UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Disentangling deceptive communication : situation and person characteristics as
determinants of lying in everyday life
Backbier, E.H.F.
Publication date
2001
Link to publication
Citation for published version (APA):
Backbier, E. H. F. (2001). Disentangling deceptive communication : situation and person
characteristics as determinants of lying in everyday life. Thela Thesis.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
Thee Extended Theory of Planned
Behaviourr as a Framework for the
Studyy of Lying Intentions
Mostt research on deceptive communication has focused either on verbal and non-verbal cuess to deception or on factors that enhance or inhibit people's ability to detect deception (e.g.,, Ekman & Friesen, 1974; Zuckerman, DePaulo & Rosenthal, 1981), as was already statedd in Chapter 1. In Chapter 2 we concluded that in addition to situational factors, personall factors are of importance to the judgement of the acceptability of telling lies. It is likelyy that personal factors are also of importance to the decision whether or not to tell a lie.. Little is known about the cognitive considerations that precede deceptive communication1.. What factors do people weigh when involved in the decision whether or nott to lie? Why do some people intend to lie in a specific situation and others not? Should thiss be attributed to differences in personality, to differences in cognitions, or to differencess in the perception of the situation, or perhaps to a mixture of these determinants? ?
11 Although the term lying is ordinarily used to refer to the utterance of false statements, we feel it is correct too include omission of information. Miller and Stiff (1993), who prefer to use the more general term 'deceptivee communication' because of the inclusion of the omission of information, define this term as
"message"message distortion resulting from deliberate falsification or omission of information by a communicator withwith the intent of stimulating in another, or others, a belief a belief that the communicator himself or herself does not beliefbelief'(p.'(p. 20). In the context of the present study we endorse Miller and Stiffs definition.
AA theoretical framework intended to provide a parsimonious explanation of informational andd motivational influences on behaviour is the theory of planned behaviour (Ajzen, 1985; Ajzenn & Madden, 1986; Madden, Ellen & Ajzen, 1992). The Theory of Planned Behaviour (TPB)) represents an extension of the original Theory of Reasoned Action (TRA). The TRA wass designed to be applied to behaviour entirely under the volitional control of the individuall (Fishbein & Ajzen, 1975; Ajzen & Fishbein, 1980). In the TPB a perceived controll variable has been added to account for behaviour not completely under volitional control.. Both theoretical models have been widely and successfully applied to account for aa variety of behaviours (see for a review on the TPB: Ajzen, 1991; see for a meta-analysis onn the TRA: Van den Putte, 1993).
Too our knowledge, only three studies are reported in the literature that tested the predictive powerr of the theories of reasoned action and planned behaviour with respect to deceptive communicationn (Beck & Ajzen, 1991; Randall & Gibson, 1991; Kurland, 1995). Beck and Ajzenn (1991), operationalised lying as 'using a false excuse to get out of taking a test or of turningg in an assignment on time'. Randall and Gibson (1991) asked nurses whether they wouldd report health professionals responsible for inadequate patient care. Kurland (1995) studiedd the ethical decision as to whether sales persons would disclose all of the available, ethicallyy relevant product information to a client, regardless of whether it was favourable orr unfavourable to the client's interest. When studying the determinants of deliberate falsificationn of information (Beck & Ajzen, 1991) or deliberate omission of information (Randalll & Gibson, 1991; Kurland, 1995) the theory of planned behaviour was successfullyy extended with the addition of a variable called 'moral obligation'.
Althoughh the results of these earlier studies are promising, we are still at the beginning of studyingg deceptive communication. We feel that more empirical testing is needed in order too establish what the role of perceived behavioural control and moral obligation are in the predictionn of lying (intentions). The present paper provides an additional study on the determinantss of the intention to deliberately falsify information, within the framework of thee theory of planned behaviour.
TheoryTheory of Planned Behaviour
Thee Theory of Planned Behaviour was designed to account for behaviour that is not completelyy under volitional control (Ajzen & Madden, 1986; Ajzen, 1991). Behaviour may
bee said to be completely under a person's control if the person can decide at will to perform itt or not perform it. Conversely, the more the performance of behaviour is contingent on thee presence of appropriate opportunities or on the possession of adequate resources (time, money,, skills, co-operation of other people, etc.), the less the behaviour is under volitional control.. The more resources and opportunities individuals think they possess, and the fewer obstacless or impediments they anticipate, the greater should be their perceived control over thee behaviour. Behavioural control can thus best be viewed as a continuum (Ajzen & Madden,, 1986). The TPB incorporates Fishbein and Ajzen's (1975) Theory of Reasoned Actionn designed for behaviour completely under behavioural control. The TRA assumes thatt a particular behaviour results from the intention to perform it. The behavioural intentionn is held to be determined by two conceptually independent factors: the evaluation off favourability or unfavourability of the behaviour (attitude) and the perceived social pressuree to perform or not perform the behaviour (subjective norm). In the case of behaviourr that is not completely under volitional control, perceived behavioural control, in additionn to attitude and subjective norm, will determine the behavioural intention.
TheThe Relevance of Moral Obligation
Whenn studying dishonest or moral behaviour some researchers successfully extended the theoryy of planned behaviour with a variable called 'moral obligation' (Gorsuch & Ortberg, 1983;; Beck & Ajzen, 1991; Randall & Gibson, 1991; Kurland, 1995: Manstead & Parker, 1995).. It refers to the personal feelings of moral obligation or responsibility to perform or nott perform a certain behaviour. In moral situations Gorsuch and Ortberg (1983) found that aa measure of perceived moral obligation was more highly correlated with intention than attitudee and subjective norm. Entering an index of perceived moral obligation in the regressionn equation after the TPB constructs, significantly increased explained variation in intentt (from as little as 1.6% in Randall and Gibson's (1991) study of 'ethical decision making'' to as much as 15.3% in Manstead and Parker's (1995) study of'overtaking on the inside').. Moral obligation appeared to be the most powerful predictor of intention in Kurland'ss (1995) study, while in Beck and Ajzen's (1991) study, perceived moral obligationn together with perceived behavioural control appeared sufficient to predict the intentionn to lie.
AssumptionsAssumptions and Hypotheses in the Context of Lying
Attitude,, subjective norm, perceived behavioural control and moral obligation are expected too be of varying importance to the prediction of lying intentions. Previous studies have shownn that some of the predictors add little or nothing to the explanation of the intention to omitt or falsify information. In order to enable our findings to be compared with the findingss of Beck and Ajzen (1991), Randall and Gibson (1991) and Kurland (1995), the contributionn of perceived behavioural control and moral obligation to the prediction of lyingg intentions will be tested hierarchically.
Withh regard to behavioural control it is conceivable that in some instances it may be sufficientt to have the will to tell or not to tell a lie, while in other instances control will dependd on internal and/or external factors. In those instances one will need to consider for instance:: whether or not one will blush, whether or not one will succeed in composing a believablee story, whether or not one has enough time to prepare a story, or whether or not thee other person will ask difficult questions or try to verify your story. Therefore, we hypothesisee that perceived behavioural control, in addition to attitude and subjective norm, willl add in varying degrees to the prediction of lying intentions.
Itt is assumed that the decision whether or not to tell a lie is salient when telling the truth or tellingg a lie will lead to different outcomes. Or to be more precise, when telling the truth willl be hurtful and/or against one's own self-interests and telling a lie will be kind and/or beneficiall for one's own self-interests. Because a choice has to be made between two optionss that are usually regarded as negative, a measure of personal moral values is expectedd to add to the explanation of lying intentions. While attitude, subjective norm and perceivedd behavioural control are assumed to function as facilitators of intention, we think thatt moral obligation will function as an inhibitor of intention. As a possible inhibitor of intention,, the direction of moral obligation is expected to be crucial to the intention to lie. Inn addition to testing the contribution of moral obligation to the explanatory power of the modell of planned behaviour, we will test the following hypotheses: 1) If one's moral values inhibitt one from lying about something or some reason, one will intend not to lie; and 2) If one'ss moral values do not inhibit one from lying, lying intentions will be greater with more positivee attitude and perceived behavioural control scores and less negative social and personall disapproval scores.
TheThe Influence of the Situation
Inn order to be able to predict lying (intentions) successfully, one should also consider the situation.. Lying in general usually is regarded reprehensible, while within a certain situationn it may be regarded acceptable or even preferable (by some individuals) to lie (see Backbier,, 1994; Chapter 2). People tend to lie in situations in which a communicative dilemmaa arises (e.g. Bavelas, Black, Chovill & Mullit, 1990); when the dilemma forces an individuall to choose between either a negative outcome or a negative act. Which answering strategyy one chooses depends in part on the kind of relationship people have (Backbier, 1994;; Metts, 1989). According to Wish and colleagues interpersonal relations vary on four dimensionss (Wish, Deutsch & Kaplan, 1976). Probably most relevant to the study of deceptivee communication is whether a relationship can be characterised as task-oriented or socio-emotionall and whether interactants are of equal or unequal status (see Meerum Terwogtt - Kouwenhoven, 1992).
Inn order to test the effect of situational characteristics on the cognitive determinants of deliberatelyy falsifying information, we decided to manipulate the interactional goal and the relationshipp between interactants using a scenario methodology (as did Randall & Gibson, 19911 and Kurland, 1995). We created a scenario in which the subject had to decide about engagingg or not engaging in a lie in order to get a job (scenario 1) and a scenario in which thee subject had to decide about engaging or not engaging in a lie in order not to miss a partyy (scenario 2). It is expected that lying about the length of the period you are available too do a job will be perceived as more negative, more disapproved of, more difficult, and moree morally wrong, and this will result in a weaker tendency to lie compared to the situationn in which the reason for lying is not wanting to miss a party. To be able to test whetherr a difference in 'legitimate' status between the liar and the deceived will cause a shiftt in the cognitive determinants of lying, in the second scenario the subject had to decide aboutt engaging or not engaging in a lie either to his or her father (an interaction partner of unequall status), or his or her sibling (an interaction partner of equal status). It is expected thatt lying to a father will be perceived as more negative, more disapproved of, more difficult,, and more morally wrong, and this will result in a weaker tendency to lie than whenn a sibling is involved.
Too summarise, the aim of the present study is to arrive at a better understanding of what peoplee take into consideration when in a given social situation they do or do not intend to
lie.. To identity relevant determinants of lying we will make use of an extension of the modell of planned behaviour (Ajzen & Fishbein, 1980: Ajzen, 1991; Beck & Ajzen, 1991). Wee will assess the relevancy of attitude, subjective norm, perceived behaviour control and morall obligation with respect to lying intentions in three different fictitious situations. The situationss will be constructed such that identification of the research participants (students) withh the description of the dilemma and the communication strategy is facilitated. By measuringg the extended TPB model towards lying intention for three different situations, wee are able to investigate what determinants' influences remain stable among the situations andd what determinants' influences change depending on the situation under study.
M e t h o d d
Participants Participants
AA total of 92 first year psychology students of the University of Amsterdam participated in thee study as part of a course requirement. About one third of them was male and two thirds female.. Data collection took place during one session halfway through the second trimester. .
ProcedureProcedure and Scenarios
Subjectss filled out two questionnaires, each relating the different variables of the extended modell of planned behaviour to a specific scenario in which one person lies to another. The firstt scenario was the same for all subjects (lying in order to get a job). In the second scenarioo (lying in order not to miss a party) the relationship between the liar and the deceivedd was varied in order to assess the influence of this difference on the variables in thee model. As a result, in one case the status of the interactants could be described as equal (sibling),, in the other as unequal (father). Subjects were asked to read the scenario carefully,, to imagine themselves in the situation, and to keep the scenario in mind while answeringg the questions. On the top of each page of the questionnaire the same scenario wass repeated. The first scenario in which the interactants are of unequal status, read as follows: :
'Friends'Friends of your parents offer you a job for six weeks during the summer holidays. AlthoughAlthough you know in advance that you will be available for only three weeks, you lielie to them in order to get the job, saying you will be therefor the whole period'.
Thee scenario provided as second was randomly assigned to subjects (scenario 2a: n=46 and scenarioo 2b: n=46). The second scenario read as follows (the equal status condition written inn between brackets):
'Your'Your father (brother/sister) will celebrate his (her) birthday next Saturday night. ThatThat same night, however, you are invited to a party you wouldn 't want to miss out on.on. You decide to call your father (brother/sister) and to tell him (her) that, since youyou have to prepare for an important exam, you will not be able to come.'
Forr each scenario, intention, attitude, subjective norm, perceived behavioural control and morall obligation were measured respectively.
IndependentIndependent Measures
AttitudeAttitude is defined as the sum of five items of the following form: 'Lying in this situation
wouldd feel....', followed by the evaluative adjectives 'bad-good', 'unpleasant-pleasant', 'boring-nice',, 'unattractive-attractive', and 'unworthwhile-worthwhile', all scored from 1 to 7.. Cronbach's alpha was .78 for the items measured with regard to the first scenario and .85 withh regard to the second scenario.
SubjectiveSubjective Norm was measured by the item 'Most people that are important to me
wouldd disapprove if I lied in this situation' and was rated from 1 'extremely unlikely' to 7 'extremelyy likely'.
PerceivedPerceived Behavioural Control was defined as the sum of the following two items:
1)) Lying in this situation is easy for me, and 2) Even if I had good reasons, I would not be ablee to lie in this situation. Item 1 was scored from 1 'extremely unlikely' to 7 'extremely likely',, the score of item 2 was reversed. Cronbach's alpha for the two items was .74 for the firstt scenario and .83 for the second scenario.
Followingg Beck and Ajzen (1991), Moral Obligation to refrain from lying was measuredd by means of the following three items: 1) 'Lying in this situation goes against my principles',, 2) 'It would be morally wrong for me to lie in this situation', and 3) 'I would not feell guilty if I lied in this situation'. The answers were rated on a scale ranging from 1 'extremelyy untrue' to 7 'extremely true' (the score of item 3 reversed). Moral obligation was definedd as the sum of the responses to these three items. Cronbach's alpha was .66 for the firstt scenario and .74 for the second scenario.
DependentDependent Measure
IntentionIntention was measured by the following item: 'If I were the person in this story (scenario
1)) I would lie in order to get the job', or, (scenario 2) I would lie in order to go to the party'. Intentionn was rated on a scale ranging from 1 'extremely unlikely' to 7 'extremely likely'.
Thus,, higher scale-scores for attitude, perceived behavioural control and intention and lowerr scale-scores for subjective norm and moral obligation reflect a more positive evaluationn of lying.
R e s u l t s s
DifferencesDifferences in Determinants between Scenarios
Tablee 3.1 provides the means and standard deviations of the direct measures of the extendedd TPB model split by scenario. One-sided t-tests revealed significant differences in thee hypothesised direction between scenario 2a and scenario 2b for perceived behaviour control,, moral obligation and intention. Lying to a sibling was perceived to be more controllablee (/ (90)= -3.00, p < .01) and less morally wrong (/ (90)= 2.10, p < .05) than lyingg to a father. Also subjects were more inclined to lie to a sibling than to a father (/(90)= -2.05,, p < .05). Although hypothesised, no significant differences between the two scenarioss were found for attitude and subjective norm.
Threee significant differences were found when paired one-sided t-tests were calculated comparingg the subgroup means of scenario 1 with the mean of scenario 2a, respectively scenarioo 2b. Again the hypothesised differences between the two scenarios for attitude and subjectivee norm were non-significant. As hypothesised, the mean intention to lie to one's fatherr in order not to miss a party was significantly greater than the mean intention to lie to one'ss parents' friends in order to get a job (t (45)= 1.79, p < .05). Lying to a sibling in order nott to miss a party was perceived less socially disapproved of (/ (45)= 1.95, p < .05) and lesss morally wrong (/ (45)= 2.09, p < .05) than lying to one's parents' friends in order to get aa job.
Tablee 3.1 Means and Standard Deviations of the Behavioural Determinants of Lying split
byby Scenario
Scenarioo 1: Scenario 2a: Scenario 2b:
tyingtying lying to a tying to a inin father in order sibling in order orderorder to get a job not to miss a party not to miss a party
MM SD M SD M SD
Attitude e Subjectivee Norm
Perceivedd Behavioural Control Morall Obligation Intention n 2.42 2 5.79 9 3.37 7 5.58 8 2.18 8 0.97 7 1.46 6 1.62 2 1.15 5 1.43 3 2.50 0 5.57 7 2.98a a 5.60a a 2.26a a 1.13 3 1.34 4 1.70 0 1.13 3 1.42 2 2.75 5 5.35 5 4.026 6 5.026 6 2.96A A 1.18 8 1.42 2 1.63 3 1.50 0 1.81 1
Note:: The means with a significantly differ from the means with b: t-test for independent samples, /K.05. .
RegressionRegression Analysis
Too examine the unique contribution of each of the four predictors in explaining the intentionn to lie in each of the three situations, the data were subjected to hierarchical regressionn analysis. First, the predictors part of the theory of reasoned action were entered inn the regression equation. Second, perceived behavioural control was added to attitude and subjectivee norm. Third, moral obligation was added to the predictors of the theory of plannedd behaviour. The results are presented in Table 3.2.
Withh respect to lying in order to get a job (scenario 1), attitude and moral obligation made aa significant contribution to the prediction of intention. The four predictors of the extended TPBB explained approximately 59% of the variance in lying intention (F(4, 87)= 30.71,/? < .001).. With respect to lying to a father in order not to miss a party (scenario 2a), only attitudee and subjective norm added significantly to the regression equation. The four predictorss of the extended TPB explained approximately 64% of the variance in lying
intentionn ( F (4, 41)= 18.01, p< .001). When the same lie is told to a sibling (scenario 2b), onlyy attitude added significantly to the prediction of intention. The total amount of variancee explained by the four predictors in the equation was approximately 7 1 %
(F(F(4,(4, 41)= 25.18,/; <.001).
Tablee 3.2 Hierarchical Regression Analysis for Lying Intentions split by Scenario
Scenarioo 1:
lying lying in in
orderorder to get a joh
rr b overall R2
Scenarioo 2a:
lyinglying to a fatherfather in order notnot to miss a party
rr h overall R2
Scenarioo 2b:
lyinglying to a siblingsibling in order notnot to miss a party
rr b overall R2
StepStep 1: Theory of Reasoned Action
Attitudee .69 .59** Subjectivee Norm -.51 -.17 .50 72 2 60 0 .56** * -.34* * .61 1 .82 2 -.40 0 90** * .12 2 .69 9 StepStep 2: Theory of Planned Behaviour
Attitudee .69 .50** Subjectivee Norm -.51 -.06 PBCC .57 .29* .55 72 2 .60 0 52 2 .51** * -.32* * .11 1 .61 1 .82 2 -.40 0 .61 1 .83** * .15 5 .122 .70
StepStep 3: Moral Obligation
Attitudee .69 .42** Subjectivee Norm -.51 -.01 PBCC .57 .16 Morall Obligation -.68 -.30* .59 72 2 .60 0 52 2 60 0 .44** * -.26* * .09 9 -.19 9 .64 4 .82 2 -.40 0 .61 1 -.74 4 .70** * .17 7 .07 7 -.211 .71 Note:: PBC = Perceived Behavioural Control. **p <.01; *p <.05
Tablee 3 J Means and Analyses of Variance Results for Non-rejecters versus Rejecters Split
bvbv Scenario
Non-rejecters s (Loww MO)
Rejecters s
(Highh MO) eta2 2
ScenarioScenario 1: lying in order to get a job
Intention n Attitude e Subjectivee Norm
Perceivedd Behavioural Control Morall Obligation 2.86 6 2.94 4 5.06 6 4.22 2 Ml Ml 1.38 8 1.81 1 6.67 7 2.37 7 6.54 4 33.06 6 45.42 2 38.94 4 44.09 9 132.57 7 .000 0 .000 0 .000 0 .000 0 .000 0 .268 8 .335 5 .302 2 .328 8 .595 5
ScenarioScenario 2a: lying to a father in order not to miss a party
F(5,86)) = 27.36,/>< .001
Intention n Attitude e Subjectivee Norm
Perceivedd Behavioural Control Morall Obligation 2.95 5 2.84 4 5.05 5 3.84 4 4.64 4 1.62 2 2.19 9 6.04 4 2.19 9 6.49 9 12.64 4 4.00 0 7.16 6 14.00 0 92.69 9 .001 1 .052 2 .010 0 .001 1 .000 0 .223 3 .083 3 .140 0 .241 1 .678 8
ScenarioScenario 2b: lying to a sibling in order not to miss a party
F(5,40)) = 26.97,p<.001
Intention n Attitude e Subjectivee Norm
Perceivedd Behavioural Control Morall Obligation 3.86 6 3.39 9 4.93 3 4.87 7 4.05 5 1.56 6 1.77 7 6.00 0 2.69 9 6.54 4 28.41 1 36.80 0 7.10 0 33.76 6 91.54 4 F(5,40)= = .0000 .392 .0000 .455 .0111 .139 .0000 .434 .0000 .675 == 19.51,/K.001
TheThe Role of Moral Obligation
Inn order to be able to examine the inhibiting influence of moral obligation on the variables inn the model, the sample was split as near as possible to the 50th percentile of the scenario 11 moral obligation scale. In this way, 42 subjects were classified as rejecters (moral obligationn scores greater than 5.67) and 50 subjects as non-rejecters (moral obligation scoress less or equal 5.67) of lying in order to get a job (scenario 1). Twenty-two subjects weree classified as rejecters and 24 as non-rejecters of lying to a father in order not to miss aa party (scenario 2a), while 28 subjects were classified as rejecters and 18 as non-rejecters
off lying to a sibling in order not to miss a party (scenario 2b). First, the distribution of intentionn scores for rejecters and non-rejecters was inspected. It appeared that subjects scoringg high on moral obligation with regard to scenario I did not intend to lie in order to gett a job (intention scores were all under 3). The scores of non-rejecters ranged from 1 to 6.. Intention scores of rejecters with regard to scenario 2a ranged from 1 to 4 while those of
non-rejectersnon-rejecters ranged from 1 to 6. The same picture was found for intention scores with
regardd to scenario 2b. Intention scores of rejecters ranged from 1 to 3 and intention scores off non-rejecters from 1 to 7. In contrast to the results of the previous regression analyses, thesee findings thus imply that moral obligation does play an important role in determining thee intention (not) to tell a lie in all three situations.
Too show that scores of subjects who do morally reject lying within a specific situation differr significantly from scores of subjects who do not morally reject lying within the same situation,, multivariate analyses of variance were conducted on the mean scores of the five variabless of the extended TPB with personal moral norm (low/high moral obligation) as a between-subjectss factor. The results of these analyses are shown in Table 3.3. Rejecters do havee a more negative attitude, expect more social disapproval, perceive less behavioural controll and more personal disapproval than non-rejecters. The weak to strong associations (eta22 .083 to .678) between high and low moral obligation subgroups and the five variables off the extended TPB reflect that some variables are more important to the group differencess than others.
D i s c u s s i o n n
Thiss chapter reports on the predictive power of the extended theory of planned behaviour too lying intentions using a scenario methodology. The main objective was to arrive at an understandingg of which determinants remain stable and which change as a function of the situationn under study.
Thee means of the determinants of lying clearly showed our subjects' negative evaluation of lyingg as an option to solve the communicative dilemma sketched in the scenarios. Despite thiss general disapproval, however, some significant differences were found. Lying to a siblingg was perceived as more controllable and less morally wrong than lying to a father. In
linee with this, more subjects intended to lie to a sibling than to a father. The intention to lie too one's parents' friends in order to get a job appeared to be even lower than the intention to liee to one's father in order not to miss a party. Finally, lying to a sibling in order not to miss aa party was perceived less socially and morally wrong than lying to one's parents friends in orderr to get a job. This pattern indicates initial support for the hypothesised relationships betweenn situational factors and the cognitive determinants of lying.
Attitudee proved to be the most important predictor of lying intentions in all three situations.. Personal feelings of moral obligation had some predictive value for the intention too lie to one's parents' friends in order to get a job. Subjective norm had some predictive valuee for the intention to lie to one's father because one did not want to miss a party. When thee same lie was told to a sibling, however, neither moral obligation nor subjective norm seemedd to play an important role. In this situation, attitude was the only substantial predictorr of the intention to lie. Perceived behavioural control did contribute to the predictionn of lying in order to get a job until moral obligation entered the equation. At none off the steps did it contribute to the prediction of lying in order not to miss a party. Althoughh these findings seem to suggest that lying to a father or a sibling in order not to misss a party is completely under volitional control, the failure to reach the status of a significantt predictor probably should be attributed to the relatively strong inter-predictor correlationss (correlations range from from -.38 to -.81). Moral obligation treated as an inhibiting factor,, showed that people who do reject lying within a specific situation perceived relativelyy little behavioural control while people who do not reject lying within the same situationn perceived relatively much control. Given the fact that the level of perceived behaviourall control and moral obligation do vary with intention, we can only conclude that bothh factors are salient in the decision process.
Althoughh the TPB does correctly answer the question which determinants best predict the reportedd lying intentions, it does not necessarily give a correct picture regarding the relevancerelevance of the predictors. Stated otherwise, the TPB has clear disadvantages as a regressionn model. When the correlation between attitude and intention as well as the
correlationn between attitude and the other three predictors are rather strong, as is the case inn the present study, the role of moral obligation and perceived behavioural control are obscuredd by attitude already being in the regression equation. Because the studies of Beck andd Ajzen (1991), Randall and Gibson (1991), and Kurland (1995) also revealed strong
inter-predictorr correlations, the regression analyses they conducted could have lead to incorrectt conclusions regarding the salience of the predictors of lying intentions as well. Moreover,, because of the repeated occurrence of strong inter-predictor correlations we believee the problem to be rather inevitable. Since lying is generally disapproved of by most people,, while at the same time being a behaviour people have a lifetime of experience with,, occasions in which there will be little agreement between the four determinants of lyingg intention are probably very rare. In other words, we expect the different conclusions regardingg the relative impact of the four predictors from the regression analyses and the strategyy of splitting the subjects into rejecters and non-rejecters of lying in a specific situationn to return.
Beckk and Ajzen (1991) studied the determinants of three kinds of dishonest behaviour (cheating,, shoplifting and lying), while we studied one kind of dishonest behaviour (lying) inn three different situations. This difference in approach reflects a difference in opinion withh regard to the causes of dishonest behaviour. Beck and Ajzen consider dishonest behaviourr to result from a disposition to behave dishonestly, implying that individuals with aa disposition for cheating, shoplifting and/or lying, will cheat, shoplift and/or lie whenever theyy have the opportunity to do so. We, on the other hand, consider the perceived necessity forr cheating, shoplifting or lying to be of much more importance than the opportunity to do so.. For instance, the perceived necessity for cheating will be preceded by the failure to preparee for an exam, the necessity for shoplifting by a bad financial situation and the necessityy for lying by a conflict between one's own interests and the interests of another person.. Moreover, the need for behaving dishonestly can easily be in conflict with one's generall opinion towards a specific kind of dishonest behaviour. The situation one ends up inn may for instance (temporarily) alter one's opinion about lying. That is, people who generallyy condemn lying to friends, will end up lying to a friend when in that specific instancee they consider it the only way to prevent their friend from being hurt. As our study wass able to show by varying situations instead of behaviours, some people who intended to liee towards their parents' friends did not intent to lie to their father (and vice versa). Therefore,, not only the individual him- or herself but also the situational demands precedingg the communication of a truthful or untruthful message are an important cause of lying. .
