Page 524 - Cultures and Organizations

Notes 489

masculinity stressed patience, courtesy, and kindness; those in feminine countries
stressed patriotism and righteousness. These associations are surprising; one would
have expected the poles to be reversed. Geert has always retained a suspicion that in
the data processing a minus sign was lost. If the analysis was right, we have to inter-
pret this result as an example of the difference between the desired and the desirable,
described in Chapter 2.
8. Across forty-eight common countries, MAS correlated with r 0.30* with asser-
tiveness “as is” and with a multiple R 0.43*** with assertiveness “as is” plus asser-
tiveness “should be,” both taken positively. This is remarkable, because assertiveness

“as is” and “should be” between themselves were significantly negatively correlated:
0.33*.
9. In a comparison with other GLOBE dimensions, humane orientation “as is” corre-
lated strongly with a combination of small assertiveness and small power distance “as
is.” It was not an independent dimension. Humane orientation “should be” was not an
independent dimension either: it correlated strongly with low uncertainty avoidance,
small institutional collectivism, and small power distance, all of them “should be.”
10. Culture’s Consequences, 2001, p. 266. The multiple correlation with IDV and MAS
was R 0.86***.
11. Bem, 1975, p. 636.
12. The percentage of women in the IBM survey population varied from 4.0 in
Pakistan to 16.2 in Finland. In Culture’s Consequences, 2001, p. 286, the MAS scores
have been recalculated, keeping the percentage of women constant for all countries.
The effect on the scores was minimal, as the percentages of women in themselves were
correlated with femininity.
13. Gray, 1993. We owe this observation to Marieke de Mooij.
14. Lynn, 1991; van de Vliert, 1998, Table 7.2; Culture’s Consequences, 2001, p. 308.
15. Based on Culture’s Consequences, 2001, pp. 289–91.
16. Stevens, 1973; Gonzalez, 1982; Culture’s Consequences, 2001, p. 309.
17. Based on Etcoff, Orbach, Scott, & Agostino, 2006. The correlation of MAS with
the choice “father” was r 0.79**; with “mother” it was r 0.75** (n 10). In
a stepwise regression, MAS explained 57 percent of the variance for “mother”; MAS
plus PDI explained 76 percent. The correlation of MAS with the choice “media” was
r 0.88***; with “celebrities” it was r 0.62* (n 10). (Calculations courtesy of

Marieke de Mooij.) See also Chapters 4 and 7.
18. Flash Eurobarometer 247, 2008. Across eighteen countries with a 2007 GNI
per capita over 19,000 euros, high MAS explained 24 percent of the differences in per-
centages of families with one parent working and one parent caring for the household
full-time. (Calculations courtesy of Marieke de Mooij.)
19. van Rossum, 1998; Culture’s Consequences, 2001, p. 300.
20. Culture’s Consequences, 2001, p. 302.
21. Hofstede, 1996b; Culture’s Consequences, 2001, p. 302.
22. Mead, 1962 [1950], pp. 271ff.
23. Culture’s Consequences, 2001, p. 309.