersen, 1979, p. 4) and Greek men swore “only by gods (e.g. Apollo)”
(Gregersen, 1979, p. 4). Although women could easily use the man’s way of taking a
god’s name in vain, a man would never refer to a goddess (Gregersen, 1979). Even in
current times and in languages throughout the world there exists a female and differing
male dialect. An example of this is as follows:

These differences range from most features of pronunciation to

morphological distinctions, such as those found in Japanese or Chiquita,

where men have a gender system overtly marked in nouns and women lack

it (Gregersen, 1979. p. 4).

A woman in Japanese society when interacting with men in a nontraditional situation may

have to adapt by speaking English, which is considered neutral (Gregersen, 1979). This

10
is because there is no good way for a Japanese female to use the commonly accepted
male dialect without being seen as offensive. Yet, the Japanese female language is not
one used to communicate respect but instead subservience (Gregersen, 1979). Therefore,
a Japanese woman working within a business with other men is in a very difficult
situation. She either is offensive with her speech or is subservient. Differences between
women and men also exist when bilingual cross-sex dyads interact. This is illustrated as
follows:

Men whose first language was either Spanish or both Guarani and Spanish

tended ‘to use more Guarani with other men, but to use Spanish with women

who are their intimates. Women, on the other hand, whose first language was
either Spanish or both, tend to use Spanish to both male and female intimates’

(Gregersen, 1979, p. 15).

In this particular situation, men varied their language according to the sex of the recipient
and women did not. This variance and non-variance of language can also be found in the
North American culture (Kemper, 1984).

The differences between what is acceptable for women and men are known at a
very early age. Young children display diversity in their interpretation of other’s use of
language. Edelsky (1976) demonstrated that first graders are able to distinguish between
what society would deem appropriate language for a woman and what would be
acceptable for a man (Edelsky, 1976). Words and types of phrases which were
considered to be masculine were: “Damn it... Damn + Adjective... I’ll be damned... (and

a) Command” (Edelsky, 1976, p. 51). The feminine words and types of phrases were as

follows: “Adorable... Oh dear... My goodness... Won’t you please... (a) Tag Question...

11
So... Very... (and) Just” (Edelsky, 1976, p. 51). Children in the first grade were able to
distinguish “Adorable” as a woman’s word and “Damn it” as a phrase that a man would
use (Edelsky, 1976). By the third grade they had figured out all but “Tag Question... So...
Very... Just... (and) Command” (Edelsky, 1976, p. 51). In the sixth grade they had
attributed all of the words and phrases to the appropriate gender (Edelsky, 1976). When
contrasted with adults, all of the words and phrases were attributed the same as the sixth
graders with the exception of the “Command” phrase. This phrase was considered to be
used equally between men and women and was considered acceptable for both.

Although much research has been done on adults and gender-specific language,
there exists no “generally accepted theoretical framework from which to view the existing
data on gender differences in language use” (Simkins-Bullock & Wildman, 1991 , p. 151).
To create such a framework, Simkins-Bullock & Wildman (1991) utilized three
perspectives which consistently show up in the literature.

The first perspective focuses on the different ways that women use language
compared to men. Women are found to communicate less directly, to be more tentative
and unsure than men (Lakoff, 1975; Simkins-Bullock & Wildman, 1991). Men are
viewed as more direct and tougher than women (Kemper, 1984; Lakoff, 1975; Simkins-
Bullock & Wildman, 1991). Research in this area focuses on tag questions, qualifiers,
disclaimers, and women’s use of formal language (Kemper, 1984; Simkins-Bullock &

Wildman, 1991).

12

A second perspective on the differences between the genders’ use of language
focuses on the intent for communicating (Simkins-Bullock & Wildman, 1991). Women
communicate to fill social needs (Duck et al., 1991; Ickes, 1993) and to show caring for
another person such as empathy (Dalton, 1983; Hoffman, 1977b) and support (Clark,
1993). Men are more task-oriented and instrumental in their communication (Hoffman,
1977a; Ickes, 1993; Simkins-Bullock & Wildman, 1991).

The third perspective on the communication differences between the sexes
focuses on power in relationships (Simkins-Bullock & Wildman, 1991). It is socially
perceived that men have more power and status than women (Simkins-Bullock &
Wildman, 1991) and that they are more oriented toward power in a relationship than with
affiliation (Ickes, 1993). Women are socialized to defer to males (Gregersen, 1979;
Hoffman, 1977a; Simkins-Bullock & Wildman, 1991). In communication, this power
disparity is illustrated by interruptions, tum-taking, nonverbals such as touching and eye
contact (Simkins-Bullock & Wildman, 1991), Quantity of talk, and control of topics
(Edwards etal., 1989; Simkins-Bullock & Wildman, 1991). In all of these situations,
men are perceived to dominate (Edwards et al., 1989; Simkins-Bullock & Wildman,
1991)

All three perspectives are supported by research. However, the focus of these
perspectives was not on close intimate relationships. The research focus here is on just
such a situation, communication between intimates. Also, it is obvious that there exists a

wealth of information on how women and men are socialized to fit; language differently

13
There does not exist, however, such a vast amount of research in terms of their different
perceptions about communication.
Women and Men and the Perception of Language within an Intimate Relationship

In intimate relationships, women think differently than men about interactions.
When daydreaming about potential discussions, women see themselves as talking more
than they actually do and as being more satisfied with the outcomes of the imagined
interaction (Edwards, Honeycutt, & Zagacki, 1989). Both men and women think about
the same variety of t0pics that they want to discuss with their significant other but women
think about these potential interactions more than men do (Edwards, Honeycutt, &
Zagacki, 1989).

Also, interpretations of actual conversations differ between the genders. Levine,
McComack, and Avery (1992) claimed that beliefs about the exchange of information in
relationships is different between women and men. Duck, Rutt, Hurst, and Strejc (1991)
found that women “place more value than men upon the information that is exchanged on
an everyday basis within relationships” (p. 289). Women were also more likely to view a
specific conversation as airing the relationship in some way while men saw the same
interaction as one denoting spam! in the relationship (Duck, Rutt, Hurst, & Strejc,
1991). Women look more for information about the long-term goals of the relationship
than men and focus on how these goals can be integrated into relational communication
(Levine, McComack, & Avery, 1992; O’Keefe, 1988). One conversation can mean very

different things to a woman than to the man with whom she is speaking. If a woman sees

14
the conversations that she has with her significant other from a different framework than
the man she is trying to communicate with, there is no wonder that there are
communication break-downs between the sexes.

The first perspective in the previous section applies to the emphasis of this
research. It focuses on the social expectations of the ways in which women and men
communicate with each other. Women are consistently expected to be polite and formal
in their conversational attempts (Edelsky, 1976; Gregersen, 1979; Kemper, 1984)
whereas men have more allowance for profanity and rougher talking (Edelsky, 1976;
Gregersen, 1979; Kemper, 1984). As already discussed, children as young as first graders
are already aware of these types of stereotypes (Edelsky, 1976). The research on these
young individuals found that the female-appropriate modes of speaking were sweet
adjectives or polite ways of explaining or asking for something (Edelsky, 1976). For
males, swearing and commanding were considered acceptable to the young children
(Edelsky, 1976).

With adults the expectations become more diverse. Women in general are
expected to use a more formal or polite way of speaking (Edelsky, 1976; Gregersen,
1979; Kemper, 1984; Lakoff, 1973; Lakoff, 1975; Lakoff, 1977). This politeness
expectation continues whether a woman is speaking to a man or another woman, and
also, regardless of whether the task is considered to be a masculine or feminine one

(Kemper, 1984). This is best explained as follows:

15

...women are supposed to speak more politely than men...women don’t use off-

color or indelicate expressions; women are the experts at euphemism; more

positively, women are the repositories of tact and know the right things to say to
other people, while men carelessly blurt out whatever they are thinking. Women
are supposed to be particularly careful to say ‘please’ and ‘thank you’ (Lakoff,

1975, p. 55) (Taken from Kemper, 1984, p. 435).

Although women are expected to consistently speak politely, men are expected to
modify their way of speaking with the gender of the person they wish to address and the
type of goal - masculine or feminine (Kemper, 1984). This is explained as follows:

Whether women are attempting to achieve feminine goals (e.g. getting the room

dusted) or masculine goals (e.g., getting the leaves raked), they are expected to

use ‘please’ and to be polite. However, the form or request a man is expected to
use depends on whether he is requesting a masculine, feminine, or neutral
action...Men are expected to use impolite forms of requests to achieve masculine
goals (e. g. getting the car door fixed)...Men who seek feminine goals (e.g. getting
tea made) are expected to speak like women - to be polite, to use ‘please’

(Kemper, 1984, p. 442).

Men are expected to change their Manner of speaking according to the situation and
gender of the person they are talking to.

These expected norms of language usage are evaluated by other individuals.
What matters is not how well constructed a message is, but how it is received. A
statement given by an individual is only as valuable as how the information gets encoded
by the receiver. Women and men think differently about interactions and about what is
acceptable language to use to speak to each other. But, do these perceptions of acceptable

messages within interactions in average conversations also appear in deceptive messages?

Evidence suggests that there are some deceptive messages are considered more competent

l6
and honest than others (McComack et. al., 1992), however, women and men were not
separated out in this research. If women and men think differently about interactions
within close intimate relationships, they may also perceive deceptive messages differently
in terms of honesty, competence, and the new construct, acceptability.

But what about the individuals who define themselves as not following the norms
prescribed for women and men? Women and men do not always follow all of the rules
(Ickes, 1993; Kemper, 1984). In fact, Ickes (1993) in his study of couples’ satisfaction,
found that the more androgynous the individuals, the more satisfaction was reported,
particularly if both sexes acted in typically feminine ways toward each other. When both
of the heterosexual partners took on feminine attributes, they were loving and considerate
in their speech patterns.

In summary, there are different expectations for language usage between women
and men. What is acceptable for men to say is not acceptable for women. Women’s
language is more polite across all situations than men’s. Not only do men and women
speak differently, they think about their conversations with intimates differently.
Information which is important to women may or may not be important to men. Lastly,

they deal with conflict within romantic relationships in Manners which are divergent.

Chapter 3: INFORMATION MANIPULATION THEORY

McComack (1992) created the Information Manipulation Theory in order to
examine deceptive messages. Deceptive messages were defined as those messages which
“mislead listeners through covertly violating the principles that underlie and guide
conversational understanding” (McComack, 1992, p. 2). Information Manipulation
Theory is “a framework for describing the different ways that information can be
manipulated to accomplish deceit” (McComack et al., 1992, p. 17). Borrowing from
Ekman (1985), McComack (1992) stated that “there are two primary ways to lie: to
conceal and falsify” (p. 3). This can be further explained as follows:

Individuals can ‘play’ with the information they disclose in at least two

different ways: they can adjust the amount of information that is disclosed,

and they can choose to disclose false information (McComack, 1992, p. 3).
Information Manipulation Theory was built around the belief that changing or playing
with the information results in violating basic conversation rules or principles
(McComack, 1992). This is all done secretly or covertly as mentioned above which is
what makes it deceptive (McComack, 1992), in that the purpose is to disseminate false
information and make it look like the truth.

Borrowing from Grice (1989), McComack (1992) built “four primary ways” (p.
4) of “manipulating information” (p. 4). These were as follows:

there are at least four dimensions along which information can be varied in

the production of deceptive messages: manipulations of Quantig,...manipulations

of Qualig,...manipulations of Relation,...and manipulations of Manner
(McComack, 1992, p. 6).

 

l7

18

Violations of Quantity (McComack, 1992) are described as “assumptions
includ(ing) expectations regarding the M of information that should be provided”
(McComack et al., 1992, p. 18). These violations include messages which contain some
of the important and truthful information but edit out either all of or some of the
important facts (McComack, 1992). An example of this is a response to the situation in
which a romantic partner stops by as the source is preparing to go out to meet another
potential romantic interest. The message created is not completely dishonest, it just
leaves out some of the truth and it is as follows:

“Jo, thanks for stopping by! That was really sweet and thoughtful of you.

I really want to see you, but not tonight. Sorry!” (McComack, 1992, p. 9).

In this situation the source can “disclose some of the ‘sensitive’ information...Yet, these
messages are deceptive because they fail to disclose the critical piece of the contextually-
relevant sensitive information” (McComack, 1992, p. 10).

Qu_alifi violations (McComack, 1992) refer to “the M of the information
presented” (McComack et al., 1992, p. 18). These messages are “perhaps the most
prototypical ‘deceptive messages,’ in terms of how theorists and laypersons have
traditionally thought of deception” (McComack, 1992, p. 10). In the situation mentioned
above, the Quality violation would be as follows:

“Jo, Hi! How are you? Jo, it’s really nice of you to come over tonight, but I

really don’t feel very well and I’m staying in tonight by myself. I really need a

night on my own to think about some of my problems. I will give you a call
tomorrow” (McComack, 1992, p. 10).

19
In this type of deceptive message “distorted versions of the sensitive information, or the
presentation of completely fabricated information” (McComack, 1992, p. 11) is the norm.
In other words, the person makes up something which may or may not be remotely
related to the truth. Information which could be considered relationally relevant is not
changed but is completed deleted.

m violations (McComack, 1992) are those which effect “the Relevance of
information within conversational contributions” (McComack et al., 1992, p. 18). This
type of deceptive message is an attempt to “initiate new topics in order to divert the
direction of the conversation away from its projected course” (McComack, 1992, p.. 12).
An example of this message (in the same situation as above) is as follows:

“(Jo!) Why didn’t you tell me you were coming!? I mean, I know that you get

paranoid some times, but driving all the way down here just to check up on me

is a bit ridiculous, don’t you think? How would you like it if I paid a sneak visit

to you and acted like a bitch by surprising you and asking you what you had

been doing?” (McComack, 1992, p. 12).

The source created this message to take the attention away from what really was
happening and put the emphasis on the receiver’s (Jo’s) actions. This could be
considered a method of avoidance.

Finally, Mpann_er manipulations (McComack, 1992) are those in which the “ng
of information provided within messages (McComack et al., 1992, p. 18) is violated. In
fact these violations are referred to as “Clarity Violation(s)” in subsequent research

(McComack, 1992, p. 12). The example for this (in again the same situation as above) is

as follows:

2O

“Whaaa??! Uh, Jo, I really appreciate the surprise and would like to spend

some time with you, but I have plans for tonight. May I call you tomorrow so we

can set something up?” (McComack, 1992, p. 12).

In this type of message, the source “discloses some of the information that is possessed,
but does so in an ambiguous fashion” (McComack, 1992, p. 12). Some of the relationally
relevant information may be included but the message remains ambiguous.

Understandably, these deceptive message types can be created independently or
an individual may create a message which is a blend of two or perhaps more of these
violations. For example, an individual may use a Quantity violation with a violation of
Relevance - also called Relation (McComack et al., 1992). In this case, information
would be deleted out of the deceptive message and the source would also attempt to
change the subject.

McComack et al., (1992) tested Information Manipulation Theory by presenting
receivers with deceptive messages which fit into each of the categories mentioned above.
These messages were created for each of the three romantic situations which were
gender-free (McComack etal., 1992). The participants were asked to rate the messages
generated in response to the situations for “perceived deceptiveness and competence”
(McComack et al., 1992, p. 17).

In terms of ratings of veracity, the “Completely Disclosive” (McComack et al.,
1992, p. 22) messages, which were the control, were found to be the most competent and
honest. This is different from other findings, perhaps because respondents were given all

of the information up front and because the situations were romantic ones where honesty

21
is considered to be essential (McComack et al., 1992). Of the four deceptive dimensions,
the violations of Qpalijy (McComack et al., 1992) were seen to be the most dishonest and
the least competent. However, 9pm violations (McComack et al., 1992) were not

only perceived to be the most honest of all of the deceptive message types, but were also

 

seen to be the most competent and acceptable. This was reasoned to be because a “simple
‘white lie’ may be superior” (McComack et al., 1992, p. 24) to totally distorting the
information (Quality violation).

The second most dishonest and incompetent deceptive message was the violation
of Relevance (McComack et al., 1992). This type of message is constructed by the
source so as to change the subject. (McComack, 1992). The third most incompetent or
dishonest message can be looked at from the other direction - violations of Lanty
(McComack et al., 1992) - follow Quantity violations (McComack et al., 1992) in terms
of perceived honesty and competence. Overall, Quantity and Clarity violations were
found to be the most honest and competent of the four deceptive types of messages
(McComack et al., 1992). The violations of My and Relevance were seen to be the
least honest and competent (McComack et al., 1992). Perhaps this illustrates that lies
which give some information or are ambiguous are more acceptable than those which are
completely untruthful or avoid the issues entirely.

A rank-order for competence and honesty was found for the types of deceptive

messages described by IMT. This rank order is as follows:

22

Quantity Violations Most Honest and Competent
Clarity Violations

Relevance Violations I

Quality Violations Least Honest and Competent

This test of the IMT (Information Manipulation Theory) not only supported the
usefulness of the theory but also illustrated which messages may be considered more
honest and competent (McComack et. al., 1992). IMT, has not, however, been used to
assess the differences between women and men and their perceptions of deceptive
messages. As discussed above, the socialization of an individual as well as their gender
identity may influence her/his perceptions of language use. The following chapter links

these ideas together in terms of hypotheses and research questions.

Chapter 4: SEX/GENDER AND PERCEPTIONS OF DECEPTIVE MESSAGES

In the test of the Information Manipulation Theory, Quantig violations were
found to be the most honest and competent of the deceptive messages (McComack et. al.,
1992). Some of the information is included in the message and some is not. These types
of messages are not totally honest or competent but are more so than the other types of
deceptive messages. Since these “simple ‘white lie(s)” (McComack et. al., 1992, p. 24)
are the least offensive of any of the types of lies, both women and men will consider

uanti manipulations (McComack, 1992) to be equally acceptable, competent, and
honest. The strong effects for this message as being the most competent and honest as
well as the fact that, for this violation there appears to be no reason as to why women and
men would perceive this message differently, support the belief that no differences will
be found between women and men and their perceptions of Quantity violations. This also
includes individuals who are untraditional or traditional in their gender-definition.
Although not as competent or honest as a completely disclosive message, the Quantig
violation will be found to be equally acceptable, honest, and competent between all
groups of individuals. However, differences between groups will be found for messages
which contain manipulations of My. This is because violations of ME!
(McComack, 1992), or completely “distorted versions of the sensitive information”
(McComack, 1992, p. l 1), have been found to be the _m_os_t dishonest and le_ast commtent

deceptive messages (McComack et. al., 1992). Since women are socialized to be more

23

24
polite in conversing with other individuals across all situations (Edelsky, 1976;
Gregersen, 1979; Kemper, 1984) than men, they will view these messages harsher than
will men. This is because this type of message violates more conversational norms the
most (McComack, 1992). Further, women view information which is exchanged in
conversations as more important than do men (Levine et. al., 1992). Manipulations of
QM leave out or distort the information more than the other types of deceptive
messages (McComack, 1992). Therefore we would expect that:

H1: Women will find Quality Violations to be less competent,
honest, and acceptable than men.

The differences between the self-described traditional and untraditional
individuals are of interest here, however, there is not enough information available on
their perceptions of communication within close intimate relationships. Therefore, it
would be impossible to postulate what their perceptions of deceptive message types
might be. The following research question will be raised:

R1: Will gender role (i.e., traditional or untraditional individual) have effects
upon perceptions of Quality violations?

It is important to note that the comparisons between traditional and untraditional
individuals are between individuals of the same sex who vary in their masculine or
feminine orientation.

Violations of Relation refer to deceptive messages in which the source, or
deceiver, change the subject or focus on the receiver’s actions in order to divert attention

away from the relevant information which could have been exchanged (McComack,

25
1992). Because women feel that information is important in communicating (Levine et.
al., 1992), they may feel that these messages are less acceptable than will men. Further,
men have been found to use avoidance or withdrawal categories in conflict situations
within heterosexual intimate relationships (Markman, Silvem, Clements, and Kraft-
Hanak, 1993). Perhaps because men may want to avoid a difficult situation, they may
feel that a deceptive message which emphasizes avoidance such as the violation of
Relation will be more acceptable than will women. This type of deceptive message -
along with violations of Manley (or M - did not differ as strongly from any other
violations in terms of acceptability as did Quantig and Qua—lity (McComack, et. al.,
1992). There does not exist strong enough information to hypothesize a relationship,
however, a research question may get at some enlightening information. Thus the
following will be examined:

R2: Will gender role (traditianal/untraditional) have effects upon perceptions of

Relevance violations?

Mappg or Cl_ari_ty manipulations are those in which the deceiver “discloses some
of the information that is possessed, but does so in an ambiguous fashion” (McComack,
1992, p. 12). These violations were found to be the second most acceptable deceptive
message after Quantity violations (McComack, 1992). Women would find this
acceptable because there is still some information disclosed (Levine et. al., 1992) and
because of its ambiguity. An ambiguous message may be considered to be much more

“polite.” A woman may feel that the important information was enclosed in just such a

26
message yet done so in a Manner which could offset a negative interpretation. They have
tried to be “nice” about their message.

There is a wealth of information on the norms of politeness for women and this
message was found to be acceptable after violations of Quantig (McComack, 1992).
Therefore, we can expect that:

H2: There will be a significant sex difference in perceptions of Mel violations
such that women will find My violations to be more honest, competent,
and acceptable than men.

Applying this information to gender-identity, the following research question was

formed:

R3: Will gender role (traditional/untraditional) have effects upon perceptions of
Manner violations?

Women and men think about communication with their significant others in
different ways. They are also socialized to use language in differently. But, does this
affect deceptive message and the perceptions of appropriateness of these messages? To
better understand the measurements which will be used to assess these perceptions, the
constructs of honesty, competence, and acceptability will be explicated and defined in the

following chapter.

Chapter 5: THE DEFINITION, EXPLICATION, AND OPERATIONALIZATION

OF HONESTY, COMPETENCE, AND ACCEPTABILITY

Honesty and competence were constructs used to test IMT (Information
Manipulation Theory) (McComack, 1992). Within the deception literature, acceptance
has not been a focal concept. Acceptance appears to have been an assumed aspect of
competence yet, it was a distinct construct. Within the deception literature, acceptance
has not been a focal concept. It has been viewed as a global construct used to measure
the overall acceptability of deception. Honesty has been described from a global or a
specific level and competence has only been used to test the deceptive message types
from IMT. Acceptance is a unique construct which could help differentiate the deceptive
message types proposed by IMT (McComack, 1992).

Other Conceptualizations

The three constructs have varied from being defined globally (or broadly) to being
comparisons to specific other constructs. Research prior to IMT assumed that honesty
was a primitive term, therefore, needed no definition. It was globally (or broadly)
described as the overall veracity of a relational partner (deTurck et. al., 1990, Ekman &
Friesen, 1969, McComack & Levine, 1990, McComack & Parks, 1990, McComack &
Parks, 1986, Riggio & Friedman, 1983).

From a more specific level, IMT (McComack, 1992) was tested with both honesty

and competence constructs (McComack et. al., 1992, Jacobs, Dawson, & Brashers,

27

28

1996). McComack (1992) defined both as they pertained to IMT’s deceptive message
types. Since these message types differed in terms of the way information was
manipulated within them, honesty was described as the extent to which individuals
believed that they were receiving all of the relevant information from a conversational
partner (McComack, 1992, Jacobs, Dawson, & Brashers, 1996). Different message types
were perceived as more or less honest based on the manipulation of information within
them.

Competence was described in a similar fashion, with its Relevance to deceptive

message types. Competence was described as the appropriateness or face-maintenance of

 

a message (McComack et. al., 1992). This definition focused on behavior within
situations. Appropriateness referred to how suitable the behavior was for the situation.
Face-maintenance referred to how the behavior saved the partner from embarrassment,
again within a particular situation. It is important to note that Jacobs, Dawson, and
Brashers (1996) did not include competence ratings for their test of IMT because they
believed them to be uninterpretable and unrelated to what they were investigating. They
did, however, utilize honesty as a construct and defined it in the same Manner as
McComack et. a1. (1992).

Acceptance was defined from a global perspective in terms of the overall
acceptability of deception (Levine, McComack, & Avery, 1992). For example, if
someone lied to an individual, how satisfactory would that behavior be? In particular it

was described as the view that deception was “a significant relational transgression (i.e.,

29
an unacceptable form of behavior” (Levine, McComack, & Avery, 1992, p. 291). It was
not situation specific but simply how acceptable, moral, and excusable lying behavior
was considered to be.
Other operationalizations

Along the same lines as the definitions of the constructs, the three again varied in
respect to what they measured: global perceptions about deception or specifically tying
the constructs to other variables.

Honesty has been operationalized in many different ways, most of these to
determine global or broad perceptions. Most often individuals were simply asked to
make a dichotomous assessment of their partner’s veracity based entirely on nonverbal
behaviors (Ekman & Friesen, 1969, Riggio, & Friedman, 1983) or their partner’s verbal
responses to a scale (deTurck et. al., 1990, McComack & Levine, 1990, McComack &
Parks, 1990, McComack & Parks, 1986). Buller, Strzyzewski, and Hunsaker (1991)
operationalized honesty assessments in three ways: “an initial honesty rating scale” (p.
30) which was based on perceived sincerity or honesty; a “seven-item trust dimension of
the Relational Communication Scale” (p. 30) which was used to develop a “less direct
measure of honest attributions” (p. 30); and an “honesty judgment” (p. 30) which was a
dichotomous assessment of the veracity of the potential deceiver. Stiff et. a1 (1990)

operationalized honesty as follows:

30
Participants’ ratings of overall truthfulness and deceptiveness were
measured with a single, seven-point semantic differential-type scale.
Ratings of verbal message content were obtained using a lO-item
semantic differential-type scale. These items were developed to
reflect characteristics of truthful and deceptive verbal content...Scale items

included such word pairs as: consistent-inconsistent, plausible-implausible, and
direct-evasive (p. 218).

I

More specifically, McComack et. a1. (1992) and Jacobs, Dawson, and Brashers
(1996) utilized measurements of honesty to describe differences between IMT’s
deceptive message types. Honesty was assessed via a “four-item semantic differential
scale using seven-point response formats” (p. 21). The measures were as follows:
“Dishonest/Honest, Deceitful/Truthful, Deceptive/Not deceptive, Misleading/Not
Misleading” (p. 29). These measures were compared to the four deceptive message types
postulated by IMT to determine whether the messages differed in perceived honesty.

Also, within the first test of IMT, competence was measured with a specific
perspective by McComack et. al. (1992). McComack et. a1. (1992) used the same format
that was used to assess honesty. The measures were as follows: “Ineffective/Effective,
Inept/Skillful, Incompetent/Competent, Mismanaged/W ell Managed” (p. 29). These
measures were also compared to the four deceptive message types of IMT. Jacobs,
Dawson, and Brashers (1996) as mentioned above chose to not utilize competence as a
construct.

The acceptability of lies was assessed from a global perspective by Levine,
McComack, and Avery (1992) by comparing women to men with the belief that women

would rate lies as less acceptable than would men. The scale used was created by Levine

31
and McComack (1991) and utilized “11 Likert-type items” (p. 292) (see Appendix D).
The items focused on how acceptable, moral, and excusable lying behavior was in the
mind of the subject.

Acceptance was only examined from a broader context but should be considered
to be a valid addition to the competence and honesty constructs. The reliability of the Lie
Acceptability Scale created by Levine and McComack (1991) was supported and the
items were different than those used for competence and honesty. The only shortcoming
of the scale was that it only assessed global perceptions. To test IMT this must be
brought to a more specific level. Therefore, items will need to be altered to incorporate a

situation and not a broad assessment.

Definitions of honesgl, competence, and acceptability

 

Honesty has been considered to be a primitive, or basic, term which refers to an
individual’s veracity. Veracity is a synonym for honesty and is considered to represent
the truthfulness of an individual. For the purposes of this research this definition will be
applied to specific messages. A message will be evaluated along the parameters of how
honest it is. This is possible because even deceptive messages contain some elements of
truth. Individuals may perceive a message with more truth in it to be more honest than a
message with less truth.

Competence is a derived term which refers to the appropriateness of the messages
tested by IMT (McComack, 1992). Appropriateness is a primitive term which describes

the suitability of actions or messages. For example, a message may be more suitable or

32

fitting in one particular situation but not in another. Individuals expect their
conversational partners to use messages which are appropriate to a situation, i.e., it makes
sense that they say what they do. When an individual does not firlfill this expectation
they will be perceived as less competent than an individual who does. Therefore, the
appropriateness of a message in a particular situation will indicate how conversationally
competent the individual who uses the message is. Competence is not to be confused
with acceptance, however. Acceptance refers to the more value-laden aspects of
perceptions about deceptive messages. For example, where competence ratings may
demonstrate an evaluation pertaining to what is appropriate in a situation, acceptance
measures would illustrate what is considered right and wrong.
Op_erationalization of constructs

For the purpose of this research, the same parameters used by McComack et. a1.
(1992) will be employed to assess the honesty and competence of deceptive message
types in one of two hypothetical situations. However, instead of semantic differentials,
7-item Likert scales were adapted from the original Semantic Differential Scales. This is
so that the unidimensionality of the scales can be tested. This test will demonstrate if any
items need to be thrown out which do not support a unidimensional solution. Also, items
which demonstrate a factor loading of less than .40 will be deleted (as consistent with the
test of the Lie Acceptability Scale by Levine, McComack, and Avery, 1992). This will
make it possible to demonstrate if the competence and honesty items measure what they

are supposed to. The acceptance ratings will be tested in the same Manner. It is

33
particularly important to check the factor loadings for the acceptance items to make sure
that as they were converted from global to specific measures that they are measuring the
same construct. Also, this will demonstrate that the competence and acceptance items are
differentiated.
The Lie Acceptability Scale will also be utilized for this research in order to

determine if women or men perceive deception as an overall behavior as less acceptable.

34

Chapter 6: METHOD

The Procedure

The students were first informed of the confidentiality of their responses. They
were also told that the purpose of the study was to assess perceptions of communication
within a romantic relationship. They were then asked to sign a consent form. The
consent form reminded them that their answers were completely confidential and
anonymous. After signing the consent form, they then were given the questionnaires and
asked to be completely honest with their responses. After completion of the
questionnaire, participants were thanked for their time.

The In_strument (Questionnaire)

The first section of the questionnaire contained a situation which remained
consistent across all conditions. This situation was adapted from McComack et. al.
(1992) in that the name of the individual who was being deceptive was gender neutral
(Chris) and that the situation was a romantic one. Because a situation which included
infidelity may have caused different reactions between women and men, the original
situation was altered (see Appendix A for the original situation from McComack et. al.,
1992). This was so that the messages in response to the situation would be where the
differences between the sexes/genders existed. The situation created described Chris
going home to visit her/his significant other’s family. Chris was overheard by her/his

partner making derogatory statements about her/his family and about the evening spent

35
with them (see Appendix A). The partner confronted Chris at a later time and Chris’s
response was one of the five messages.

Following the situation, participants received one of 5 messages in response.
Four of the messages represented the deceptive message types described in the IMT
(McComack, 1992). The fifth message was a completely disclosive message which was
the control (see Appendix B). The participants then completed items assessing the
acceptability, competence, and honesty (see Appendix C) of the message they received.
Each of these items were Likert-type.

In the second section of the questionnaire the participants were asked to complete
the Bem Sex Role inventory (1974). They were not informed that this was the name of
the scale.

Following these sections, participants were asked to answer questions from the
Lie Acceptability Scale (Levine, McComack, & Avery, 1992) (see Appendix D). This
was given to determine the global or broad perceptions regarding deceptive behavior.

Lastly, the participants were asked to give demographic information such as sex
and age.

The Participants

The participants were college students at Michigan State University. Two
hundred and seventy four individuals (104 males and 169 females) completed the
questionnaire but one case was deleted from analysis due to missingresponses. Of the

sample used for analysis, 231 were Caucasian, 24 African American, 3 Hispanic, 9 Asian

36

American, 1 Native American, and 5 other. The age of the respondents ranged from 18 to
35 (M = 20). Gender orientation was broken down as follows: untraditional males = 51,
traditional males = 92, untraditional females = 94, and traditional females = 124.

Participants were not told the entire contents of the study but instead were asked
to participate in research regarding communication in intimate or close interpersonal
relationships. Participants were ensured that their participation was entirely voluntary
and confidential and that they would be given extra credit in their college class for their

participation.

Chapter 7: RESULTS

Scale reliability was assessed by confirmatory factor analysis. All of the scales
were assessed for internal consistency and parallelism. Acceptability and competence
were found to be measuring the same variable, thus the items were collapsed and a new
factor, lie appropriateness was created. This new factor contained acceptability items 1
and 4 (Chris’s message was reasonable and acceptable) and competence items 2, 3, and 4
(Chris’s message was appropriate, Lkfllul, and competent). The Cronbach’s alpha for the
lie appropriateness scale was .80. All of the honesty items were maintained with an alpha
of .84. Four of the 10 items from the Lie Acceptability Scale were maintained, items 3,
5, 6, 8, 9, and 10 (see Appendix D). The alpha for the scale was .71. All items below a
factor loading of .60 were thrown out for the lie appropriateness and honesty scales. A
minimum factor loading of .42 was required for the items of the Lie Acceptability scale.
This was consistent with a prior test of the scale in which a factor loading of .40 was
considered acceptable (Levine, & McComack, 1991).

Bem’s Sex Role Inventory (1974) was used to determine the gender orientation of
the participants. The ratings for masculine attributes were summed and averaged. If the
mean of an individual’s score was greater than 4.89, s/he was described as demonstrating
high levels of masculine traits. Feminine attributes were computed in the same Manner
with means greater than 4.76 representing individuals with high levels of feminine traits.

Traditional males were males with high levels of masculine attributes. Untraditional

37

38
males rated high on femininity. Traditional females demonstrated feminine traits and
untraditional females, masculine traits. Traditional and untraditional individuals’
perceptions of messages were compared.
Perceptions of Lie Appropriateness and Honesty

AN OVA Analysis and Unegual Cell Sizes. Analysis of variance was used to test
all hypotheses. Because the interaction effects were unable to be assessed via SPSS
programs, contrast effects were used when necessary to test for the hypothesized
interactions. ANOVA tests only for disordinal interactions and the hypothesized
interactions were not disordinal.

Also, the original data created extremely unequal cell sizes. When using AN OVA
with unequal cell sizes, Type one error (demonstrating an effect when there is none)
becomes more likely. Therefore, a higher significance value was mandated. Instead of
significance being considered at p = .01 , a p value of .001 was necessary for any effect to
be considered significant.

Further, in terms of design, a 5 (message type) x 2 (male or female) was used to
test for effects from biological sex. For gender, a 5 (message type) x 2 (masculine or
feminine) design was used within the conditions of male or female. This was so that
untraditional and traditional individuals within each sex could be compared.

Honesg. A significant main effect was found message type upon honesty (F=
101.46, df = 4, 261 , p=.0001, eta squared .61) in that Completely disclosive messages

were found to be the most honest and Quality violations, the least. The pattern of means

39
(see Appendix E for the means for the Honesty conditions) and the main effect supported

the findings of McComack et. al. (1992). The similarities between the two studies were

 

 

 

as follows:

McComack etpl. (1992): Current Research:
Quantity Violations Most Honest Quantity Violations
Clarity/Manner Violations Clarity/Manner Violations
Relevance Violations. Relevance Violations
Quality Violations Leas: Honest Quality Violations

Figure 1: Comparison of McComack et.al. to Current Research

The rank order of perceptions of honesty supported the findings of McComack et. al.
(1992). When comparing gender orientation, the same main effect was found for
untraditional and traditional women (F= 62.82, df = 4, 201, p = .0001, eta squared = .59).
as well as for untraditional and traditional men (F = 31.20, p = .0001 , eta squared = .63).
HypothesesaLnd Research Questions. True to assumptions, women and men were
found to perceive Quantity violations as equally honest. Support was found for this affect
in that the ANOVA demonstrated that there were no significant differences between the
sexes in perceptions of honesty. Hypothesis one postulated that women would find
Quality messages to be less honest than would men, no support was found for this
hypothesis. Hypothesis two posited that women would find Manner violations to be
rpm honest than men. Again, no support was found for this hypothesis. Further, the

gender orientation of women or of men (traditional compared to untraditional) did not

40

influence their perceptions of the deceptive messages (Research questions 1, 2, and 3).
Overall a rank-order for the honesty of the messages was found.
Lie Appropriateness

Analysis of variance did not initially support an interaction between sex and lie
appropriateness. However, the hypothesized interaction would not be picked up by a
standard AN OVA because it would look for a disordinal interaction. The hypothesized
interaction was supported by the means (see Appendix F for the lie appropriateness
means). Contrast analysis was done based on the pattern of the means. Table 1 and

Figure 2 demonstrate the contrasts used for analysis:

Table l: Lie Appropriateness of Messages by Sex

 

Message:
1 2 3 4 5
Males 1 2 1 1 -1
Females -2 2 1 -2 -3

 

(Note: Message 1=Completely Disclosive, 2=Quantity Violations, 3=Manner Violations,
=Relevance Violations, and 5=Quality Violations.)

The following graph illustrates the differences in perceptions of message appropriateness

between men and women:

41

 

 

0. .lVales l

_1 W D FerMbs i

 

Lie Appropriateness
N
U

 

 

 

 

Figure 2: Lie Appropriateness of Messages by Sex

This model was found to be significant at F (1, 263) = 65.11 (p=.00001, eta squared =
.20). The contrast model explained more of the effect than the main effect for message
type (F = 12.82, df = 4, 263, p = .001, eta squared = .16). Therefore, the model was
accepted as significantly representing the differences between the messages by sex.
The model supported two rank orders for the appropriateness of the deceptive
message types. One rank order was found for women and one for men. Figure 3

illustrates these rank orders.

m Most Appropriate Women
Quantity Violations Quantity Violations
Comp. Dis/Manner/Relevance I Manner Violations
Quality Violations Relevance/Comp. Dis.

Least Appropriate Quality Violations

Figure 3: Rank Order of Appropriateness of Messages (Men and Women)

42

Hypothesis 1. It was hypothesized that women would perceive Quality violations
to be less competent and acceptable than would men. These assessments were collapsed
into the new variable, lie appropriateness. Women found Quality violations to be
significantly less appropriate than men and hypothesis two was supported.

Hypothesis 2. Although it was believed that women would rate Manner violations
as more appropriate (competent and acceptable) than men, no significant difference in
perceptions was found. This hypothesis was not supported.

Research Questions. Two separate analyses were completed on gender
orientation. Perceptions of lie appropriateness of the traditional and untraditional
individuals’ were compared for both men and women. ANOVAs completed for both
males and females - traditional vs. untraditional - demonstrated differing effects. No
main effects were found for untraditional and traditional males (F = 1.87, df = 5, 84, p =
.108, eta squared = .09). For women, a main effect for message type was found (F =
15.01, df = 4, 141, p = .001, eta squared = .29). However, the patterns of the means
(Appendix F) in both cases demonstrated interactions that standard ANOVA would not
test for. Therefore, contrast analyses were used to determine if the interactions between
the means were significant.

The contrasts in Table 2 and Figure 4 illustrate the interaction effects anticipated.

43

Table 2: Lie Appropriateness of Message Type by Gender Orientation (Females)

 

Message:
1 2 3 4 5
Trad. -1 1 l -1 -1
Untrad. 0 2 2 -1 -2

 

The following graph illustrates the differences in perceptions between traditional and

untraditional females:

 

 

4
a 3 2 2
g 2 J 1 1
1 t m""'
g 0 1.1T“. l
n- -1 4 2 3 - 1' . ..4 ,,____- .
a -1 - -
< -2 ..
.2
'J -3
-4

 

 

 

Figure 4: Lie Appropriateness of Message Type by Gender Orientation (Females)

The test of this model was significant at F = 71.13 (df = 1, 141, p = .00001, eta squared =
.32). The model explained more of the variance than the combined main effects found by
the initial ANOVA (F = 12.26, df= 5, 141, p = .0001, eta squared = .30) and was

considered to represent a better fit with the data.

44
Again, two rank orders were discovered, one for feminine women and the other

for masculine women. These two rank orders are demonstrated in Figure 5.

Traditional/Feminine Women Most Appropriate Untraditional/Masculine
Quantity/Manner Quantity/Manner Violations
Comp. Dis./Relevance/Qual. 1 Comp. Dis. Violations
Relevance Violations

Least Appropriate Quality Violations

Figure 5: Rank Order of Appropriateness of Messages (Traditional and

Untraditional Females)

Untraditional women rated themselves higher on masculinity traits and traditional
women rate themselves as higher on feminine traits. The model demonstrated that
untraditional, or masculine, women rated Quality violations (Research Question 1) as
less appropriate than traditional, feminine, females. Untraditional women also believed
that completely disclosive messages (Research Question 3) were mpg appropriate than
did traditional women. No differences in perceptions of appropriateness was found for
Relevance violations (Research Question 2), although both traditional and untraditional
women rated Relevance violations as less appropriate than Manner violations.
Untraditional females rated Manner violations (Research Question 3) as more acceptable
than traditional. Further, untraditional women believed that Manner violations were as
appropriate as Quantity violations. Untraditional females rated the Quantity and Manner
violations as the _rr_ro_st appropriate messages and, they rated these messages as more

appropriate than traditional women.

45
One of these patterns were similar to the interactions found between traditional
and untraditional males. Again, contrast analysis was performed based on the patterns
demonstrated by the means (Appendix F). Table 3 and Figure 6 illustrate the contrasts

used.

Table 3: Lie Appropriateness for Message Type by Gender Orientation (Males)

 

Message:
1 2 3 4 5
Trad. 0 2 O l -l
Untrad. -l l l -l -2

 

(Note: Message 1=Completely Disclosive, 2=Quantity Violations, 3=Manner Violations,
4=Relevance Violations, and 5=Quality Violations.)

46

 

Lie Appropriateness

 

 

 

 

Figure 6: Lie Appropriateness for Message Type by Gender Orientation (Males)

The model was found to be significant at F = 15.70 (df = 1, 84, p = .001, eta squared =

.15). In fact, the combined main effects were found to be insignificant at F = 1.87 (df =
5,84, p = .108, eta squared = .09). The model explained more of the variance explained
than either of the insignificant main effects (message type or sex). Therefore, the model

was accepted as a better fit with the data.

Differing rank orders were again found, this time between masculine and feminine

men. Figure 7 illustrates these rank orders.

47

 

Traditional/Masculine Men Most Appropriate Untraditional/Feminine Men
Quantity Violations Manner/Quantity
Relevance I Comp. Dis/Relevance
Comp. Dis/Manner Quality Violations
Quality Violations Least Appropriate

Figure 7: Rank Order of Appropriateness of Messages (Traditional and
Untraditional Males)

Traditional males were those individuals who exhibited higher levels of masculine
traits. Untraditional males demonstrated higher levels of feminine attributes. Similar to
untraditional vs. traditional females, untraditional males rated Quality violations as less
appropriate than did traditional males (Research Question 1). Also, untraditional males
rated completely disclosive and Quantity violations as lgs_s appropriate than did
traditional males. Untraditional males perceived Manner violations as mg; appropriate
than traditional males (Research Question 3), and as appropriate as Quantity violations.
This pattern is consistent with the untraditional females. Further, a difference in
perceptions was found for Relevance violations (Research Question 2) in that
untraditional males rated these violations as less acceptable than traditional males.

What was consistent between these interactions when comparing women to men
was that untraditional men and traditional women rated Quantity violations as less
appropriate than traditional males or untraditional females but all individuals - regardless
of sex or gender orientation, rated Quantity violations as more appropriate than
completely disclosive violations. Also, untraditional men and women rated Quality

violations as significantly leg appropriate than did their traditional counterparts.

48
Lie Acceptability Seek
A significant main effect was found for sex at F = 6.30 (df = l, 217, p = .01)
although the effect was small (eta squared = .02). For gender, results approached
significance with F = 5.30 (df = 1, 243, p = .02) but the effect size was also small (eta

squared = .02). Based on the low effect sizes no direct effects can be hypothesized.

Chapter 8: DISCUSSION

The two hypothesized effects did not find support in this research. However, the
research questions concerning whether there was a rank order for the appropriateness of
deceptive message types produced some unique and unexpected data. Further, the
distinctions found between women and men were found to be different when comparing
untraditional to traditional individuals of either sex. Lastly, the acceptability and
competency measurements were found to measure the same construct, appropriateness.

Of all of the observed trends in the data, four specific effects were particularly
intriguing. First of all, honesty assessments were all found to be consistent with prior
research on IMT and appropriateness assessments were not (McComack, et. al., 1992).
This is important in that IMT has been criticized for lacking explanatory power across
situations (Stiff, 1996). In this test a different situation was used and similar effects were
found for honesty perceptions of deceptive messages and dissimilar effects were found
for appropriateness evaluations of these messages. This implies that the honesty
perceptions of deceptive messages may remain consistent across situations, whereas,
perceptions of appropriateness may not. Individuals may perceive different messages to
be appropriate in different situations. This has extremely important implications for
further testing of IMT in that by varying the context of the deception, the appropriateness

of each deceptive message may change. By testing multiple situations, predictions could

49

50
be made of what to expect in other tests of similar situations. A taxonomy of the
appropriateness of certain types of deceptive messages within situations could be created.

Secondly, the rank order of the honesty assessments was a further important
finding. All individuals were able to distinguish the completely disclosive message as the
most honest, however, no individuals rated this type of message as the most appropriate.
In fact, the Quantity violation (“white lie”) was considered by all individuals to be more
appropriate than the completely disclosive message. This may indicate that in a situation
such as this it may be considered more appropriate to tell a little white lie than to tell the
truth. Perhaps in a situation such as the one given respondents in this research, it is more
appropriate to “save face” than to be completely honest. This has important implications
in that different deceptive messages may be preferred and that being truthful may not
always be preferred.

Third, although unexpected interactions occurred between both individuals of
differing sexes and gender orientations, one unexpected effect occurred between
traditional and untraditional individuals with regard to Quality violations. Females rated
these messages as significantly less appropriate than did men, therefore, it would be
expected that more feminine individuals would rate these messages as less appropriate.
However, this was not found to be the case. The Quality violation was considered to be
significantly more inappropriate for untraditional individuals of either sex. By definition,
untraditional men exhibited feminine traits and untraditional females demonstrated

masculine traits. Therefore, it appears that perhaps what seemed like a simple sex effect

51
may not be so simple. Perhaps individuals who adopt untraditional behaviors become
more sensitive to the feelings of other individuals and believe that to totally lie to another
individual could be hurtful and thus, unacceptable. Or, perhaps by being more open
about the kind of individual that s/he wants to be, regardless of socialization, an
untraditional person is in effect, more open and honest. Untraditional individuals may
place a value on honesty that is thus different from more traditional individuals who
remain in roles dictated to them by society.

This has extremely important implications in that individuals do not perceive the
world based on their sex but on the gender they identify with. Perhaps research such as
this can emphasize the importance of including the gender orientation of individuals and
not just their sex.

Fourth, individuals perceived the construct of acceptability as measuring the same
construct as competence. Appropriateness came from the blend of these constructs,
although they were believed to measure different perceptions. The acceptability of a
message was intended to assess whether a deceptive message was morally right or wrong.
Competence was initially intended to measure the perceived effectiveness of the
deceptive message. These evaluations appear to measure distinct constructs, however,
perhaps in deceptive situations, these constructs overlap. A message that may be
perceived to be more unacceptable or “wrong” may also be perceived to be less effective
and competent. This may be because deception creates an emotional response. If a

message is perceived as morally “wrong” it could not possibly be considered competent

52
as well. Also, the reverse could be true, i.e., when a message is considered to be very
ineffective it may also be perceived as “wrong.”

What is also interesting about these ratings is the fact that even with the
completely disclosive message, the assessments of competence and acceptability were
consistent. This may lend support to the fact that these constructs may not in fact be
distinct. However, this may also demonstrate the powerful effect of the context of the
messages. Completely disclosive messages were found to be less appropriate, i.e.,
competent and acceptable, when they could possibly be hurtful for the receiver.
Therefore, a completely honest message may be deemed incompetent for the same reason
that some deceptive messages were. An individual may perceive that a message which is
hurtful is “wrong” and thus, is also not competent.

The findings of this research are very thought provoking, however, two
limitations exist which suggest further research and analysis. One major limitation of this
research was that although significant differences were found between individuals of
differing sex and gender orientation, racial differences were not examined. It is well
accepted that individuals’ perceptions of their world are deeply influenced by their ethnic
origins. The author, being primarily of Polish descent, carries certain biases which alter
the way information is perceived. Also, being a Caucasian effects the way that
information is processed. This investigation was primarily on Caucasians, with other
racial groups not being well represented. Therefore, this may be a study of what

Caucasian individuals believe to be honest and appropriate in terms of deceptive

53
messages. This may be a serious limitation of this research which could be remedied
with future exploration. By making the effort to closely investigate the differences
between individuals of various ethnic backgrounds, perhaps unique and useful
information on the perceptions of deceptive messages will be gained.

A second limitation is the need to examine the perceptions of the androgynous
individuals. These individuals were not the focus of this research because this effort was
to look for the greatest differences between individuals of the same biological sex in
order to demonstrate that such differences did exist. This demonstrated, the door is open
for closer examination of the androgynous individuals which can be done by extracting
their data out from the current data set. This exploratory analysis may create interesting
rank orders of honesty and appropriateness which may or may not be similar to the other
findings.

This research demonstrates the importance of including variables such as sex and
gender orientation and perhaps even race, when looking at the perceptions of language.
The perception of a message is not a simple process, there may be much in an
individual’s past which will influence the way a message is decoded. It is important to

realize that this is a complex process and should be treated as such.

APPENDIX A

The Situation

You have known Chris for a year and have been dating for several months. This is the
closest romantic relationship to which you have been involved. Your family’s impression
of the people you date is very important to you. You decide that because of the
seriousness of your relationship with Chris, you decide to take her/him home to meet the
family. Mom prepared a special meal, one of your favorites. You thought that
everything was fine and your family really liked Chris. Chris seemed to have a good
time. As you were walking up to Chris and one of her/his fiiends after class just the other
day you overheard Chris making fun of your family. Chris did not know that you
overheard her/him. S/he said that your family was very nice to her/him. But, the food
was really bad and they acted really weird. Chris also said that s/he could not believe that
you were from the same family. You are very upset and you don’t know what to say.
Later at your place you and Chris are studying and it is still bothering you. You look at
Chris and say, “I really enjoyed you coming home with me. My family really liked you.
What did you think of them?” Chris looks at you and says:

54

APPENDIX B

Deceptive Messages:

Quantity Violation:
“I thought they treated me really nicely.”

Quality Violation:
“I had a great time! The meal was fantastic, and it was a blast hanging out with such
smart, educated people for an evening. They are totally like you, and I can see why you
are so close to them. You are just like them!”

Relation Violation:
“You asking me about your family reminds me, my folks want to know whether you want
to come spend part of Winter Break at their cottage. Are you interested? We would love
to have you.”

Manner Violation:
“I thought they were unique.”

Completely Disclosive Message:

“I am glad you asked, because it has been bugging me ever since that night, and I wanted
to tell you what I really thought, but was afraid that you would never forgive me if I was
honest. Maybe you won’t still, but I have to tell you. Your family treated me very
nicely, but to tell the truth, I thought your Mom’s cooking was terrible. I don’t want to
hurt your feelings or anything, and I hope you don’t dump me for saying this, but your
family seemed really stupid and uneducated. I can’t believe that you are from the same
family. Are you adopted?”

55

APPENDIX C

Acceptability:

I believe that her/his reasonable.

Chris’s response was immoral.

Chris’s message was polite.

I believe that her/his message was acceptable.
Chris’s response was wrong.

MPPNI“

Competence:

l. I think that Chris’s response was ineffective.

2. I feel that Chris’s message was appropriate.

3. I think that Chris’s message was skillful.

4. I think that Chris’s message was competent.

5. I think that Chris’s message was well managed.

Honesty:
l. I think that Chris’s message was misleading.
2. I think that her/his response was not deceptive.

3. I think that Chris’s response was truthful.
4. Chris’s message was honest.

56

APPENDIX D

Lie Acceptability Scale (Levine, McComack, & Avery, 1992):

Using the scale below, write in the answer that most appropriately represses your attitude
toward each of the items below.

Completely Disagree 1 2 3 4 5 6 7 Completely Agree

1. Never tell anyone the real reason that you do something unless it is useful to do
so.

2. Lying is immoral.

3. It is okay to lie in order to achieve one’s goals.

4. What people don’t know can’t hurt them.

5. The best way to handle people is to tell them what they want to hear.

6. There is no excuse for lying to someone else.

7. Honesty is the best policy in all cases.

8. It is often better to lie than to hurt someone’s feelings.

9. Lying is a cowardly thing to do.

10. There is nothing wrong with lying as long as you don’t get caught.

57

Males

Females

Trad.

Untrad.

Trad.

Untrad.

APPENDIX E

Assessments of Honesty of Messages: Means.

Comparisons by Sex:

Message:
1 2 3 4
23.91 14.17 12.89 11.61
N=23 N=18 N=26 N=18
23.81 14.08 14.00 10.50
N=32 N=37 N=27 N=34

Comparison of Traditional and Untraditional Males:

Message:
1 2 3 4
23.53 13.23 13.14 11.87
N=17 N=13 N=22 N=15
24,58 12.73 11.60 11.29
N=12 N=11 N=10 N=7

Comparison of Traditional and Untraditional Females:

Message:
1 2 3 4
24.29 14.17 13.64 10.86
N=24 N=29 N=l4 N=28
24.00 15.15 14.77 10.75
N=1 3 N=20 N=1 7 N=20

58

7.68
N=l9

6.70
N=37

7.82
N=17

8.64
N=11

6.26
N=27

7.05
N=19

59

APPENDIX F

Assessments of Lie Appropriateness of Messages: Means.

Comparisons by Sex:

Message:
1 2 3 4 5
Males 18.35 21.83 19.12 19.11 16.47
N=23 N=18 N=26 N=l 8 N=l 9
Females 15.22 21.55 19.93 14.83 13.73
N=32 N=38 N=27 N=35 N=37
Comparison of Untraditional and Traditional Males:
Message:
1 2 3 4 5
Trad. 18.88 21.69 18.91 19.67 16.82
N=17 N=21 N=22 N=15 N=17
Untrad. 16.58 20.09 21.00 16.14 15.64
N=12 N=11 N=10 N=7 N=ll
Comparison of Untraditional and Traditional Females:
Message:
1 2 3 4 5
Trad. 14.75 21.03 21.36 14.62 14.93
N=24 N=30 N=l4 N=29 N=27
Untrad. 16.54 22.95 22.00 14.60 13.74
N=17 N=21 N=17 N=20 N=19

REFERENCES

Aronfreed, J. (1968). Conduct and con_science/The socialization of internplized control
over behivior. New York, NY: Academic Press

Barbee, A.P., Cunningham, M.R., Winstead, B.A., Derlega, V.J., Gulley, M.R.,
Yankeelov, RA, and Druen, RB. (1993). Effects of gender role expectations on
the social support process. Joumpl of Social Issues. 49, 175-190.

Bern, S. L. (1974). The measurement of psychological androgyny. Journal of
Consulting and Clinical PsychologY. 42. 155-162.

Bond, C. F., Kahler, K. N., & Paolicelli, L. M. (1985). The miscommunication of
deception: An adaptive perspective. Joum_2_rl of Experimental Psychology, 21,
331-345.

Buller, D. B. & Burgoon, J. K. (1996). Another look at the information management:
A rejoinder to McComack, Levine, Morrison, and Lapinski. Communication

Monoggaphs, 63, 92-97.

Buller, D.B., Strzyzewski, K.D., and Hunsaker, PG. (1991). Interpersonal deception: II.
The inferiority of conversational participants as deception detectors.

Communication Monographs, 58, 25-40.

Clark, RA. (1993). Men’s and women’s self-confidence in persuasive, comforting, and
justificatory communicative tasks. Sex Roles 28 553-567.

 

Dalton, J .E. (1983). Sex differences in communication skills as measured by a modified
relationship inventory. Sex Roles 9 195-204.

 

deTurck, M.A., Harszlak, J .J ., Bodhom, D.J., and Texter, LA. (1990). The effects of
training social perceivers to detect deception from behavioral cues.

Communication Quarterly, 38, 189-199.

Duck, 8., Rutt, D.J., Hurst, M.H., Strejc, H. (1991). Some evident truths about
conversations in everyday relationships/ All communications are not created
equal. Human Communication Research 18 228-267.

 

Edelsky, C. (1976). The acquisition of communicative competence: Recognition of
linguistic correlates of sex roles. Merrill-Palmer Quprterlv. 22, 47-59.

60

61

Edwards, R., Honeycutt, J .M., and Zagacki, KS. (1989). Sex differences in imagined
interactions. Sex Rolgs. 21, 263-272.

Ekman, P. (1985). Telling lies: Clues to deceit in the marketplace, politics, and
marriage. New York: W.W. Norton.

Ekman, P., and Friesen, W. (1969). Nonverbal leakage and clues to deception.
Psychgrtrv. 32. 88-106.

Gregersen, E. (1979). Sexual linguistics. In J. Qrison, M.K. Slater, and L. Loeb Adler
(Eds), Language, sexgapdgenfier/Doesgla, differen_ce make a difference? New
York: Annals of the New York Academy of Sciences.

Grice, P. (1989). Studies in the way of words. Cambridge, Mass: Harvard University
Press.

Hoffman, ML. (1977). Personality and social development. Annual Review of
Psychology. 28, 295-321.

Hoffman, ML. (1977). Sex differences in empathy and related behaviors.
Psychological Bulletin. 84. 712-722.

Ickes, W. (1993). Traditional gender roles: Do they make, and then break, our
relationships? Journflf Social Issues. 49, p. 71-85.

Jacobs, S., Dawson, E. J ., & Brashers, D. (1996). Information manipulation theory: A
replication and assessment. Communication Monographs, 63, 72-82.

Kemper, S. (1984). When to speak like a lady. Sex Roles 10 435-443.

 

Lakoff, R. (1973). Language and woman’s place. Language in Society. 2. 45-79.
Lakoff, R. (1975). Language and wompn’s plgc_e. New York: Harper & Row.

Lakoff, R. (1977). What you can do with words: Politeness, pragmatics, and
perforrnatives. In A. Rodgers, B. Wall, and J .P. Murphy (Eds), Proceedings of

the Texas Conference on Performatives, Presuppositions, and Implications.
Arlington, VA: Center for Applied Linguistics.

Levine, T. R., & McComack, S. A. (1991). The dark side of trust: conceptualizing and
measuring types of communication suspicion. Communication Quarterly. 39.
325-340.

62

Levine, T.R., McComack, S.A., and Baldwin Avery, P. (1992). Sex differences in
emotional reactions to discovered deception. Communication Quarterly, 40, 289-
296.

Markman, H.J., Silvem, L., Clements, M., Kraft-Hanak, S. (1993). Men and women
dealing with conflict in heterosexual relationships. Journal of Social Issues. 49,
107-125.

McComack, SA. (1992). Information Manipulation Theory. Communication
Monographs. 59. 1-16.

McComack, S. A., & Levine, T. R. (1990). When lovers become leery: The relationship
between suspicion and accuracy in detecting deception. Communication
Monographs. 57, 220-230.

McComack, S.A., Levine, T.R., Solowczuk, K.A., Torres, H.I., and Campbell, D.M.
(1992). When the alteration of information is viewed as deception: An empirical
test of information manipulation theory. Communication Monographs. 59, 17-29.

McComack, S.A., and Parks, MR. (1986). Deception detection and relationship
development: The other side of trust. In M.L. McLaughlin (Ed.),
Communication Yearbook 9 (pp. 377-3 89). Beverly Hills, CA: Sage.

 

McComack, S. A., & Parks, M. R. (1990). What women know that men don’t: Sex
differences in determining the truth behind deceptive messages. Journal of
Social and Personal Relationships, 7, 107-118.

Q’Keefe, B. (1988). The logic of message design: Individual differences in reasoning
about communication. Communication Monoggaphs, 55, 80-103.

The Radom House College Dictionfl. (1982). New York: Random House, Inc.

Riggio, R., and Friedman, H. (1983). Individual differences and cues to deception.
Journal of Personality and Social Psychologv. 45, 899-915.

Simkins-Bullock, J .A., Wildman, BC. (1991). An investigation into the relationships
between gender and language. Sex Roles. 24. 149-160.

Stiff, J. B. (1996). Theoretical approaches to the study of deception: Comments on the
interpersonal deception theory. Communication Theog, 3, 289-296.

63

Stiff, J.B., Hale, J .L., Garlick, R., and Rogan, R.G. (1990). Effect of cue incongruence
and social normative influences on individual judgements of honesty and deceit.
The Southern Communication Journal 55 206-229.

 

Worell, J. (1993). Gender in close relationships: Public policy vs. personal prerogative.

Journal of Social Issues 49 203-218.

 

p:-_--:-.- ‘-

MICHIGAN srnTE UNIV. LIBRARIES
111111111111111111111!!!1111111111111111111111111111
31293015578051