A Reconsideration of Hofstede’s Fifth Dimension: New Flexibility Versus Monumentalism Data From 54 Countries

Abstract

Hofstede’s “long-term orientation” (LTO) may be one of the most important dimensions of national culture, as it highlights differences on a continuum from East Asia to Africa and Latin America, strongly associated with differences in educational achievement. However, LTO’s structure lacks theoretical coherence. We show that a statistically similar, and theoretically more focused and coherent, dimension of national culture, called “flexibility versus monumentalism,” or vice versa, can be extracted from national differences in self-enhancement and self-stability or self-consistency, as well as a willingness to help people. Using data from nearly 53,000 respondents recruited probabilistically from 54 countries, we provide a new national flexibility-versus-monumentalism index that measures key cultural differences on the world’s East–West geographic axis and predicts educational achievement better than LTO or any other known dimension of national culture.

Keywords

national culture Hofstede’s model long-term orientation monumentalism self-enhancement educational achievement

Introduction

In a recent article, van Witteloostuijn (2016) criticized what he called the anti-Popperian nature of modern social science. In particular, he deplored the scarcity of replication studies. In this regard, despite the enormous influence of Hofstede’s (2001) model of national culture, very few attempts have been made to replicate his dimensions. Hofstede’s followers take the reliability of his model for granted, whereas detractors sometimes criticize it in an empirical vacuum. This study is devoted to a replication and reconceptualization of one of the most important dimensions in Hofstede’s model: long-term orientation (LTO) versus short-term orientation (STO), known simply as LTO. As LTO highlights salient cultural differences between East Asia at one extreme and Africa and Latin America at the other, and because of that dimension’s reportedly strong association with average national educational achievement (Hofstede, 2001; Minkov & Hofstede, 2012), the importance of LTO cannot be overstated. Yet, the dimension is still shrouded in some obscurity as it is conceptually problematic. Also, the highest number of countries across which LTO has been studied is only 38 (Minkov & Hofstede, 2012). There is a need for a larger study that elucidates some of the main cultural differences mapping countries approximately on the Earth’s East–West geographic axis and their implications.

LTO From Its Original Perspective

The Chinese Culture Connection (1987) reported a dimension of national culture called “Confucian work dynamism” (CWD). It was extracted from a nation-level analysis of responses to questions about personal values suggested by Chinese scholars. Given its empirical distinctiveness from his original four dimensions of national culture, Hofstede (2001) later adopted CWD as a fifth dimension in his popular model of national culture under the name LTO. Minkov and Hofstede (2012) replicated LTO with World Values Survey items. That study demonstrated that a Chinese approach and a Western approach to the study of culture can yield very similar results. More importantly, Hofstede (2001), Minkov (2008), and Minkov and Hofstede (2012) reported that CWD/LTO was a strong predictor of average national cognitive achievement assessed as performance on standardized mathematics tests designed by the international project Trends in International Mathematics and Science Study (TIMSS).

The original CWD (Chinese Culture Connection, 1987) and Minkov and Hofstede’s (2012) LTO demonstrate a significant conceptual overlap, suggesting that the main facets at the positive pole of the CWD/LTO dimension (high CWD or high LTO) are thrift and determination or persistence. Those at the negative pole (low CWD or STO) are self-stability or self-consistency (having an invariant self that does not change across situations), positive self-enhancing feelings (such as pride), and a willingness to do favors or render services to people. The logical relationship between these facets has not been explained satisfactorily, and this incoherence has generated criticisms (Ashkanasy, Gupta, Mayfield, & Trevor-Roberts, 2004; Fang, 2003). Minkov and Hofstede have acknowledged that there is a lack of a convincing explanation of “what holds together such seemingly diverse values as thrift and persistence and opposes them to personal stability, tradition, and a willingness to help others” (p. 11). A new look at LTO is needed to elucidate its structure.

An Alternative Perspective

To improve the comprehensibility of LTO, Minkov (2008, 2011, 2013) proposed a new approach: The best starting point for making sense of the LTO dimension might be Steven Heine’s self-enhancement and self-stability theory. Heine (2003) defines self-enhancement as “the tendency to overly dwell on, elaborate, and exaggerate positive aspects of the self, relative to one’s weaknesses” (p. 101). Later, Heine and Hamamura (2007) reviewed 91 cross-cultural comparisons in terms of self-enhancement and tabulated all the different methods that were used to operationalize that concept. They present self-enhancement as a complex syndrome of positive self-concepts, such as high self-regard, self-liking, satisfaction with self, self-confidence, and optimism. In sum, the core of this syndrome is a high opinion of one’s self, including one’s abilities, which is associated with self-confidence.

Heine (2003) theorized that self-enhancement should be related to self-stability or self-consistency: the tendency to ascribe immutable traits to the human self and value the existence of such traits. If one believes that abilities are the result of stable personal factors, it becomes beneficial to view the self in the most positive light. Discoveries of weaknesses in the stable self would have a psychologically unsettling effect, because they would be seen as irremediable. Conversely, if one views the self as fluid and malleable, there is no imperative need for self-enhancement. Instead, one can recognize one’s deficiencies and make efforts to correct them.

It is important to note that Heine (2003) sees self-enhancement and self-stability as negative predictors of self-improvement, such as a willingness to make strong efforts in education. The reason for this is that self-enhancers should be satisfied with their selves and discount the need for self-improvement. As they believe in immutable selves, they may see activities aimed at self-improvement as futile.

Some indirect support for these associations is provided by Green, Pinter, and Sedikides (2005). They found that people are more likely to neglect negative information about themselves when that information concerns traits perceived as unmodifiable. In other words, it is important to maintain a positive self-image concerning stable aspects of the self. Blackwell, Trzesniewski, and Dweck (2007) found that students who believed that intelligence is malleable subsequently tended to achieve an upward trajectory in their school grades, whereas those who believed that it is a fixed trait achieved a flat trajectory. Mueller and Dweck (1998) found that students who are praised for their intelligence (and thus receive self-enhancing stimuli) have a tendency to believe that intelligence is a fixed trait and lose their achievement motivation.

Heine’s theory was developed to explain individual-level phenomena. Yet, it seems attractive as a starting point for an explanation of the core structure of LTO as a dimension of national culture, because Heine’s theory addresses the concept of self-stability, which seems central to LTO. Another key component of LTO—pride and a high concern for dignity (Minkov & Hofstede, 2012)—is an aspect of self-enhancement. Yet, it is not clear why these concepts should be related at the societal level. Nor is it obvious what they have in common with “service to others” (in Minkov and Hofstede’s operationalization), or “reciprocation of favors and gifts” (in the operationalization of the Chinese Culture Connection) and “thrift,” which are also key components of the LTO dimension.

We propose that if a particular society encourages pride and a sense of dignity, it is logical that it should emphasize self-stability and consistency as well. Proud and dignified individuals are supposed to maintain their pride and dignity across various situations rather than be variable and fickle. The old French saying “honneur oblige” is an example of the perception that an honorable person has the obligation always to maintain an honorable behavior. We must note that it is unlikely that any human culture considers dignity unimportant while valuing fickleness. Cross-cultural differences in the prevalence of these traits are not about presence versus absence but about relative frequency or relative importance.

A sense of dignity and high self-esteem can be derived from socially useful activities, such as “service to others.” Societies that emphasize seemingly selfless help must in fact provide some incentive for such contributions, for instance, by making them a source of public respect and high esteem. For the same reason, in such societies, sharing one’s possessions with others (which is a form of help and service to others) and gift-giving may be more laudable than saving and keeping what one owns for oneself. Lee (1979) provides an example of such a society from his fieldwork with the !Kung San of Southern Africa. In the 1960s, when Lee did his work, the !Kung San equated saving with stinginess and hoarding and considered this a major moral flaw: “The corrective for this, in the Kung view, is to make the hoarder give ‘till it hurts,’ that is to make him give generously and without stint until everyone can see that he is truly cleaned out” (p. 458).

Societies that encourage exchange of services (STO societies) seem to de-emphasize self-reliance. Minkov and Hofstede (2012) reported a high positive correlation between LTO and self-reliance as measured by Green, Deschamps, and Paez (2005) with items such as “Only those who depend on themselves get ahead in life.” The highest score on self-reliance was obtained in China, the lowest in various Latin American countries. Despite the authors’ intention to measure variants of individualism-collectivism, their nation-level measure of self-reliance was not associated with either Hofstede’s (2001) or Project GLOBE’s (Gelfand, Bhawuk, Nishii, & Bechtold, 2004) measures of that construct.

LTO is also highly and positively associated with the “independence” item (coded v12) in the 2005-2009 wave (the largest that provides overlapping data) of the World Values Survey section asking the respondents to select important child qualities: r = .64 (p < .001, n = 21) with LTO in Minkov and Hofstede (2012). The “obedience” item (coded v21) is highly and negatively associated with LTO in Minkov and Hofstede (2012): r = –.69 (p = .001, n = 21). Just like self-reliance, the “independence” item is not associated with individualism as measured by Hofstede or Project Globe, whereas “obedience” is weakly associated with each of them. “Independence” versus “obedience” create a geographic pattern that is similar to that of LTO versus STO. East Asia, the Scandinavian countries, and the German-speaking countries have the highest scores on independence and the lowest on obedience, whereas Africa, Latin America, and the Middle East exhibit the opposite pattern. This configuration is quite logical: Obedience goes hand in hand with dependence and both are more typical in traditional-conservative societies with large percentages of poorly educated people. Vice versa, there is more independence in societies with large percentages of highly educated people who can get good jobs and can take care of themselves. This is particularly true with respect to women. In highly educated nations, women are much more independent than in poorly educated nations, where they are often completely dependent on male relatives and husbands.

Figure 1 illustrates the association between importance of “independence” and “obedience” across the World Values Survey countries in 2005-2009. The 2011-2014 wave yields a very similar pattern.

Figure 1.

Visualization of the relationship between importance of “independence” and “obedience” as desirable traits for children, items v12 and v21, World Values Survey 2005-2009.

Societies that promote interdependence through an exchange of services and mutual help are likely to value individuals who are predictable and reliable, possessing strong values that guide their admirable behaviors across situations and guarantee behavioral consistency. Individuals with variable selves are likely to be seen as a potential threat to the system of interdependence. If a society has a lot of such individuals, interdependence is compromised as one does not know when and to what extent one can depend on others.

We still have one seemingly important LTO aspect unexplained: perseverance (persistence) and determination. We note that neither of the two latest sets of national scores (2005-2009 and 2011-2014) yielded by the corresponding World Values Survey item (coded v18) is associated with the original CWD index in the Chinese Culture Connection (1987). Minkov and Hofstede (2012) chose to include that item in their operationalization of LTO, imitating the Chinese Culture Connection, yet the results of the 2011-2014 wave of the World Values Survey cast doubt on the appropriateness of that decision. While Japan systematically scores high on that item in the recent waves, China and Taiwan systematically have very low scores, whereas South Korea shows some inconsistency. Thus, that item does not seem to highlight shared East Asian cultural characteristics that distinguish East Asia from other cultural clusters. In the absence of any other evidence about national differences in determination–perseverance, it is best for the time being to leave that topic out of our discussion.

Objectives of the Present Study

Our goal is to measure aspects of self-enhancement, self-stability (self-consistency), and willingness to help people, and test the following hypotheses:

Hypothesis 1: Our measures of self-enhancement, self-stability, and willingness to help people will yield a strong single factor at the national level (although they may not necessarily be associated at the individual level), meaning that societies that emphasize any one of these three characteristics emphasize the other two as well. This would support the theory that we proposed previously: Societies that promote interdependence through mutual help create self-enhancing tendencies and value self-stability.

Hypothesis 2: The national index that the single factor will yield will be highly correlated with reported measures of CWD and LTO, thus demonstrating that our new cultural measure has not only conceptual communalities with CWD and LTO but also statistical ones. Thus, our dimension of national culture will highlight cultural differences on a geographic continuum, ranging from East Asia to Africa, and Latin America, and will be a refined variant of CWD-LTO.

Hypothesis 3: Consistent with Heine’s theory and the known predictive properties of CWD and LTO, our dimension of national culture will be a strong predictor of national educational achievement in a crucially important subject: mathematics.

Materials and Method

Samples

We used a database from a large cross-cultural study of personality and culture, carried out in 2015-2016. The goal of the study was to test the replicability of Hofstede’s dimensions of national culture, as well as to measure other constructs, such as the Big Five traits of personality, across nationally representative samples, rather than matched samples of students or other individuals, or convenience samples of Internet users, which is the approach of nearly all cross-cultural studies of culture and personality across the world, with the exception of the World Values Survey.

The study yielded data from 44,374 respondents, chosen probabilistically from each country’s main regions, main ethnic and language groups, typical age groups, and diverse occupations in 55 countries. Respondents were informed that the survey was completely voluntary and anonymous. If they felt uncomfortable with any question, they could abandon the survey at any time. A small financial compensation was promised and subsequently provided to all those who completed the whole survey. The survey was financed by MediaCom, an international media agency. Notwithstanding this financial assistance, MediaCom did not participate in the study design or data analysis in any way, entrusting the first author of this study with all academic responsibilities related to the survey.

In addition, a project sponsored by the Government of the Republic of Kazakhstan used parts of the same questionnaire and provided data from another 8,478 respondents from Kazakhstan. Volunteers in Peru, Puerto Rico, Dominican Republic, and Myanmar collected data for those countries. Thus, the total number of respondents available for the analyses in this study is 52,852.

The sample structures approximate those of the national census in economically developed countries. In developing countries, there was a focus on respondents with higher education, as less educated respondents were harder to reach and in some cases might have difficulty answering questionnaires, being unused to them.

In all but three countries, the data were collected online and stored in a computer in Finland. In Kazakhstan, the data were collected face to face. In Puerto Rico and the Dominican Republic, they were collected by phone. Myanmar is the only country where no effort was made to obtain probabilistic samples. The respondents were recruited mostly through Facebook and are inhabitants of the largest Burmese cities. Detailed sample characteristics for all countries are available from Itim International (www.itim.org), a cross-cultural management consultancy, or from the corresponding author of this study. Concise information about the samples is provided in the appendix at the end of this article.

To correct for the education-level inconsistency across the national samples, we split each sample into two groups: respondents without higher education and respondents with higher education. An exception was made for Egypt and Myanmar, as we had too few respondents without higher education from those countries. Our main analysis is based on national samples without higher education. We used the national samples of university-educated respondents for a backup study and found that dimensions of national culture from samples of university-educated respondents correlate with their counterparts from respondents without higher education at .95 or higher. Thus, differences in educational level matter very little in large-scale comparisons of cultural differences.

As the sample composition in South Africa does not reflect the actual ethnic composition of the diverse population of that country, and because South Africans of native African descent differed enormously on their responses from the responses of those of European descent, we used only the sample of individuals of native African descent inasmuch as they account for more than 80% of South Africa’s population.

Reviewers of this article noted that whereas most samples are large (20 national samples exceed 1,000 respondents), a few are small, creating a discrepancy. We decided that 250 respondents would be a good cutoff point as it is reasonably high and we would lose only two countries if we adopted that criterion: Serbia and the Dominican Republic. We also dropped Puerto Rico from our initial analysis as we discovered technical issues in two of the items, making Puerto Rico an improbable outlier on those two items. However, the remaining five items that we used in this study did not pose problems and could be used to calculate a score for Puerto Rico after the initial analysis.

Questionnaire

The whole questionnaire contained 104 items grouped in several sections. The original questionnaire was written in English by the first author of this article. It was discussed in circles of representatives of diverse cultures and tested in pilot studies in the United States and Singapore. A Spanish version was fielded in Puerto Rico and the Dominican Republic, and a Russian version was used in Kazakhstan, prior to the main study. These tests suggested various interventions in the wordings of some items. The final questionnaire was translated into native local languages and back-translated into English. An exception was made for countries where English is not a predominant native language but is an official language of instruction at school and is sufficiently well known and spoken by educated citizens: India, Kenya, Nigeria, Singapore, and South Africa.

The items used in this study are from a section consisting of 52 items, targeting primarily the Big Five traits of personality. A few items were written specifically to capture LTO concepts, including self-enhancement and self-stability. The items elicit self-construals that are similar to self-descriptions in personality studies. The respondents are not asked what they value or consider important, as in the Chinese Culture Connection (1987) or in Minkov and Hofstede (2012). Instead, they are instructed to describe what kind of people they are in terms of traits that they possess or in terms of their typical behaviors. Although many values and traits may overlap, there can be a difference between a value and a corresponding trait. It is impossible to really value honesty as a guiding principle in one’s life while being a crook, but it is quite possible to value creativity or emotional stability, while not possessing these traits.

To avoid issues with Likert-type scales in cross-cultural research (Heine, Lehman, Peng, & Greenholtz, 2002), and especially to avoid extreme responding and acquiescence, the items do not use Likert-type scales at all. Negatively worded items are also avoided as these have been found to produce different effects in different cultures (Kim, Peng, & Chiu, 2008). Instead, all items ask the respondents to choose between two opposites, such as “usually shy” or “usually bold,” with a third, evasive or neutral option: “I am somewhere here, in between these two.”

This response format does not guarantee a lack of national differences in response style as there may be a tendency in some nations to prefer a particular position on the scale, despite its shortness and the clear labeling of the response options. To address potential concerns that our measures may be contaminated with systematic response style, we did two analyses: across raw items and across ipsatized items. Ipsatization was achieved by averaging national scores on all 52 items in the personality section of our questionnaire and subtracting that all-item average from each raw score.

We started from an analysis of selected items, written to target the concept of self-enhancement or self-confidence and self-stability or self-consistency, as well as helping people. We conceptualized self-enhancement as a sense of personal superiority (which incorporates a sense of high self-esteem), a willingness to demonstrate such superiority (for instance, through competition), and self-confidence. We conceptualized self-stability as genuineness and behavioral consistency, based on strong personal values, as opposed to personal duality or changeability.

To measure self-enhancement and self-confidence, we chose the following items:

1. I have some good qualities or skills that most other people do not have. 2. I am somewhere here, in between these two. 3. I am an ordinary person, like most other people.

1. I like to compete with people. 2. I am somewhere here, in between these two. 3. I hate to compete with people.

1. Most of the good things that happen to me come from my own actions. 2. I am somewhere here, in between these two. 3. When something good happens to me, I feel it is just good luck.

For self-stability or self-consistency, we chose the following items:

1. I would feel bad if I had to pretend and act like a different person. 2. I am somewhere here, in between these two. 3. I can pretend that I am somebody else without feeling bad about that.

1. I am usually the same person at home and outside (at work, at school, in public places). 2. I am somewhere here, in between these two. 3. I am often quite different at home and outside.

1. I have strong values and beliefs. They guide my behavior in most situations. 2. I am somewhere here, in between these two. 3. My behavior depends on the situation, not so much on my values and beliefs.

To measure willingness to help people, we used the following item:

1. I like to help people, even if I have to do something difficult. 2. I am somewhere here, in between these two. 3. I rarely agree to do something difficult to help people.

All items were scored as if they involved a scale: 1 for the first response option, 2 for the second, 3 for the third. Aggregation to the national level involved calculation of a national average, ranging between 1 and 3. We also tried working with percentages of respondents who have selected either of the two categorical answer options, but that method did not produce substantially different results.

Results

We factor-analyzed mean national scores on the seven items, using the generalized least square method. Alternative factor analysis methods and principal components analysis (taking into consideration the nonshared variance as well) produced nearly identical results. The factor analysis yielded a strong single factor with an eigenvalue of 4.49, explaining 64.02% of variance. The item loadings are provided in Table 1.

Table 1.

Factor Loadings on the Flexibility Versus Monumentalism Factor at the National Level.

Item		Factor loading
Short version of the monumentalist response option for each item (scored 1)	Short version of the flexibility response option for each item (scored 3)	Factor loading
I would feel bad if I had to pretend	I can pretend without feeling bad	.87
I have strong values guiding my behavior	My behavior depends on the situation	.83
I like to help	I rarely agree to help	.81
I have unique good qualities	I am an ordinary person	.80
I like to compete	I hate to compete	.74
Good things come from my actions	Good things are luck	.71
I am always the same person	I am different at home and outside	.64

Note. As the flexibility response options are scored higher than the monumentalism response options, and as all factor loadings are high and positive, high factor scores correspond to high flexibility and low monumentalism, corresponding to high LTO and low STO in Hofstede’s paradigm. LTO = long-term orientation; STO = short-term orientation.

We repeated this analysis with ipsatized items. The resulting single factor had an eigenvalue of 3.23 and explained 46.02% of variance. It correlated with the factor from raw items at .94. Obviously, ipsatization results in practically the same factor and same country positions on it. This demonstrates that the result of the analysis of raw items is not due to a failure to account for national differences in response style. While national differences in response style certainly exist, removing those differences from our dataset through a classic statistical procedure (ipsatization) does not have an appreciable effect on the final result.

The results in Table 1 support our first hypothesis. All items load on a strong single factor and obviously measure the same nation-level construct: a combination of self-enhancement and self-stability, associated with a willingness to help people. This dimension is conceptually less close to LTO than it is to Minkov’s (2011, 2013) “flexibility” versus “monumentalism,” and we decided to adopt those names for it. Monumentalism is a metaphor for a cultural tendency to encourage people to be like a monolithic monument: proud, stable, and consistent (made of the same substance outside and inside). Flexibility is the opposite cultural tendency, favoring a modest self-regard, duality, and adaptability.

The country scores that the factor analysis yielded were used as a dependent variable in linear regression to estimate a score for Puerto Rico. This estimated score should be accepted with caution.¹

Country scores on the single factor are provided in Table 2. These are flexibility-versus-monumentalism scores. In other words, high-scoring countries are characterized by flexibility (corresponding to high LTO in Hofstede’s paradigm), whereas low-scoring ones are characterized by monumentalism (low LTO in Hofstede’s paradigm). This is an outcome of the response format of the items: More points are given for choosing responses indicating flexibility than for responses indicating monumentalism.

Table 2.

Scores on Flexibility Versus Monumentalism for 54 Countries (Factor Scores × 100).

Country	Abbreviation	Score
1. Japan	JAPN	234
2. Hong Kong	HONG	199
3. Taiwan	TAIW	175
4. South Korea	KORE	174
5. China	CHIN	134
6. Singapore	SING	89
7. Thailand	THAI	87
8. Netherlands	NETH	87
9. Finland	FINL	71
10. Belgium	BELG	63
11. Kazakhstan	KAZK	62
12. France	FRAN	57
13. United Kingdom	UK	56
14. Denmark	DENM	51
15. Russia	RUSS	48
16. Germany	GERM	46
17. Australia	AUST	41
18. New Zealand	NEWZ	37
19. Norway	NORW	36
20. Canada	CANA	31
21. Sweden	SWED	21
22. Ukraine	UKRN	19
23. Ireland	IREL	17
24. Greece	GREE	17
25. India	INDI	17
26. Czech Republic	CZEC	17
27. United States	US	11
28. Poland	POLN	9
29. Spain	SPAI	2
30. Israel	ISRA	2
31. Vietnam	VIET	−1
32. Malaysia	MALS	−1
33. Austria	OSTR	−2
34. Philippines	PHIL	−4
35. Hungary	HUNG	−9
36. Switzerland	SWIT	−21
37. Portugal	PORT	−24
38. Italy	ITAL	−36
39. Turkey	TURK	−52
40. Myanmar	MYAN	−55
41. Romania	ROMN	−64
42. Indonesia	INDN	−78
43. Argentina	ARGN	−80
44. Brazil	BRAZ	−83
45. Puerto Rico	PUER	−90
46. Mexico	MEXC	−104
47. Egypt	EGPT	−111
48. South Africa	SAFR	−126
49. Chile	CHIL	−153
50. Kenya	KENY	−158
51. Nigeria	NIGR	−173
52. Colombia	COLO	−182
53. Peru	PERU	−187
54. Venezuela	VENZ	−207

Note. A high score reflects high flexibility and low monumentalism, corresponding to high LTO and low STO in Hofstede’s paradigm. A low score reflects high monumentalism and low flexibility, corresponding to high STO and low LTO in Hofstede’s paradigm. LTO = long-term orientation; STO = short-term orientation.

Table 3 shows the nomological network of flexibility versus monumentalism. It is highly and positively associated with CWD and LTO, and highly and negatively with measures of self-consistency and self-esteem. It is also strongly and positively associated with national measures of cognitive achievement: performance on standardized mathematics tests in the eighth grade or at age 15, as measured by the nationally representative projects TIMSS and the Program for International Student Assessment of the Organization for Economic Cooperation and Development (PISA OECD). As all recent TIMSS and PISA measures of achievement in mathematics are intercorrelated at .90 or higher, and because the two projects use the same scale, it is possible to merge the results of TIMSS 2015, TIMSS 2011, PISA 2015, and PISA 2012 to increase the number of countries for the analysis. We used each country’s highest score in any of these four studies. Because of the high stability of the national scores across time and across studies, this procedure is not essentially different from calculating average scores across the four studies.

Table 3.

Validation of the National Flexibility-Monumentalism Index.

Variable	r	n (countries)
Related dimensions of national culture
Monumentalism, based on World Values Survey items (Minkov, 2011)	−.86	32
Self-enhancement (Loughnan et al., 2011)	−.81	14
Long-term orientation, based on World Values Survey items (Minkov & Hofstede, 2012)	.77	21
Confucian work dynamism (Chinese Culture Connection, 1987; Hofstede, 2001)	.68	20
Measures of self-consistency or self-esteem
Consistency versus variability (Vignoles et al., 2016)	−.71	24
General self-esteem on Rosenberg’s scale (Schmitt & Allik, 2005)	−.66	34
Cognitive achievement
Average math achievement, 8th grade, on TIMSS tests in 2015 (Mullis, Martin, Foy, & Hooper, 2016)	.84	24
Same, after controlling for GDP per person in 2014 (The World Bank, 2016)	.81	20
Average math achievement, 8th grade, on TIMSS tests in 2011 (Mullis, Martin, Foy, & Arora, 2012)	.82	25
Same, after controlling for GDP per person in 2014 (The World Bank, 2016)	.77	21
Average math achievement on PISA tests of mathematics in 2015 (PISA, 2016)	.79	42
Same, after controlling for GDP per person in 2014 (The World Bank, 2016)	.76	38
Average math achievement on PISA tests of mathematics in 2012 (PISA, 2014)	.79	44
Same, after controlling for GDP per person in 2014 (The World Bank, 2016)	.75	40
Highest math achievement on any of TIMSS 2015, TIMSS 2011, PISA 2015, and PISA 2012	.86	47
Same, after controlling for GDP per person in 2014 (The World Bank, 2016)	.82	43

Note. All correlations are significant at the .001 level. TIMSS = Trends in International Mathematics and Science Study; PISA = Program for International Student Assessment.

The results in Table 3 validate our second and third hypotheses. The flexibility-monumentalism index is strongly correlated in the expected direction with measures of CWD, LTO, self-esteem, self-enhancement, and self-consistency. It is also an exceptionally strong positive predictor of national differences in school achievement in mathematics across 47 countries, yielding correlations of about .80 or more. Countries whose cultures are characterized by high flexibility and low monumentalism have high mathematics scores. As Table 3 shows, this association is reduced very weakly after controlling for national differences in economic wealth.

The LTO index in Minkov and Hofstede (2012) correlates with the variable in the penultimate row of Table 3 (highest mathematics achievement on any of the two latest TIMSS or PISA tests) at only .67 (p <.001, n = 26), whereas CWD in the Chinese Culture Connection correlates with that variable at only .51 (p = .038, n = 17). Our new flexibility-monumentalism index is thus the best cultural predictor of national differences in mathematics achievement. As it is well-known that national achievement in mathematics is strongly associated with national achievement in other domains, such as science achievement as measured by TIMSS and reading achievement as measured by PISA, flexibility-monumentalism is strongly correlated with general national educational achievement.

Table 4 provides correlations between the flexibility-monumentalism index and measures of national culture in two of the best-known models: Hofstede’s (2001) and Schwartz’s (2008). We also provide correlations with the national scores for all measures of self-construals in Vignoles et al. (2016) as one of those measures—consistency versus variability—is strongly and logically associated with our measure of flexibility versus monumentalism.

Table 4.

Correlations Between Flexibility-Monumentalism and Relevant Measures of National Culture.

Dimension	r	n (countries)
Hofstede’s original dimensions of national culture (Hofstede, 2001)
Individualism	.32*	42
Power distance	−.29	42
Uncertainty avoidance	−.21	42
Masculinity-femininity	−.02	42
Schwartz’s nation-level domain scores on measures of personal values (S. Schwartz, unpublished scores, personally provided in August 2016)
Harmony	−.01	49
Embeddedness	−.34*	49
Hierarchy	.12	49
Mastery	.22	49
Affective autonomy	.38**	49
Intellectual autonomy	.26	49
Egalitarianism	−.21	49
Models of selfhood (self-construals; Vignoles et al., 2016)
Difference–similarity	−.26	24
Self-containment–connection to others	.01	24
Self-direction–reception to influence	.21	24
Self-reliance–dependence	.33	24
Consistency–variability	−.71**	24
Self-expression–harmony	−.41**	24
Self-interest–commitment to others	−.27	24

Correlation significant at the .05 level. **Correlation significant at the .01 level.

While Hofstede’s scores are well-known and used in countless publications, Shalom Schwartz has not published his latest scores yet. Nevertheless, he kindly provided those scores to us and asked us not to use his previously published scores for any purposes. The low correlations between flexibility-monumentalism and the measures of Hofstede and Schwartz suggest that none of the measures in those models (except CWD/LTO) is closely associated with the construct discussed in this study.

Figure 2 visualizes the most important implication of this study: the relationship between flexibility-monumentalism and highest national achievement in mathematics.

Figure 2.

Visualization of the relationship between flexibility-monumentalism and schoolchildren’s best performance on math tests: TIMSS 2011, TIMSS 2015, PISA 2012, and PISA 2015.

Discussion

The goal of our study was to resolve the conceptual problems that plague Hofstede’s LTO dimension. We took a more focused approach, concentrating on selected LTO facets, which in our view highlight some of the most salient cultural characteristics on the world’s East–West geographic axis, from East Asia to Africa and Latin America. Those facets capture differences in self-enhancement and self-stability: a cultural propensity to encourage high self-esteem and self-confidence, as well as an invariant self that maintains its genuineness and consistency across situations. These two characteristics form a single dimension at the national level, even if they are not necessarily related at the individual level. We propose that at the root of these cultural traits is a societal tendency to exchange help and services, which is a component of the same dimension. Regular provision and exchange of help within a social network creates images of admirable persons and boosts the self-esteem of the helpers while creating societal interdependence. The latter requires people to have an invariant self so as to be reliable and predictable across situations. We support this view with Steven Heine’s theory of self-enhancement and self-stability and previous findings of a statistical relationship between LTO and national measures of self-reliance and independence.

As time orientation is not central to the concept that our new national index measures, we propose the name “flexibility versus monumentalism” for it, borrowed from the work of Minkov (2011, 2013). The motto of monumentalist cultures seems to be “Always be the same: feel good about yourself, and make others feel good about you.” Flexible cultures emphasize adaptability and a modest opinion of one’s self. This type of culture de-emphasizes help. Instead, it promotes self-reliance and independence.

It is noteworthy that our flexibility-monumentalism index is not associated with the self-reliance measure in Vignoles et al. (2016). The latter is a measure of the degree to which individuals in different societies feel comfortable or uncomfortable with self-reliance versus dependence on others. This measure is quite different from the measure in Green et al. (2005), measures of service to others and help as personal values in the World Values Survey, and our measure of the degree to which people like to help others (not the degree to which they like or dislike receiving help). It is quite possible to imagine societies in which there is some exchange of services, help, and interdependence, yet many people resent this arrangement. At the same time, there may be societies where interdependence tends to be seen as acceptable.

Our study is mostly about the structure and measurement of flexibility-monumentalism, as an improved variant of LTO. It does not focus on a crucial question: Why does this dimension predict national achievement in mathematics and other fields of modern education so well? This question warrants a study in its own right. Here, we can only advance a few hypotheses. The first is based on Heine’s theory: Self-enhancement creates complacency as self-enhancers would not feel a strong need for self-improvement. Heine’s theory targets the individual level of analysis but may also be applicable to the societal level. Complacency at that level would account not only for students who believe they know enough and need not study hard but also low parental pressure and lenient school practices. Minkov (2008, 2011) notes an interesting phenomenon revealed by the TIMSS studies. East Asian students have the most negative attitudes toward mathematics, whereas students from the lowest-achieving countries have the most positive attitudes toward that subject. One possible explanation is that East Asians are under strong societal pressure to excel in mathematics, resulting in high achievement and a negative attitude. In low-achieving societies, there is little pressure and consequently a positive attitude toward mathematics. This is consistent with the conclusions of van de Gaer Grisay, Schulz, and Gebhardt (2012): Country differences in educational benchmarks, standards, and norms explain country differences in achievement on PISA OECD tests as well as in the students’ self-concepts.

We must admit that the direction of the cause-and-effect relationship between flexibility-monumentalism and cognitive achievement is not fully clear. We may be right in assuming that monumentalist complacency depresses school achievement. But a reverse causality is not implausible. In societies with low average cognitive achievement, many people may not be aware of their deficiencies and disregard them in their self-assessments. Also, we cannot rule out the hypothesis that, because of their low cognitive achievement, many people in monumentalist societies overestimate their self-stability. It is cognitively simpler to see and present oneself as being always the same person than to assess properly the situations in which one acts differently. It is also possible that people with lower cognitive achievement are indeed less flexible and less adaptable to diverse situations (Minkov, 2017). Finally, monumentalism and low educational achievement may be in a complex two-way cause-and-effect relationship, reinforcing each other.

Another interesting question for future research is the origin of the national differences on the flexibility-monumentalism dimension and its plasticity. Does flexibility versus monumentalism evolve over time, how, and as a function of what ecosocial conditions?

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was sponsored financially by MediaCom Ltd., a multinational media agency, and was carried out with the participation of the Hofstede Center at Itim International, a cross-cultural management consultancy. MediaCom and Itim International have not influenced the study design or the analysis in this study in any way.

Notes

Author Biographies

Michael Minkov is professor of cross-cultural awareness at Varna University of Management (VUM), Sofia campus, Bulgaria. He holds a PhD in social anthropology from the State University of Sofia (Kliment Ohridski) and an MA degree in linguistics, also from that university. His main research interests and main publications are in the field of national cultures and societal differences. He is also interested in research methodology and data analysis.

Michael H. Bond is a Canadian male social psychologist intrigued by culture. He was born during World War II and intrigued by the ways in which our behaviors become different as a consequence of being born in a different time and place from one another. He was fortunate to be educated at the universities of Toronto and Stanford, whose reputation for excellence enabled him to gain access to the wider intellectual world from which he continues to learn about the forces in life that enable us to live together in this 21st century.

Pinaki Dutt is a research and strategy professional with close to 20 years of experience in understanding consumers in and across markets globally. With specific interest in consumer behavior and how choices are influenced, he specializes in the applied aspect of research and in predicting future possibilities of, and from, consumers. He has been specifically interested in understanding the deep and yet unseen forces of culture that have an implicit influence on everything that a person in a society does—in everyday life, conversation, and how they decide and choose.

Michael Schachner is operations manager at Itim International’s Hofstede Centre based in Helsinki, Finland, where he is in charge of the development of managerial tools and research projects in the fields of intercultural management and organizational culture. His main academic interest is the interrelationship of organizational culture and national culture. He graduated as “magister” in economics and in international economic and business studies from the University of Innsbruck, Austria.

Oswaldo Morales, PhD, is professor in administration, and director of online education. He holds a PhD in International Studies, from the Graduate School of Asia-Pacific Studies, University of Waseda, Japan, an MBA from ESAN, Peru, Master in Economics and Public Service Regulation from the University of Barcelona, Spain, Master in Business Law from the University of Lima and a law degree from the University of Lima, Peru. Extensive experience at a company level and consultant on international projects. His interests are in administration and organization, intercultural management, strategic analysis, competence development, and entrepreneurship. His multidisciplinary experience allows him to approach human capital development from a creative perspective, focusing on generating value. He is currently working on ways to incorporate cultural elements in project development in a global context.

Carlos Sanchez is director of the Institute of Research and Polls at the Universidad del Turabo. He does research in the fields of public opinion and predictive analysis in the Caribbean and Spanish-speaking United States. His main academic interests are in the fields of cross-cultural research, international management, and subjective well-being. He has a doctorate in business administration from Universidad del Turabo in Puerto Rico.

Janar Jandosova is a head of the Sange Research Center in Astana, Kazakhstan. She defended her PhD in general psychology at the Saint Petersburg State University. Since 1998, she has worked mostly on the field of sociology, doing nationwide surveys for the government, international organizations, and private companies. Her main academic interests and publications are linked to cultural and regional differences in corruption and entrepreneurship. She was a Fulbright Scholar at the University of California, Riverside in 1997-1998.

Yerlan Khassenbekov is a business consultant who received a Bachelor’s degree at the Almaty Institute of Management and Economics “Daulet” and an MBA from the A.B. Freeman School of Business at Tulane University. He specializes in studies of entrepreneurship. In 2015, he was in charge of a research project that discovered regional cultural features of entrepreneurship development in Kazakhstan. In 2017, he took part in projects delivered by the World Bank Group and the Asian Development Bank for the government of the Republic of Kazakhstan.

Ben Mudd holds an MA in screen translation from the University of Sheffield and has had a varied career in the IT sector, Teaching English as a Foreign Language (TRFL), and nongovernmental organizations (NGOs). His main academic interests are translation, intercultural communication, and sports for development.

References

Ashkanasy

Gupta

Mayfield

M. S.

Trevor-Roberts

(2004). Future orientation. In House

R. J.

Hanges

P. J.

Mansour

Dorfman

P. W.

Gupta

(Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 282-342). Thousand Oaks, CA: Sage.

Blackwell

L. S.

Trzesniewski

K. H.

Dweck

C. S.

(2007). Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development, 78, 246-263.

Chinese Culture Connection. (1987). Chinese values and the search for culture-free dimensions of culture. Journal of Cross-Cultural Psychology, 18, 143-164.

Fang

(2003). A critique of Hofstede’s fifth national culture dimension. International Journal of Cross-Cultural Management, 3, 347-368.

Gelfand

Bhawuk

Nishii

L. H.

Bechtold

(2004). Individualism and collectivism. In House

R. J.

Hanges

P. J.

Javidan

Dorfman

P. W.

Gupta

(Eds.), Culture, leadership, and organizations: The GLOBE study of 62 societies (pp. 437-512). Thousand Oaks, CA: Sage.

Green

E. G. T.

Deschamps

J. C.

Paez

(2005). Variation of individualism and collectivism within and between 20 countries: A typological analysis. Journal of Cross-Cultural Psychology, 36, 321-339.

Green

J. D.

Pinter

Sedikides

(2005). Mnemic neglect and self-threat: Trait modifiability moderates self-protection. European Journal of Social Psychology, 35, 225-235.

Heine

S. J.

(2003). An exploration of cultural variation in self-enhancing and self-improving motivations. In Murphy-Berman

Berman

J. J.

(Eds.), Nebraska symposium on motivation: Cross-cultural differences in perspectives on the self (Vol. 49, pp. 101-128). Lincoln: University of Nebraska Press.

Heine

S. J.

Hamamura

(2007). In search of East Asian self-enhancement. Personality and Social Psychology Review, 11(1), 4-27.

10.

Heine

S. J.

Lehman

D. R.

Peng

Greenholtz

(2002). What is wrong with cross-cultural comparisons of subjective Likert scales: The reference-group effect. Journal of Personality and Social Psychology, 82, 903-918.

11.

Hofstede

(2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations (2nd ed.). Thousand Oaks, CA: Sage.

12.

Kim

Y. H.

Pen Chiu

C.Y.

(2008). Explaining self-esteem differences between Chinese and North Americans: Dialectical self (vs self-consistency) or lack of positive self-regard. Self and Identity, 7(2), 113-128.

13.

Lee

R. B.

(1979). The !Kung San: Men, women, and work in a foraging society. Cambridge, UK: Cambridge University Press.

14.

Loughnan

Kuppens

Allik

Balazs

de Lemus

Dumont

. . . Haslam

(2011). Economic inequality is linked to biased self-perception. Psychological Science, 22, 1254-1258.

15.

Minkov

(2008). Self-enhancement and self-stability predict school achievement at the national level. Cross-Cultural Research, 42, 172-196.

16.

Minkov

(2011). Cultural differences in a globalizing world. Bingley, UK: Emerald.

17.

Minkov

(2013). Cross-cultural analysis: The science and art of comparing the world’s modern societies and their cultures (with contributions by Geert Hofstede). Thousand Oaks, CA: Sage.

18.

Minkov

(2017). Middle responding: An unobtrusive measure of national cognitive ability and personality. Personality and Individual Differences, 113, 187-192.

19.

Minkov

Hofstede

(2012). Hofstede’s fifth dimension: New evidence from the World Values Survey. Journal of Cross-Cultural Psychology, 43(1), 3-14.

20.

Mueller

C. M.

Dweck

C. S.

(1998). Praise for intelligence can undermine children’s motivation and performance. Journal of Personality and Social Psychology, 75, 33-52.

21.

Mullis

I. V. S.

Martin

M. O.

Foy

Arora

(2012). TIMSS 2011 international results in mathematics. Chestnut Hill, MA: TIMSS and PIRLS International Study Center.

22.

Mullis

I. V. S.

Martin

M. O.

Foy

Hooper

(2016). TIMSS 2015 international results in mathematics. Chestnut Hill, MA: TIMSS and PIRLS International Study Center.

23.

PISA. (2014). About PISA. Retrieved from http://www.oecd.org/pisa/aboutpisa

24.

PISA. (2016). PISA 2015 results (Volume I): Excellence and equity in education. Retrieved from http://www.oecd.org/pisa/aboutpisa

25.

Schmitt

D. P.

Allik

(2005). Simultaneous administration of the Rosenberg Self-Esteem Scale in 53 nations: Exploring the universal and culture-specific features of global self-esteem. Journal of Personality and Social Psychology, 89, 623-642.

26.

Schwartz

S. H.

(2008). Cultural value orientations: Nature and implications of national differences. Moscow, Russia: State University—Higher School of Economics Press.

27.

Van de Gaer Grisay

Schulz

Gebhardt

(2012). The reference group effect: An explanation of the paradoxical relationship between academic achievement and self-confidence across countries. Journal of Cross-Cultural Psychology, 43, 1205-1228.

28.

van Witteloostuijn

. (2016). What happened to Popperian falsification? Publishing neutral and negative findings: Moving away from biased publication practices. Cross Cultural & Strategic Management, 23, 481-508.

29.

Vignoles

V. L.

Owe

Becker

Smith

P. B.

Easterbrook

Gonzalez

. . . Bond

M. H.

(2016). Beyond the “East-West” dichotomy: Global variation in cultural models of selfhood. Journal of Experimental Psychology: General, 145, 966-1000.

30.

The World Bank. (2016). World Bank open data. Available from http://data.worldbank.org