A Test Of The Argument Engagement Model In Romania

University of Windsor
University of Windsor
Scholarship at UWindsor
Scholarship at UWindsor
OSSA Conference Archive OSSA 9
May 18th, 9:00 AM - May 21st, 5:00 PM
A test of the argument engagement model in Romania
Ioana Cionea
University of Maryland
Dale Hample
Fabio Paglieri
Goal-Oriented Agents Laboratory
Lilian Bermejo-Luque
Follow this and additional works at: https://scholar.uwindsor.ca/ossaarchive
Part of the Philosophy Commons
Cionea, Ioana; Hample, Dale; Paglieri, Fabio; and Bermejo-Luque, Lilian, "A test of the argument
engagement model in Romania" (2011). OSSA Conference Archive. 6.
https://scholar.uwindsor.ca/ossaarchive/OSSA9/papersandcommentaries/6
This Paper is brought to you for free and open access by the Conferences and Conference Proceedings at
Scholarship at UWindsor. It has been accepted for inclusion in OSSA Conference Archive by an authorized
conference organizer of Scholarship at UWindsor. For more information, please contact scholarship@uwindsor.ca.

Zenker, F. (ed.). Argumentation: Cognition and Community. Proceedings of the 9th
International Conference of the
Ontario Society for the Study of Argumentation (OSSA), May 18-21, 2011. Windsor, ON (CD ROM), pp. 1-15.
IOANA A. CIONEA, DALE HAMPLE, FABIO PAGLIERI
Department of Communication
College Park, MD 20742-7635
United States
icionea@umd.edu
Department of Communication
College Park, MD 20742-7635
United States
dhample@umd.edu
Goal-Oriented Agents Laboratory
Istituto di Scienze e Tecnologie della Cognizione
Consiglio Nazionale delle Ricerche
Via S. Martino della Battaglia
44, 00185 Roma
Italy
fabio.paglieri@istc.cnr.it
ABSTRACT: Hample, Paglieri, and Na’s (2010) model of argument engagement proposes that people en-
gage in arguments when they perceive the benefits of arguing to be greater than the costs of doing so. This
paper tests the model in Romania, a different culture than the one in which the model was developed, by
using a 2 (other arguer: friend or romantic partner) x 2 (topic of argument: private or public issue) design.
KEYWORDS: argument engagement, costs and benefits of arguing, arguing in Romania.
1. INTRODUCTION
A recent line of research (Hample, Paglieri, & Na 2010; Paglieri 2009; Paglieri &
Castelfranchi 2010) has begun investigating the reasons and consequences of engaging in
arguments. Our everyday interactions with others present numerous opportunities for ar-
guing, but we don’t follow up on all these possibilities. The scholars above suggest that
we pick our fights based on a cost-benefit analysis. People decide to engage in an argu-
ment if the perceived benefits are greater than the perceived costs of doing so.
Hample, Paglieri, and Na (2010) proposed a model of argument engagement in
which a person’s intent to engage in an argument can be predicted based on situational
factors (i.e., the argument topic), traits of the arguer (i.e., argumentativeness and verbal
aggressiveness tendencies), the expected costs of engaging in an argument (i.e., the cog-
nitive effort involved), and the expected benefits of arguing (i.e., the perceived gains).
Likelihood of winning the argument, perceived appropriateness of arguing within the
given situation, the expected level of civility of the argument, the perceived resolvability
of the argument, and the expected reasonability of the other person were also proposed as
factors that affect one’s intent to engage in an argument or not. The model proposed re-

CIONEA, HAMPLE & PAGLIERI
2
ceived partial support when tested in the United States, in the context of a private, public,
and workplace argumentation topic.
The purpose of this study is to test the model of argument engagement in a dif-
ferent culture and in a different argumentation context. Romania is a culture of conven-
ience, but if the model is correct, it should receive support in this culture too. The argu-
mentation context includes a private and a public topic of argument. Johnson (2002)
found that people showed different levels of involvement in an argument depending on
whether the topic was a private one (e.g., how to spend time together) or a public one
(e.g., the death penalty). The relationship between arguers is either of friends or of ro-
mantic partners. Our main goal is to test whether behavioral intent to engage in an argu-
ment can be predicted based on the factors proposed by Hample et al. (2010): argumenta-
tiveness, verbal aggressiveness, resolvability, appropriateness, civility, other’s reasonabil-
ity, cost, benefits, and likelihood of winning the argument. Our secondary goal is to offer
a comparison of Romanian and U.S. American argumentative and verbal aggressive
traits. There are no reports about Romanians on these measures, so we take advantage of
our data to extend the available knowledge about the population of our study.
2. ROMANIA: AN OVERVIEW
Romania is a country with an area of 238,391 square kilometers and approximately 22
million inhabitants, situated in Eastern Europe, neighboring Ukraine, Hungary, Serbia,
Bulgaria, and Moldova (The World Factbook 2011). The Romanian people was formed
after Roman legions conquered the ancient province of Dacia in 105-106 A.D. The Ro-
manian language has strong Latin influences and is part of the family of romance lan-
guages (Benedict 1972).
Romanian history has been tumultuous, marked by a constant struggle to defend
the territory from the expansionist tendencies of neighbors such as the Ottoman Empire
and the Austro-Hungarian Empire. The current state of Romania was formed in 1918
when the provinces of Walachia, Moldova, and Transylvania united (Calafeteanu, n. d.).
After War World II, Romania became a communist country. Nicolae Ceauşescu’s regime
brought social and moral degradation (Calafeteanu, n. d.), restricted the freedom of ex-
pression and civic involvement of citizens and isolated Romania from the Western world.
In 1989 Romania overturned this political dictatorship, but the transition from the old re-
gime proved to be very difficult. Although more than twenty years have passed since that
revolution, Romania is still in a state of transition, with acute economic and political issues.
During Ceauşescu’s communist regime public debate, intellectual public dia-
logue, and self-expression were almost non-existent. The political, social, and educational
changes that occurred in post-communist Romania created a new environment for the
young generation who grew up in the aftermath of the 1989 revolution. Scholars found
changes in youth’s system of values (Friedlmeier 2006; Săcară & Iacob 2002) and have
argued youth have more individualistic orientations, guided by influences of Western cul-
tures (Albu 2006). Eastern European adolescents (Romanians included) were found to be
similar in respect to their future-orientations to American adolescents (Alsaker & Flam-
mer 1999). In light of such conclusions, we expect that Romanian youth will exhibit a
willingness to express their mind, to stand up for themselves, and defend their ideas in
situations inviting arguing.

A TEST OF THE ARGUMENT ENGAGEMENT MODEL IN ROMANIA
3
To our knowledge, there are no studies on arguing behaviors in Romania. Thus,
we can speculate, at best, about the predictions of our model based on arguments made by
other scholars about Romanians. We expect the cost-benefit model of argument engage-
ment to receive support given results from studies on youth values. Săcară and Iacob
(2002) found that post-communist youth valued intelligence, professional competence,
and responsibility as opposed to communist youth who valued honesty, politeness, and
readiness to help. The authors argued that post-communist youth was more pragmatic
than communist youth. A cost-benefit analysis of a situation that invites arguing is no
doubt a pragmatic assessment of one’s chances in an argument. So, we can expect Roma-
nians to adopt a pragmatic approach when arguing with others.
We believe appropriateness of arguing is an important predictor of the intent to
engage in an argument among Romanians. The culture is full of social norms, customs,
and rites of how one ought to behave in various relationships and situations. Deviations
from appropriate behaviors tend to be ridiculed and shamed publicly (Albu 2006). Roma-
nians are also concerned with etiquette and polite interactions. For example, there is a
proper etiquette for how one ought to interact with the elderly (Benedict 1972). There are
also different pronouns for the second person, singular and plural, and one must be care-
ful about choosing the correct one in interactions (Albu 2006). We expect such considera-
tions to be reflected in an assessment of whether arguing is appropriate within a particular
situation, with a particular argumentation partner, and about a particular topic. However,
in the absence of more evidence about Romanian argumentation practices, we do not
have sufficient justification to formulate a strong hypothesis. We propose instead the fol-
lowing research question:
 RQ1: Is the behavioral intent to engage in an argument predicted by argumenta-
tiveness, verbal aggressiveness, cost of arguing, benefits of arguing, resolvability
of an argument, appropriateness of arguing, civility, other’s reasonability, and
likelihood of winning the argument?
In addition, we are interested in differences on the variables of interest based on whether
the argument is between friends or romantic partners and whether it is about a private or a
public topic. As such, we propose the following research question:
 RQ2: Is there a difference in the argumentativeness, verbal aggressiveness, cost
of arguing, benefits of arguing, resolvability of an argument, appropriateness of
arguing, civility, other’s reasonability, and likelihood of winning the argument
based on a) the topic of argument or b) the argumentation partner?
Finally, a secondary goal of our study is to report information about Romanians’ argu-
mentative and verbal aggressive traits. To our knowledge, such analyses have not been
conducted yet. To accomplish this goal, we compare our data from the present study with
the data collected by Hample et al. (2010). The research question we seek to answer is:
 RQ3: Are there any differences between Romanians and U. S. Americans on a)
argumentativeness and b) verbal aggressiveness?

4
3. METHOD
3.1. Participants and Sampling Methods
Participants in the study were 201 Romanians recruited in several ways. Online recruit-
ment based on the first author’s acquaintances and social networking sites yielded a sam-
ple of 61 participants. These participants completed an online version of the study. Stu-
dents recruited from courses at a large university in the North-Western part of Romania
yielded a sample of 58 participants. Finally, students recruited from an off-campus resi-
dence hall yielded a sample of 82 participants. Participants had various majors (e.g., eco-
nomic sciences, tourism, European studies, and business) and various occupations (e.g.,
doctor, nurse, sales consultant, project manager, and business manager). These latter two
samples completed a paper-and-pencil questionnaire, identical in content to the online
one.
Participants ranged in age from 18 to 64 years (M = 23.42, SD = 5.78). Partici-
pants were undergraduate students (N = 137), graduate students (N = 22), and working
adults (N = 40). One participant indicated a different occupation and another participant
did not answer this demographic question. Thirty-eight participants were male and the
remaining 163 were female. The ethnic distribution of participants was as follows: Ro-
manian (N = 188), Hungarian (N = 6), Rroma (N =1), and a combination of these ethnici-
ties (N = 4). Two participants did not indicate their ethnicity.
3.2. Procedures
A 2 (topic of argument: private or public) x 2 (relationship: friend or romantic partner)
experimental design was employed in the study. Participants were randomly assigned to
one of the four conditions describing a situation that invited arguing. Because not all pa-
per-and-pencil questionnaires were returned, the number of participants within each con-
dition was not equal. Thirty-nine participants were in the friends, private topic condition.
Sixty-nine participants were in the romantic partners, private topic condition. Twenty par-
ticipants were in the friends, public topic condition. Seventy-three participants were in
the romantic, public topic condition. The scenarios employed are described below under
Measures.
All participants completed demographic information and the argumentativeness
(Infante & Rancer, 1982) and verbal aggressiveness (Infante & Wigley, 1986) scales. Par-
ticipants then read one of the four hypothetical scenarios and answered questions about
that scenario. The questions pertained to their behavioral intentions, costs, and benefits of
engaging in an argument. Finally, participants assessed the realism of the scenario pre-
sented.
3.3. Argument Topics
Two of the hypothetical scenarios employed dealt with private topics and two dealt with
public topics. The private topic was about preference for movies whereas the public topic
was about preference for a particular political candidate. The scenarios were as follows:

5
Scenario I: Friends, private topic
You and a good friend of yours like movies a lot. You know a lot about various actors
and don’t miss a chance to go see the latest premiers. The two of you like different genres
of movies. It’s always been that way. One day, when you are spending some time togeth-
er, your friend makes a remark about how much better the movies he/she likes are, com-
pared to the ones you like, which are just terrible.
Scenario II: Romantic partners, private topic
You and your romantic partner (boyfriend, girlfriend, fiancé, husband, wife, etc.) like
movies a lot. You know a lot about various actors and don’t miss a chance to go see the
latest premiers. The two of you like different genres of movies. It’s always been that way.
One day, when you are spending some time together, your partner makes a remark about
how much better the movies he/she likes are, compared to the ones you like, which are
just terrible.
Scenario III: Friends, public topic
You and a good friend of yours keep up with what’s going on in the political arena and
often discuss politics. Elections are in the near future and it turns out you prefer different
candidates. That is not surprising as you’ve had different opinions and preferences about
political figures in the past. One day, when you are spending some time together, your
friend makes a remark about how much better the candidate he/she prefers is, compared
to the one you prefer, who is just terrible.
Scenario IV: Romantic partner, public topic
You and your romantic partner (boyfriend, girlfriend, fiancé, husband, wife, etc.) keep up
with what’s going on in the political arena and often discuss politics. Elections are in the
near future and it turns out you prefer different candidates. That is not surprising as
you’ve had different opinions and preferences about political figures in the past. One day,
when you are spending some time together, your partner makes a remark about how much
better the candidate he/she prefers is, compared to the one you prefer, who is just terrible.
3.4. Measures
The measures for the variables of interest were the same as the ones used by Hample et
al. (2010). A 1 to 5 Likert-type scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 =
agree, and 5 = strongly agree) was employed. The items were translated into Romanian
by the first author and back-translated into English by another Romanian with excellent
English language proficiency. Any disagreements were resolved by discussing the items
and agreeing on a final version of the translation.
Behavioral intent was the dependent variable. Eighteen items were employed to
measure whether participants intended to engage in an argument with the other person
about the topic in the hypothetical scenario.
Traits of the arguer consisted of argumentativeness and verbal aggressiveness.
Argumentativeness is considered a constructive trait, indicating one’s tendency to attack
another person’s position on an issue (Rancer & Avtgis 2006). The argumentativeness
scale consists of two sub-scales measuring one’s tendency to approach an argument or to

6
avoid an argument (Infante & Rancer 1982). Both sub-scales contain ten items. Verbal
aggressiveness is considered a destructive trait, indicating one’s tendency to attack “the
self-concepts of individuals” (Infante & Wigley 1986: 61). The verbal aggressiveness
scale consists of two subscales also, indicating one’s tendency for pro-social behaviors or
for anti-social behaviors. Both sub-scales contain ten items.
Cost of arguing was measured with ten items assessing the cognitive effort an
argument with the other person would involve. Benefits of arguing were measured with
six items assessing the potential benefits an argument with the other person would bring,
both personally, and for the relationships between the two people.
Resolvability was measured with six items assessing the chance of resolving the
argument if it occurred. Appropriateness was measured with seven items assessing the
appropriateness of having an argument with the particular person, at the particular mo-
ment, and on the particular topic. Civility was measured with ten items assessing the de-
gree of hostility, open-mindedness and cooperation between the two people if an argu-
ment occurred. Other’s reasonability was measured with six items assessing the degree to
which the other person would be stubborn, mature, tolerant, and willing to change his/her
mind if an argument occurred. Likelihood of winning was measured with eight items as-
sessing who had better arguments and evidence to support his/her position and who the
winner of the argument would be.
Finally, the realism of the scenarios was measured with five items assessing the
degree to which each scenario presented was realistic, reflected a real-life situation, and
whether participants were able to imagine themselves in the situation described in the hy-
pothetical scenario.
4. RESULTS
4.1. Scale Assessment
Reliability analyses and inter-item correlations were examined to assess the internal
structure of each scale used. These investigations were corroborated with confirmatory
factor analyses (CFA). The results informed which items should be retained in each scale.
Reliability analyses produced Cronbach’s alpha values. Inter-item correlations
revealed any problematic and non-significant correlations. CFA models were tested using
a covariance matrix of the scale items and their standard deviations as input data. The
maximum likelihood method was employed to estimate each model. Model fit was as-
sessed based on Hu and Bentler’s (1999) criteria: the comparative fit index (CFI) should
be greater or equal to .95, the standardized root mean square residual (SRMR) should be
less than or equal to .08, and the root mean square error of approximation (RMSEA)
should be less than or equal to .06. The minimum fit function chi-square and the contribu-
tion of each scale item to explain variance in the latent factor ( 2
R value) were examined
as well for each model. Table 1 contains the final model fit indices for each scale.
As a result of these analyses, several scales were adjusted. Four items were
dropped from the scale measuring behavioral intent. Two items were dropped from the
scale measuring argumentativeness, one item from each of the subscales. The verbal ag-
gressiveness scale’s 20 items were retained. Five items were dropped from the scale
measuring cost of arguing. One item was dropped from the scale measuring benefits of

7
arguing. One item was dropped from the scale measuring resolvability. All items measur-
ing appropriateness were retained. Five items were dropped from the scale measuring ci-
vility. Three items were dropped from the scale measuring other’s reasonability. Two
items were dropped from the scale measuring likelihood of winning. All items assessing
the realism of the scenarios were retained.
Following these analyses, aggregate variables were computed by calculating the
mean of the retained items. Table 2 contains the reliabilities, means, and standard devia-
tions of the aggregate variables both overall, for the entire sample, and within each condition.
4.2. Regression analyses1
An overall regression model with dummy-coded variables was conducted. Three dummy-
coded variables were used: one for the relationship between participants in the scenario
(friends or romantic partners), one for the topic of argument (private or public), and one
for the interaction between the dummies. Four variables predicted behavioral intent: ap-
propriateness (ß= .19, t(185) = 3.83, p < .001), civility (ß= -.11, t(185) = -1.99, p < .05),
other’s reasonability (ß = .17, t(185) = 2.97, p < .005), and likelihood of winning (ß = .20,
t(185) = 3.01, p < .005). The proportion of variance in behavioral intent explained by the-
se four variables was 33% (adjusted R2
= .33, F(15, 185) = 7.56, p < .001). Thus, the
structural equation for behavioral intention is BI = .19*Appropriateness - .11*Civility +
.17*Other Reasonability + .20*Likelihood of Win.
4.3. Path analysis
A measured variables path analysis with the first principal component of each variable of
interest was also conducted. First, we conducted a principal components analysis and re-
tained the factor scores for the component that explained the most variance in each varia-
ble. Second, we entered these principal components along with the three dummies created
in the regression analyses in a measured variables path model in which the exogenous
variables were allowed to co-vary.
The path model was just-identified, so fit indices are not available. Four paths
from appropriateness (p < .001), other reasonability (p < .01), likelihood of winning (p <
.001), and the dummy for the private argument topic (p < .05) to behavioral intent were
significant. The adjusted R2
for the model was .36. The structural equation was BI =
0.29*DummyPrivate + 0.31*Appropriateness + 0.18*Other Reasonability + 0.21 Likeli-
hood of Win.
In light of the regression analyses and the path analysis, we conclude as an an-
swer to RQ1 that behavioral intent is predicted by appropriateness of arguing, other’s rea-
sonability and likelihood of winning. The dummy for the private argument topic was sig-
nificant in the path model but it did not emerge as a significant predictor in the regression
analyses. So whether the topic of an argument makes a difference needs further research
before drawing a certain conclusion in this respect.
1
All regression coefficients reported are unstandardized.

8
4.4. ANOVAs
A one-way ANOVA was conducted to assess whether any differences existed between
the four conditions in respect to any of the variables in the study, given that the model
posits the situation will affect one’s intent to engage in an argument. Significant differ-
ences existed between groups for two variables: the avoidance dimension of argumenta-
tiveness (F(3, 197) = 3.65, p < .05) and cost of arguing (F(3, 197) = 6.37, p < .001).
Post-hoc multiple comparisons revealed that the mean of responses for argument
avoidance in the romantic partners, private topic condition was significantly lower than
the mean of responses for argument avoidance in the friends, private topic condition (M
difference = -0.40, p < .05). The mean of responses for argument avoidance in the friends,
private topic condition was significantly higher than the mean of responses for argument
avoidance in the friends, public topic condition (M difference = 0.64, p < .05). Finally,
the mean of responses for cost of arguing in the friends, public topic condition was signif-
icantly lower than the mean of responses for cost of arguing in the romantic partners,
public topic condition (M difference = -0.79, p < .005).
Thus, the answer to RQ2a is that the topic of argument makes a difference only
as far as argument avoidance is concerned (friends avoid private arguments less than they
avoid public ones). The answer to RQ2b is that the argumentation partner (friend or ro-
mantic partner) makes a difference in respect to argument avoidance (friends avoid pri-
vate arguments more than romantic partners do) and the cost of arguing (friends associate
less costs with arguing about public topics than romantic partners do).
4.5. Romanian argumentative traits
A secondary goal of our study was to compare Romanians and U.S. Americans on argu-
mentativeness and verbal aggressiveness. We compared the data for verbal aggressive-
ness and argumentativeness with the data collected by Hample et al. (2010). Significant
differences between Romanians and Americans were found for argumentativeness but not
for verbal aggressiveness. An independent samples t-test revealed that Romanians (M =
3.53, SD = 0.68) were significantly more likely to approach arguments than Americans
(M = 3.37, SD = 0.57) were, t(317) = 2.95, p < .005. Also, Romanians (M = 2.81, SD =
0.78) were significantly less likely to avoid arguments than Americans (M = 2.99, SD =
0.63) were t(308) = 2.88, p < .005. Thus, we conclude that Romanians are more argu-
mentative than U. S. Americans are (RQ3a) and that no significant differences exist re-
garding verbal aggressiveness (RQ3b).
5. DISCUSSION
The present study tested the argument engagement model proposed by Hample, Paglieri,
and Na (2010) in Romania, a culture different from the one in which the model was de-
veloped to assess whether the model’s predictions can be applied cross-culturally. Argu-
ing in a particular situation with a particular person and about a particular topic is a
choice that people make. Other options are available, including the option to avoid the
situation or the person, to refuse to engage in an argument, and to adopt some other forms
of response, such as passive aggressiveness. Our investigation suggests that people’s in-
tent to engage in an argument is affected by several important factors.

9
The model for argument engagement posited that behavioral intent to engage in
an argument is predicted by several factors: the arguer’s argumentativeness and verbal
aggressiveness, the expected cost of arguing, the perceived benefits of arguing, the likeli-
hood of winning the argument, the perceived appropriateness of arguing, the perceived
resolvability and the expected level of civility of the argument, and the expected reasona-
bility of the other person. This model received partial support in the present study. The
regression analyses and the path model revealed that, across situations and argumentation
partners, the most influential predictors of the intent to engage in an argument are the ap-
propriateness of arguing, the perceived likelihood of winning the argument, and the ex-
pected reasonability of the other person.
As previously explained, we expected appropriateness to be an important factor
that affects one’s decision to engage in an argument in Romania due to the cultural norms
that guide appropriate interactions with others. Paglieri (2009) explained that argumenta-
tion may be culturally encouraged or discouraged across different contexts. As such, cul-
tural differences in perceived appropriateness of arguing are likely to exist, with some
cultures sanctioning arguing as more appropriate than other cultures do. Our data suggest
that in Romania’s case weighing the appropriateness of arguing matters a lot.
The perceived likelihood of winning an argument as a predictor of the intention
to engage in an argument suggests that people assess their chances of coming out of such
an encounter victorious. As Hample et al. (2010) explained, winning may carry both an
instrumental goal and a positive feeling. People may evaluate their chance of winning
also in order to decide which arguments are worth pursuing. In other words, as Paglieri
(2009) put it, we “pick our fights.” If one perceives there is not a chance of winning, even
if arguing may seem appropriate, it won’t be pursued.
The expected reasonability of the other person also matters when deciding
whether to engage in an argument or not. In our study, the more reasonable the other per-
son was expected to be, the higher the chance that one would engage in an argument. This
finding suggests that when the other person is perceived to not be reasonable, people are like-
ly to decide against arguing because they deem the cause as lost. There is no point in trying to
argue with someone who will not change one’s mind despite good arguments and evidence.
In addition to these three factors, the significant path result from the dummy for
the private argument topic to behavioral intent deserves further investigation. In our
study, this path suggests that private topics are likely to determine argument engagement.
This result is consistent with Johnson’s (2002) findings that people were more engaged in
arguments that concerned private topics than in arguments that concerned public topics.
Moreover, the post-hoc comparisons suggest romantic partners are significantly less
avoidant of discussing private topics than friends are. This may be the case due to the dif-
ferent nature of the two relationships. Romantic relationships involve a process of ac-
commodation in which the two partners negotiate their relationship, including their stance
on issues such as personal preferences. Arguments on these topics reveal information
about the other person and may be necessary to develop a functional relationship. Friend-
ships, however, do not require agreement on such issues for their continuation. In fact,
according to the post-hoc comparisons, friends avoid arguing about private issues signifi-
cantly more than about public issues. Public topics arguments can be perceived as friend-
ly exchanges whereas private topics arguments may be interpreted as personal attacks or
criticisms that could damage the relationship.

10
The fact that civility was a significant predictor of the intent to engage in an ar-
gument in the regression equation needs further attention as well. The fact that the ex-
pected level of civility may contribute to one’s intent to engage in an argument makes
sense, but the negative coefficient in this equation is puzzling. In other words, we find it
puzzling that Romanians were more likely to engage in an argument when they expected
a lower level of civility of the argument. A possible interpretation of this finding is that if
people go through the trouble of starting an argument, then they want to hash it all out,
including yell at the other person, and are ready for the argument to involve lack of toler-
ance and negativity. Beyond this speculation, however, it is clear that more research is
needed to understand how Romanians approach such arguments and what their expecta-
tions about the civility of an argument are.
The results of our study did not offer any support for the personal traits of an
arguer as factors that affect one’s decision to engage in an argument. Nor were the per-
ceived costs and benefits predictors of this decision. These results are consistent though
in large proportion with Hample et al.’s (2010) model in which appropriateness of argu-
ing and likelihood of winning were the two predictors of the intent to engage in all three
conditions. However, unlike the Hample et al. (2010) study, the proportion of explained
variance in the intent to engage based on these predictors was smaller. One possible ex-
planation may be the difference in sample sizes, as our study had a much lower sample
size and contained individuals from a different culture. The R2
statistic is sample specific
(Hanushek & Jackson 1977). A second possible explanation is that our model leaves out
important variables that affect one’s decision to engage in an argument in Romania, given
that the residual error variance in the path model was significant (p < .001). Therefore, a
more thorough investigation of culturally specific factors is needed.
Finally, as the population of our study was Romanian, we draw a few conclu-
sions about arguing behaviors in Romania. Our results indicate that Romanians are more
argumentative than U. S. Americans are. Previous studies that have compared individuals
from different cultures on argumentative and verbal aggressive traits have explained their
results based on the individualism-collectivism dimension (Bresnahan, Shearman, Lee,
Ohashi, & Mosher 2002; Prunty, Klopf, & Ishii 1990, 1991). According to this dimen-
sion, Romanians are more collectivistic than U.S. Americans are (Hofstede 2001). Our
finding contradicts the conclusion that people from collectivistic cultures are less argu-
mentative than people from individualistic cultures. A possible explanation is that the ori-
entation towards individualism and collectivism among Romanians has changed from the
time Hofstede’s research was conducted. This explanation is supported by Albu’s (2006)
conclusions that Romanian youth has become more individualistic. Another possible ex-
planation is that arguing has a different role among Romanians than among Americans.
This idea needs further research, especially from an emic perspective, on the functions of
argument in Romania.
5.4. Limitations and directions for future research
The present study has several limitations that must be taken into account. First, hypothet-
ical scenarios were used rather than having participants engage in an actual interaction.
The perceived realism of the scenarios, however, gives us assurance that participants be-
lieved the situations described were realistic and could put themselves in those situations

11
because the mean scores for all scenarios are above the scales’ mean. Second, the sample
in our study was mostly female. As such, the results of the study may be better interpret-
ed as reflecting females’ perspectives on the decision to engage in arguments. Third, the
return rate for our paper questionnaires resulted in a disproportionate sample size for the
friends conditions as compared to the romantic partners conditions, which affected the
data analyses we were able to conduct. Finally, the reliability of our measures was prob-
lematic in two instances. Some of these issues may have been caused by translation inac-
curacies whereas others may be reflective of problems with our scales whose validity and
dimensionality should be further assessed.
The results of the present study have implications for the study of argument en-
gagement and for the study of arguing behaviors in Romania. Our results suggest that the
argument engagement model can be helpful in predicting people’s intent to engage in an
argument in various situations and with various people, but that the factors hypothesized
to affect this intent may need revision. More studies are needed to refine this model in
respect to the variables believes to affect the behavioral intent to engage in an argument
and in respect to the scales used to measure these variables. The results also suggest that
the model can be used across cultures. Cross-cultural tests of the model should pay care-
ful attention to the translation of the materials in the native language and should supple-
ment the core measures with culturally specific measures that can capture the peculiar
mechanisms involved in arguing in a specific culture. Finally, our study indicates that Roma-
nians are more argumentative than U. S. Americans are, which calls for more attention to the
specific understanding of arguing, its functions, and consequences in Romanian society.

12
REFERENCES
Alasker, F.D., and Flammer, A. (1999). Cross-national research in adolescent psychology: The Euronet
project. In Alasker, F.D. & Flammer, A. (eds). The adolescent experience: European and American
adolescents in the 1990s (pp. 1-14). Mahwah, NJ: Lawrence Erlbaum.
Albu, M. (2006). Aspects regarding Hofstede’s cultural dimensions model. Buletinul Universităţii
Petrol-Gaze din Ploieşti[The Bulletin of the Petrol-Gas University in Ploieşti], LVIII, 77-82. Retrieved
from http://www.upg-bulletin-se.ro/archive/2006-4/10.%20Albu.pdf
Benedict, R. (1972). Rumanian culture and behavior. Occasional Papers in Anthropology 1, 1-54.
Bresnahan, M.J., Shearman, S.M., Lee, S.Y., Ohashi, R., and Mosher, D. (2002). Personal and cultural
differences in responding to criticisms in three countries. Asian Journal of Social Psychology 5, 93-105.
Central Intelligence Agency (2011). The World Factbook: Romania. Retrieved from
https://www.cia.gov/ library/publications/the-world-factbook/geos/ro.html
Calafeteanu, I. (n. d.). Istoria românilor [Romanians’ history]. Retrieved from
http://www.casaromana.org/ istoria/r_secolulxx.html
Friedlmeier, M. (2006). Transmission of values within families in Romania (Doctoral dissertation, Konstanz
University, Germany). Retrieved from
http://www.ub.uni-konstanz.de/kops/volltexte/2008/5505/pdf/Diss_Friedlmeier.pdf
Hample, D. (2009). Commentary on: F. Paglieri's “Ruinous arguments: Escalation of disagreement and the
dangers of arguing.” In J. Ritola (ed.). Argument cultures: Proceedings of the 8th International
Conference of the Ontario Society for the Study of Argumentation. Windsor, Canada: University of
Windsor. [CD-ROM]
Hample, D. (2005). Arguing: Exchanging reasons face to face. Mahwah, NJ: Lawrence Erlbaum.
Hample, D., Paglieri, F., and Na, L. (in press). The costs and benefits of arguing: Predicting the decision
whether to engage or not. In F. H. van Eemeren, B. Garssen, D. Godden, & Mitchell, G. (eds). Proceed-
ings of the 7th International Conference of the International Society for the Study of Argumentation.
Amsterdam, The Netherlands: Sic-Sat.
Hanushek, E.A., & Jackson, J.E. (1977). Statistical methods for social scientists. San Diego, CA: Academic
Press.
Hofstede, G. (2001). Cultures’ consequences (2nd
ed). Beverly Hills, CA: Sage.
Hu, L., and Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling 6, 1-55.
Infante, D.A., and Rancer, A.S. (1982). A conceptualization and measure of argumentativeness. Journal of
Personality Assessment 46, 72-80.
Infante, D.A., and Wigley, C.J. (1986). Verbal aggressiveness: An interpersonal model and measure.
Communication Monographs 53, 61-69.
Johnson, A.J. (2002). Beliefs about arguing: A comparison of public issue and personal issue arguments.
Communication Reports 15, 99-112.
Paglieri, F. (2009). Ruinous arguments: Escalation of disagreement and the dangers of arguing. In: Ritola,
J. (ed.), Argument cultures: Proceedings of the 8th International Conference of the Ontario Society for
the Study of Argumentation. Windsor, Canada: University of Windsor. [CD-ROM]
Paglieri, F., and Castelfranchi, C. (2010). Why argue? Towards a cost-benefit analysis of argumentation.
Argument & Computation 1, 71-91.
Prunty, A.M., Klopf, D.W., & Ishii, S. (1990). Japanese and American tendencies to argue. Psychological
Reports 66, 802. Retrieved from http://www.ammonsscientific.com/AmSci/
Prunty, A.M., Klopf, D.W., & Ishii, S. (1991). Argumentativeness: Japanese and American tendencies to
approach and avoid conflict. Communication Research Reports 7, 75-79.
Rancer, A.S., and Avtgis, T.A. (2006). Argumentative and aggressive communication: Theory, research,
and application. Thousand Oaks, CA: Sage.
Săcară, L., and Iacob, L. (2002). Dinamica structurilor axiologice la adolescenţi [The dynamic of
axiological structures in adolescents]. Psihologie Socială 10, 52-77.

13
Table 1
Confirmatory factor analyses results
2
 df, N CFI SRMR RMSEA
Behavioral Intent 204.40 77, 201 .94 .07 .09
Argt. Approach 103.22 27, 201 .92 .07 .11
Argt. Avoid 63.09* 27, 201 .97 .07 .08
Verbal agress. Pro-Social 64.90* 35, 201 .93 .06 .07
Verbal agress. Anti-Social 78.77 35, 201 .94 .06 .08
Costs 51.33 5, 201 .92 .08 .22
Benefits 34.38 5, 201 .97 .03 .15
Resolvability 25.58* 5, 201 .93 .06 .14
Appropriateness 103.77 14, 201 .93 .10 .17
Civility 30.97 5, 201 .94 .07 .16
Other Reasonability*** 0 - - - -
Likelihood of Win 87.45 9, 201 .86 .08 .21
Realism 14.70* 5, 201 .97 .04 .10
All 2
 results are significant at p = .00 unless otherwise noted.
* p < .001.
** p < .05.
*** The model was just-identified. Therefore, no fit indices are available.

14
Table 2
Descriptive Statistics for Aggregate Variables*
α N Mean SD
Argt. Approach .84 201 3.53 0.68
Argt. Avoid .85 201 2.81 0.74
Verbal agress. Pro-Social .73 201 3.41 0.55
Verbal agress. Anti-Social .79 201 2.64 0.63
Behavioral Intent .87 201 3.71 0.63
Friends, public .80 20 3.66 0.61
Friends, private .91 39 3.82 0.67
Romantic, public .86 73 3.57 0.64
Romantic, private .85 69 3.80 0.60
Resolvability .75 201 3.25 0.71
Civility .82 201 3.28 0.83
Appropriateness .90 201 3.02 0.92
Benefits .90 201 3.16 0.85
Costs .84 201 2.60 0.87

15
α N Mean SD
Other Reasonability .64 201 3.23 0.73
Likelihood of Win .81 201 3.10 0.62
Realism .77 201 3.66 0.68
* All information is based on the items retained following the scale assessment analyses.

Zenker, F. (ed.). Argumentation: Cognition and Community. Proceedings of the 9th
International Conference of the
Ontario Society for the Study of Argumentation (OSSA), May 18-21, 2011. Windsor, ON (CD ROM), pp. 1-4.
Commentary on “A TEST OF THE
ARGUMENT ENGAGEMENT MODEL IN ROMANIA”
by Iona Cionea, Dale Hample, and Fabio Paglieri
LILIAN BERMEJO-LUQUE
Instituto de Filosofía, Department of Theoretical Philosophy
CCHS-CSIC
Albasanz 26-28, Madrid, 28037
Spain
lilian.bermejoluque@gmail.com
1. INTRODUCTION
The main goal of Cionea, Hample & Paglieri’s “A test of the Argument Engagement
Model in Romania” was to test Hample, Paglieri & Na’s (2010) model of argument
engagement. According to this model, a person’s intent to engage in an argument can be
predicted based on the following factors:
 Situational factors (characterized as the argument topic)
 Traits of the arguer (characterized as argumentativeness and
verbal aggressiveness tendencies)
 The expected costs of engaging in an argument (characterized in
terms of the cognitive effort involved in arguing)
 The expected benefits of arguing
 Likelihood of winning the argument
 Perceived appropriateness of arguing within the given situation
 The expected level of civility of the argument
 The perceived resolvability of the argument, and
 The expected reasonability of the other person.
The authors pointed out that the model had received some support when tested in the
United States and their aim was to test it in a different cultural and argumentative context.
This is why they design a new experiment whose sample consisted of Romanians. Yet, as
the authors explain, there are no studies on argumentative behaviors in Romania. Because
of this, their test of the model was rather brought about by dealing with the following
research questions:
RQ1: Is the behavioral intent to engage in an argument predicted by argumentativeness, verbal
aggressiveness, cost of arguing, benefits of arguing, resolvability of an argument, appropriate-
ness of arguing, civility, other’s reasonability, and likelihood of winning the argument? (…)
RQ2: Is there a difference in the argumentativeness, verbal aggressiveness, cost of arguing,
benefits of arguing, resolvability of an argument, appropriateness of arguing, civility, other’s
reasonability, and likelihood of winning the argument based on a) the topic of argument or b)
the argumentation partner? (…)
RQ3: Are there any differences between Romanians and U. S. Americans on a) argumenta-
tiveness and b) verbal aggressiveness? (Cionea et al. 2011: 3)

LILIAN BERMEJO LUQUE
2
To this end, the authors designed a 2 (topics of argument: private or public) x 2 (kinds of
relationship between arguers) experiment in which the topics of the argument were, on
the one hand, a preference for movies, and on the other, a preference for a particular
political candidate and in which the types of relationships where either friendship or
romantic partnership. This matrix determined the situational factors of the experiment.
Once they established the four possible scenarios that this matrix permitted, they
took a sampling of 201 Romanians recruited from Cionea’s acquaintances and social
networking sites (61), students from a university in the North-West of Romania (58) and
students from an off-campus residence hall (82). The age of the participants ranged from
18 to 64 years, they were undergraduate students, graduate students and working adults,
38 were men and 163 were women and most of them were Romanians. All the par-
ticipants were evaluated according to their argumentativeness and verbal aggressiveness
(measured with thirty-seven items, following Infante & Rancer (1982) and (1986) scales),
which determined participants’ tendency to approach an argument or to avoid an argu-
ment and their tendency for pro-social behaviors or for anti-social behaviors, that is, their
argumentative traits. Then they read one of the four possible scenarios and answered a
questionnaire measuring the following features:
 Behavioral intent, the dependent variable, measured with fourteen items, deter-
mined the participants’ intention to engage in argumentation in each scenario.
 Cost of arguing, measured with five items, determined the cognitive effort that
engaging in argumentation in each scenario amounted for each participant.
 Benefits of arguing, measured with five items, determined the perceived poten-
tial benefits of engaging in argumentation, both personally, and for the relation-
ships between the two people.
 Resolvability, measured with five items, determined the perceived chance of
resolving the argument if it occurred.
 Appropriateness, measured with seven items, determined the appropriateness of
having an argument in this scenario.
 Civility, measured with five items, determined the perceived degree of hostility,
open-mindedness and cooperation between the two people if an argument occurred.
 Other’s reasonability, measured with three items, measured the perceived degree
to which the other person would be stubborn, mature, tolerant, and willing to
change his/her mind if an argument occurred.
 Likelihood of winning was measured with six items assessing who had better
arguments and evidence to support his/her position and who the winner of the
argument would be.
As a result of the experiment, Cionea, Hample & Paglieri said, as an answer to RQ1, that
the intention to engage in argumentation was mostly predicted by three of the factors that
Hample, Paglieri & Na (2010) proposed, namely, the perceived appropriateness of
arguing, the other’s perceived reasonability and the likelihood of winning the argument.
On this result, the argument engagement model proposed by Hample, Paglieri, and Na
(2010) would have received only partial support.

COMMENTARY
3
In turn, the answer to RQ2a was that the topic of the argument made a difference
“only as far as argument avoidance is concerned (because friends avoided private arguments
less than they avoided public ones). The answer to RQ2b was that the type of argumentation
partner made a difference regarding 1) argument avoidance (because friends avoided private
arguments more than romantic partners did) and 2) the cost of arguing (because friends asso-
ciated less costs with arguing about public topics than romantic partners did).
Finally, regarding RQ3a the authors concluded that, according to the experiment,
Romanians were more argumentative than U.S. Americans; and regarding RQ3b, they
concluded that there was no significant difference in verbal aggressiveness between Ro-
manians and U.S. Americans.
2. COMMENTS
I take for granted that the experiment was well designed and carried out from a technical
point of view. The paper gives many details about these questions which seem to suggest that,
at least, it has been carefully done. However, statistics is not my field of expertise, so that,
unfortunately, I have to declare myself unable to evaluate this aspect of the paper. In turn, I
will concentrate on the conceptual design of the experiment—particularly, on the question of
the topics chosen- and on the conclusions that the authors draw from their results.
To begin with, the authors said that they “expected appropriateness to be an
important factor that affects one’s decision to engage in an argument in Romania due to
the cultural norms that guide appropriate interactions with others” (Cionea et al. 2011: 9).
I suppose that they were actually considering “the perceived appropriateness of arguing”,
that is, a correlation between the degree of appropriateness that the individual attributes
to engaging in argumentation in a particular scenario and her degree of willingness to
engage in argumentation. That is, the authors did not take any of the proposed scenarios to
be “actually” more appropriate for engaging in argumentation than the other. So, in order to
make more salient the differences between perceived degrees of appropriateness, I think
that it would have been convenient to propose more controversial and varied topics.
According to the experiment, another important predictor was the perceived
likelihood of winning the argument. In this regard, I also think that the election of the
topics was problematic. For example, it is plausible that a topic on which one of the
arguers could be an expert or an authority would have made a bigger difference. But the
topics proposed were, to a great extent, a matter of preferences, which give rise to
arguments that are quite difficult “to win”. On the other hand, the authors contend that “if
one perceives there is not a chance of winning, even if arguing may seem appropriate, it
won’t be pursued” (Cionea et al. 2011: 9). But I think that there are at least two ways of
understanding that “there is not a chance of winning”. One of them is that, because of the
topic of the argument (for example, a matter of preferences), there is no possibility of
“actually” winning: even if the other just gives up, we are not going to rationally
persuade her of our preferences. The other sense of “there being a chance of winning”
would be the sense in which the subject takes herself to be in a better position to defend
her standpoint (for example, because she is an expert on the matter). In both cases, the
topic of the argument is a key question.
Regarding the question of the expected reasonability of the other person, the
authors said that according to their results, there seems to be “no point in trying to argue

LILIAN BERMEJO LUQUE
4
with someone who will not change one’s mind despite good arguments and evidence.”
(Cionea et al. 2011: 9). But, again, I think that the topic of the argument, joint with an over-
instrumentalist conception of motives for arguing is a source of concern here: in principle,
it is plausible to think that if the topic in question is perceived as something crucial, or if
defending one’s standpoint is seen as a matter of principles, then individuals will engage in
argumentation regardless of their chances of persuading the other. Besides, in some
contexts we offer reasons just because of their pedagogical value, so that the expected
reasonability of the other does not play an important role. But the design of the experiment
did not seem to allow this possibility: neither the proposed topics were ones in which
individuals in the sample could take as a duty to argue nor the possible partners were ones
that would demand a pedagogical treatment of the topic in question by the individuals in
the sample. For these reasons, I think that grounding the selection of the topics on the
distinction between public and private topics was an infelicitous strategy: probably, more
significant differences on the bearing of the topic upon the willingness to engage in
argumentation would have been found if the topics had been related to distinctions such as “is
it a topic that could be taken to be a crucial question for one or both arguers, or a matter of
principles?”, “is it a topic in which one of the arguers could be an expert or an authority?”.
The authors said that the results of their study did not offer any support for the
idea that perceived costs and benefits of arguing are good predictors of the decision of
arguing (Cionea et al. 2011: 10). In principle, this is a counterintuitive result: what is a
rational decision if not a decision took after valuing costs and benefits, broadly construed?
Well, the key is precisely that, in the study, the items “perceived costs and benefits” are not
broadly construed. Particularly, as we have seen, the perceived costs of arguing are a matter
of the perceived cognitive effort that engaging in argumentation will involve.
Finally, the authors say that their “finding contradicts the conclusion that people
from collectivistic cultures are less argumentative than people from individualistic
cultures.” However, as they themselves recognize, their study is strongly limited by the
fact that their sample was mostly female: women’s disposition to verbally interact with
others is bigger than men’s, or at least, that’s the stereotype. It is probably necessary to
check that this is not the main source of the difference between the results of the test with
Romanians and U.S. Americans.
REFERENCES
Cionea, I., Hample, D., and Paglieri, F. (2011). A Test of the Argument Engagement Model in Romania
(this volume).
Hample, D., Paglieri, F., & Na, L. (2010). The costs and benefits of arguing: Predicting the decision
whether to engage or not. Paper presented at the meeting of the International Society for the Study of
Argumentation, Amsterdam, The Netherlands.
Infante, D.A., and Rancer, A.S. (1982). A conceptualization and measure of argumentativeness. Journal of
Personality Assessment 46, 72-80. doi: 10.1207/s15327752jpa4601_13
Infante, D.A., and Wigley, C. J. (1986). Verbal aggressiveness: An interpersonal model and measure.
Communication Monographs 53, 61-69. doi: 10.1080/03637758609376126

A Test Of The Argument Engagement Model In Romania

Recommended

Recommended

More Related Content

Similar to A Test Of The Argument Engagement Model In Romania

Similar to A Test Of The Argument Engagement Model In Romania (20)

More from Emma Burke

More from Emma Burke (20)

Recently uploaded

Recently uploaded (20)

A Test Of The Argument Engagement Model In Romania