Page 110 - Handbooks of Applied Linguistics Communication Competence Language and Communication Problems Practical Solutions

88 David Matsumoto, Seung Hee Yoo and Jeffrey A. LeRoux

Another source of information concerning cultural differences in ER
comes from McCrae’s multinational study of the five factor model of person-
ality (Allik and McCrae 2004; McCrae 2002; McCrae et al. 1998). In these
studies McCrae and his colleagues have used their Revised NEO-Personality
Inventory (NEO-PI-R; Costa and McCrae 1992), a 240 item questionnaire that
measures the five personality traits considered to be universal: Extraversion,
Openness, Agreeableness, Conscientiousness, and Neuroticism. To date
McCrae has reported data on this measure from 36 samples in 32 countries in-
volving both college students and adults (McCrae 2002). Although data are
collected from individuals, means on the various facet scores were computed
for each sample. The Five Factor Model replicates on the national level as
well as the individual (McCrae 2001, 2002). Based on these results McCrae
has computed country-level means for each of the five factors (and their fa-
cets) for each of the countries studied. Country scores on Neuroticism prob-
ably reflect mean levels of ER. Neuroticism is typically defined as emotional
lability, and thus high scores on Neuroticism probably reflect low scores on
emotion regulation, and vice versa. This suggests that people from countries
high on Neuroticism would experience more difficulty in intercultural adjust-
ment, and vice versa. In McCrae’s study, the three countries that scored hig-
hest on Neuroticism were Portugal, Italy and Spain; the three lowest were
Sweden, Denmark and Norway.
The notion that Hofstede’s UA and McCrae’s Neuroticism are related to
each other received empirical support by Hofstede and McCrae (2004), who
computed country-level correlations between their respective culture and per-
sonality scores. UA was correlated with Neuroticism 0.58 (and negatively with
Agreeableness –0.55), suggesting that these dimensions share a common de-
nominator. We suggest that one common denominator is ER.
One of the limitations of using the Hofstede and McCrae data to estimate
cultural differences in ER is that neither of them intended to measure ER di-
rectly. The ICAPS described earlier in this chapter, however, does, and our cur-
rent normative database includes data from approximately 11,000 individuals
around the world. We computed an exploratory factor analysis of these data,
after doubly standardizing both within individuals and countries in order to
eliminate positioning effects and to produce a pancultural solution (Leung and
Bond 1989). As previously, the first factor to emerge in these analyses was ER.
We then created scale scores for the raw data using the highest loading items on
this factor (11 items), and computed means on this scale for each country rep-
resented in the data set. (Respondents rate each item on a 7-point scale; means
therefore range from 1–7.) Like the Hofstede and McCrae data sets, these data
(Table 2) also demonstrate considerable variability across cultures in ER. The
three countries with the highest ICAPS ER scores were Sweden, Norway, and
Finland; the three lowest were Japan, Malaysia, and China.