Minkov indb

Understanding “Culture”
Therefore, it might be useful that those 
who present cultural analyses explain 
how they conceptualize culture, specifying 
its contents and boundaries. This could 
help avoid a situation described by Child 
(1981), who pointed out that there is a 
danger of inferring culture as a national 
phenomenon from virtually any contrasts 
that emerge from a comparison of orga-
nizations in different countries: “Even if 
such contrasts are unambiguously national 
in scope, they could possibly be due to 
other non-cultural phenomena such as 
national wealth, level of industrialization, 
or even climate” (p. 328). 
A comment by Fischer (2009) illustrates 
another practical reason to define culture. 
In his view, if researchers do not focus 
on the shared aspect of culture (see 2.1.), 
there is no need to investigate agreement 
among the members of a national culture 
who provide information to a researcher. 
But if one adopts a definition of culture in 
which sharedness is emphasized, such an 
investigation becomes necessary. 
Leung and van de Vijver (2008) dis-
cuss two approaches to culture: holistic 
and causal. The first approach is taken 
by those who view culture as consisting 
of inseparable phenomena that cannot 
cause each other. Those who prefer the 
second approach may say that one cultural 
characteristic shapes another. If this is so, 
cultural researchers may need to explain 
how they conceive of culture: holistically 
or causally. 
There are also other reasons for defin-
ing culture. Some methodologists working 
in the domain of cross-cultural psychology 
have treated culture as a variable resem-
bling some kind of noise that needs to be 
reduced or eliminated. Poortinga and van 
de Vijver (1987) suggested a procedure for 
explaining measured differences between 
societies by introducing various relevant 
variables, each of which explains part of the 
observed variance, until the effect of cul-
ture disappears: “The consequence of our 
argument is that a cross-cultural psycholo-
gist is not interested in the variable culture 
per se, but only in specific context vari-
ables that can explain observed differences 
on some dependent variable” (p. 272), and 
“In the ideal study the set of context vari-
ables will be chosen in such a way that the 
remaining effect for culture will be zero” 
(p. 272). This begs the question of what 
variables can explain differences between 
groups of people but are not part of their 
cultures. 
3
Some of the clearly external variables 
with respect to culture—also known as 
“exogenous” or “extraneous”—are cli-
mate, geographic location, and patho-
gen prevalence. But what about national 
wealth, main type of economy, or degree 
of democracy? Are these cultural variables 
or not? According to van de Vijver and 
Leung (1997a), gross national product, 
educational systems, and even health care 
institutions are culture-related variables 
(p. 4). Is this position acceptable? 
Javidan and Houser (2004) describe 
two possible views: that a society’s wealth 
should not be confused with its culture 
and that wealth is an integral part of 
its culture. The position that we adopt 
may determine our research methodol-
ogy. If wealth is an extraneous variable, a 
researcher may decide to partial it out of 
cultural measures using statistical tools. If 
wealth is viewed as an integral part of cul-
ture, there is no need to control for it when 
cultural variables and the relationships 
between them are measured. Thus, the 
solution is a matter of subjective choice. 
◆