Within-Culture Variations of Collectivism in Japan

Abstract

Modeling after Vandello and Cohen’s American Collectivism Index, the researcher developed a scale to measure collectivism in prefectures (similar to U.S. states) in Japan. The new scale was evaluated against results from the Japanese General Social Survey, a national survey of individuals conducted annually, and was tested for association with common correlates of cultural syndromes. As expected, the Japanese Collectivism Scale (JCS) was reliable and was significantly associated with results of the individual-level attitudinal survey and the correlates of collectivism. The JCS also showed within-culture variations of collectivism in Japan—variations that are important to consider when interpreting cross-cultural differences in attitudes and behavior.

Keywords

within-culture variation Japan collectivism scale

Empirical studies have shown that cultural variations play a crucial role in forming patterns in one’s cognition, attitudes, and behavior. In particular, the idea of societies contrasting on the basis of individualism and collectivism has increased in popularity, not only in the field of psychology but also in anthropology, economics, history, political science, business, and sociology.

Recently, Oyserman, Coon, and Kemmelmeier (2002b) reviewed empirical evidence for cultural variations in individualism and collectivism. They concluded that the reliability and validity of existing measurements are questionable and that the concept of individualism–collectivism was construed too broadly. Nevertheless, the effects of individualism and collectivism, identified by such “questionable” measurements, have been frequently used to compare cultural or cross-cultural differences. Accordingly, Oyserman, Coon, and Kemmelmeier (2002a) urged cross-cultural researchers to search for other promising general or foundational cultural schemas. In response to those findings and conclusions, Bond (2002), Fiske (2002), Kitayama (2002), and Miller (2002) all pointed out some limitations of Oyserman et al.’s study. Yet they agreed with the suggestion of using a variety of means to capture individualism–collectivism differences.

Bond (2002), Kitayama (2002), and Triandis (2001) together attested that extra-individual factors, such as norms, ecology, history, geography, resources, and so on, should be included to capture differences in individualism and collectivism. In other words, variations in collectivism should be expected within every culture, and failure to recognize such variations within a culture will distort researchers’ understanding of differences among cultures.

Efforts by Vandello and Cohen (1999) in creating a collectivism index based on such extra-individual factors are noteworthy. They found state-level variations in individualism and collectivism within the United States, one of the most cited examples of an individualistic culture. Such an index is an extremely valuable tool that complements existing methods and that could be equally indispensable for other countries.

The creation of the Japanese Collectivism Scale (JCS) is particularly vital in demonstrating within-cultural variations for the country that has been the most intensely studied example of an extremely collectivist culture (Bond, 2002). In the current literature, the bulk of studies that have examined the effects of individualism–collectivism have come from U.S.–Japan comparisons. However, in their extensive meta-analysis of the studies on individualism–collectivism, Oyserman et al. (2002b) concluded that Japan was not as collectivistic as previously reported, and that even Americans were slightly more collectivistic than people in Japan. To explain such inconsistency, Bond (2002) used Hofstede’s (1980) taxonomy to show that Japan and the United States would have been cultural neighbors rather than opposing deviations. Therefore, he concluded, exclusive reliance on the U.S.–Japan comparison could be misleading. Another explanation for such inconsistency was first made by Chen, Lee, and Stevenson (1995), who suggested that differences in response style among American and Japanese respondents could obscure the cross-cultural findings. They found that Japanese respondents tended to rate at the mid-point of the Likert-type scale, while American respondents tended to rate at the extreme values. Indeed, Schimmack, Oishi, and Diener (2005) demonstrated that when such response styles were controlled, individualism was indeed an important and valid cultural dimension in cross-cultural studies.

One of the major aims of this study is to propose another factor to explain such inconsistency by showing within-cultural variations in Japan. Given the within-cultural variations in the United States (as shown by Vandello & Cohen, 1999), it is likely that such variations also exist in Japan, which suggests that examinations of the effects of individualism–collectivism could be influenced by where the sample is drawn. That is, if one collects data from Tokyo, the most individualistic prefecture in Japan, and compares the result to data collected in the most collectivistic areas in the United States, such as Hawaii or Louisiana, interpretations might be difficult or misleading. In fact, a recent study by Kitayama, Ishii, Imada, Takemura, and Ramaswamy (2006) showed within-cultural variations on independent agency in Japan by comparing Hokkaido-born Japanese and non-Hokkaido-born Japanese. To date, however, there is no study to show the nationwide within-cultural variation in Japan. Thus, the purpose of this study is to show cultural variations in Japan by creating an individualism–collectivism index based on prefectures¹ and then to validate the index with the survey data taken from a large number of Japanese individuals.

Study 1: Creating the Japanese Collectivism Scale

Method

In accordance with the procedures of Vandello and Cohen (1999), prefecture-level data that might be relevant to the concepts of individualism and collectivism were identified. In addition to the measures in Vandello and Cohen’s American Collectivism Index, some indicators that are theoretically relevant to and suitable for an indigenous construct of Japanese individualism and collectivism were gathered and then narrowed. The individual items would be moderately related so as to yield a scale with relatively high reliability, but would also be non-redundant with one another so as to achieve a parsimonious scale. The following items were included in the scale, with items delineated by an asterisk also being included in the American Collectivism Index.

Divorce to marriage ratio (reverse scored. Ministry of Health, Labour and Welfare, 2006)

Previous researchers found that divorce and individualism are closely associated with each other. For instance, Inglehart (1990) claimed that as individualists tend to seek fulfillment of their own self-interests, they tend to put the self first when they divorce. Furthermore, according to Toth and Kemmelmeier (2009), individualists tend to be “unwilling to sacrifice their personal fulfillment for a bad marriage, even when the act of divorce implies great emotional and financial costs” (p. 2). Indeed, previous studies, including a cross-cultural study of Japan along with 23 other nations, indicated that individualistic cultures have a higher divorce rate in comparison to collectivistic cultures (Dion & Dion, 1996; Naroll, 1983; Toth & Kemmelmeier, 2009; Vandello & Cohen, 1998). Similar to the reasons for divorce in Western societies, divorces in today’s Japan, as reported by Kumagai (1983), are more motivated by nontraditional reasons (e.g., pursuing personal happiness) than traditional reasons (e.g., focusing lineage). That is, the general meanings of marriage and divorce have shifted from collectivistic to rather individualistic ones. The value of family lineage has faded, while personal affection has become the main focus in Japanese marriages and divorces. Therefore, including divorce rate as one of the indicators of the JCS is suitable. This study used divorce to marriage ratio as a divorce rate because it is the most consistently and widely used index in Japan.

Percentage of households with three generations living together (Statistics Bureau, 2006)

Pyke (1999) attested that individualistic families have minimal expectations for providing elder care and they tend to assume that members must be self-reliant, while collectivistic families focus on close family ties, strong commitment to other family members, and a high degree of contact and interdependence. Furthermore, Pyke and Bengtson (1996) found that individualists minimize their elderly caregiving and tend to rely on formal support, while collectivists use caregiving to construct family ties. Therefore, collectivists are more likely to take personal care of their parents by sharing living quarters with them and assuming caregiving obligations. Consequently, the percentage of households with three generations living together is included as one of the indicators of the JCS.

Percentage of elderly people (aged 65+ years) living alone (reverse scored; Statistics Bureau,2006)

Historically, elderly individuals have been respected and valued in Japanese society. Furthermore, due to the deeply rooted tradition of family obligation, Japanese elderly individuals have been historically taken care of by their own family members. In contemporary Japan, however, the hospital has been a major resource for elderly care. According to Asahara, Konishi, Soyano, and Davis (1999), such a trend is due to the fact that the younger Japanese generation, born after World War II, has prioritized personal needs over collectivistic family obligations. Therefore, as Vandello and Cohen (1999) included the percentage of elderly persons living alone as part of the American Collectivism Index, the JCS also uses this issue to evaluate the individualism–collectivism dimension in Japan. This indicator is reversed scored in the JCS. It should be noted that the data on elderly persons living alone from the Statistics Bureau do not include elderly persons who reside in hospitals, nursing homes, and community facilities.

Percentage of nuclear family households (reverse scored; Statistics Bureau, 2006)

Family structure is one of the vital factors related to individualism–collectivism. Cross-cultural researchers attest that collectivists in Asian cultures prioritize the extended family structure over a nuclear one along with collectivistic values, such as obedience, duty, in-group harmony, and interdependent personal relationships (e.g., Ivey, Ivey, & Simek-Morgan, 1997; Tata & Leong, 1994). However, the Japanese family structure has shifted drastically. Japan is now characterized as a Westernized “high tech” culture that is typified by nuclear family life (Early & Gibson, 1998). Because extended family structures foster collectivistic values, such shift of family structure in Japan may encourage individualistic values. Indeed, Tsutsumi (2002) found that individuals who lived in an extended family structure tended to have a strong desire for succession of the collectivistic values. She further claimed that when the external appearance of the family structure changed, internal changes, such as ideology, also occurred. Therefore, while the households of Japanese collectivists tend to include extended family, Japanese individualists are more likely to live in nuclear family arrangements.

Percentage of people living alone (reverse scored; Statistics Bureau, 2006)

One of the main characteristics that distinguishes individualism from collectivism is one’s independence. Individualists tend to seek autonomy, self-reliance, individual privacy, freedom, and individual identity (Kağitçibasi, 1996), while collectivists tend to seek interdependence, communal living, and low freedom (Singelis, Triandis, Bhawuk, & Gelfand, 1995). Therefore, individuals who seek such independence are more likely than those with collectivistic tendencies to live alone and to view themselves as being autonomous.

Results

Data for each item were standardized across prefectures and reverse-scored where noted above. The overall collectivism score was the sum of the individual z scores for each of the five items. The z scores were then transformed by multiplying the means by 20 and adding this result to 50, thereby resulting in a collectivism score that would be positive and range from approximately 1 to 100. On this new JCS, a higher score represents higher collectivistic tendencies.

The Cronbach’s alpha for the newly developed scale was .85. With the exception of divorce rate, all items were important to the JCS, such that the deletion of an item would have decreased the overall alpha. Table 1 addresses two different types of reliability: (1) alpha if item were deleted and (2) corrected item-total correlations. Although the alpha would have increased if divorce rate were deleted (.86), this item was retained because it is important to the concept of individualism–collectivism and only minimally impacted the overall alpha. As for corrected item-total correlations, all items were above .40, which is the recommended value by Gliem and Gliem (2003). In addition, intercorrelations among all five items are demonstrated in Table 2. Table 3 represents the prefecture ranking on the JCS, and Figure 1 displays the ranking in map form.

Table 1.

Reliability Statistics for the Five Collectivism Indicators

Indicators	Corrected Item-Total Correlations	Alpha if Item Deleted
Divorce to marriage ratio (R)	.45	.86
Percentage elderly living alone (R)	.84	.78
Percentage nuclear family (R)	.77	.82
Percentage living alone (R)	.73	.80
Percentage three-generation households	.74	.80

Note. R indicates a reversed score. Total standardized α = .85.

Table 2.

Correlation Matrix for the Indicators of the Japanese Collectivism Scale

Indicator	1	2	3	4
1. Divorce to marriage ratio (R)
2. Percentage elderly living alone (R)	.49
3. Percentage nuclear family (R)	.41	.71
4. Percentage living alone (R)	.40	.79	.64
5. Percentage three-generation households	.30	.74	.75	.58

Note. R indicates a reversed score.

Table 3.

Prefecture Rankings on the Japanese Collectivism Scale

Rank	Prefecture	Score	Rank	Prefecture	Score
1	Yamagata	96.97	25	Okayama	49.30
2	Toyama	87.45	26	Kumamoto	47.72
3	Niigata	82.78	27	Yamanashi	47.57
4	Fukui	81.72	28	Kagawa	45.27
5	Gifu	74.85	29	Kochi	45.32
6	Shiga	74.55	30	Tokushima	45.09
7	Fukushima	72.25	31	Yamaguchi	44.51
8	Ishikawa	68.93	32	Oita	43.85
9	Iwate	67.72	33	Wakayama	43.53
10	Akita	67.28	34	Ehime	42.15
11	Tottori	67.15	35	Miyazaki	39.93
12	Ibaraki	65.73	36	Nagasaki	39.16
13	Shizuoka	65.51	37	Hiroshima	39.08
14	Saga	65.31	38	Chiba	35.49
15	Nagano	64.81	39	Hyogo	34.66
16	Tochigi	61.59	40	Fukuoka	33.74
17	Aomori	61.05	41	Gunma	32.52
18	Shimane	57.95	42	Kyoto	30.22
19	Mie	57.31	43	Okinawa	30.08
20	Kagoshima	55.63	44	Kanagawa	28.19
21	Miyagi	54.06	45	Osaka	17.31
22	Nara	53.63	46	Hokkaido	16.46
23	Aichi	50.98	47	Tokyo	6.28
24	Saitama	49.50

Note: Higher score represents higher collectivism.

Figure 1.

Regional Patterns of Collectivism in Japan

Study 2: Validating the Japanese Collectivism Scale at the Individual Level

Method

Another purpose of this study was to validate the JCS with responses from a large number of individuals who answered a national survey. In taking this approach, the assumption was that the prefectural differences in collectivism would lead to differing psychological orientations in individuals (Triandis, Leung, Villareal, & Clark, 1985). Similar to the process reported by Vandello and Cohen (1999), the JCS was used to predict attitudinal responses to national survey items from the 2006 Japanese General Social Survey (JGSS). The Japanese General Social Surveys are designed and carried out by the JGSS Research Center at Osaka University of Commerce (2002–2003, 2005) (Joint Usage / Research Center for Japanese General Social Surveys, accredited by the Minister of Education, Culture, Sports, Science and Technology), in collaboration with the Institute of Social Science at the University of Tokyo. The datasets are distributed by SSJ Data Archive, Institute of Social Science, and the University of Tokyo. In the 2006 JGSS, 4,254 individuals were identified through a multistage random sampling procedure that targeted adults living in households throughout Japan. The purpose of the JGSS was to solicit political, sociological, and economic information from people living in Japan. In addition, the survey included other items asking about respondents’ psychological and behavioral tendencies. A total of seven JGSS questions that are associated with individualism–collectivism were elicited. According to previous researchers (e.g., Triandis, 1989; Yamaguchi, 1994), one of the salient collectivistic characteristics was how individuals prioritize their goals in relation to group goals. That is, collectivistic individuals prioritize group goals over personal goals. Of the seven JGSS questions, one item closely paralleled collectivism (Scale 1: Question 1, listed below). The other six items were indirectly associated with collectivism (Scale 2: Questions 2 to 7, listed below). Unaggregated individual scores of the 2006 JGSS were used to validate the JCS. Because all items were answered with a 7-point Likert-type scale (1 = strongly agree to 7 = strongly disagree), prefectural mean scores of Scale 1 and Scale 2 were used for analysis. All items were reversed so that higher scores represented greater collectivistic tendencies. For the JCS, the means of the z scores for the five items were calculated. Each of the seven questions from the 2006 JGSS is listed below.

Scale 1:

One must put familial well-being and interest before one’s own.

Scale 2:

Children must make efforts to do something that would bring honor to their parents.

One should take good care of the ancestral grave and pass on the next generation.

A married adult man ought to provide financial support for his own parents.

A married adult woman ought to provide financial support for her own parents.

A married adult man ought to provide financial support for the parents of his wife.

A married adult woman ought to provide financial support for the parents of her husband.

Question 2 was selected because honoring one’s parents is one of the most salient characteristics of Asian collectivism (Chan, 1994). Another tradition of Asian culture is to worship one’s ancestors (Heppner, Heppner, Lee, Wang, Park, & Wang, 2006). Japanese people, even today, value ancestor worship and maintaining family graves. Such traditions strongly influence the Japanese family and lead to continuity of the family line and intergenerational connectedness (Tanaka, 2007). Fiske, Markus, Kitayama, and Nisbett (1998) claim that one of the characteristics of collectivism is to meet such family obligations and to conform to such traditional norms. Given that Question 3 specifically asks the extent to which participants agree with such family obligations as ancestor worship and maintaining family graves, it was included in Scale 2. According to Falicov (2001), collectivists tend to value intergenerational connectedness, and shared family financial responsibilities reflect various forms of interdependence among family members. Questions 4–7 were chosen because they measure participants’ willingness to share such responsibilities with extended family members.

Results

First, examinations of internal consistency of Scale 2 and the relationship between Scale 1 and Scale 2 were performed. The Cronbach’s alpha was .79, and as expected, both scales were moderately associated with each other, r(45) = .34, p < .0001. In addition, the intra-class correlation analyses of both Scale 1 and Scale 2 by prefecture were conducted. The results indicated that intra-class correlations for Scale 1 and Scale 2 with prefecture were r_ICC(49.30) = .04, p = .02, and r_ICC(35.53) = .02, p = .05, respectively. This result indicated that although the degree of relationship was small, individuals in each prefecture endorsed the psychological measures of collectivism in a significantly similar manner.

Second, to validate the JCS with Scales 1 and 2, two random intercept multilevel regression analyses were performed. In these analyses, Scale 1 or Scale 2 was the Level 1 (individual-level) variable, while the JCS was the Level 2 (group-level) variable. That is, the analyses showed whether Scales 1 and 2 would be predicted by the JCS. In order to show the true significance of the regional variation in Japan, unaggregated data were used. The JCS can be correlated with survey scores at the individual, unaggregated level by giving each respondent a collectivism score based on the score of his or her reported home prefecture. The first random intercept multilevel regression analysis showed that the JCS significantly predicted Scale 1 (b = .18, SE = .02, p <.0001), which measured the most paralleled item of collectivism. The second random intercept multilevel regression analysis demonstrated that Scale 2, an indirectly paralleled item of collectivism, was also significantly predicted by the JCS (b = .98, SE = .02, p < .0001). These analyses revealed that the estimated effect of the JCS on Scale 1 was approximately twice as great as on Scale 2, in the expected direction.

Study 3: Relationship of the JCS With Correlates of Collectivism

Following the procedures of Vandello and Cohen (1999), the relationship of the JCS was also tested for some correlates of individualism and collectivism. After World War II, the regional distribution of age groups in Japan was drastically altered due to economic development, as young people moved towards large cities. It is thus plausible that contemporary regional variations of collectivism in Japan may be contaminated by the variation in age distribution. Therefore, in the present exploration, age was controlled in all hypotheses tested.

Correlate 1: Amount of Gross Domestic Product per capita (Cabinet Office, 2010)

Triandis, Bontempo, Villareal, Asai, and Lucca (1988) attested that affluence is one of the important correlates of individualism because affluence implies that a person has the ability to be autonomous by being on his or her own. Furthermore, Hofstede (1980) reported a stunning correlation (.82) between Gross National Product (GNP) per capita and individualism and also claimed (Hofstede, 2001) that Gross Domestic Product per capita (GDP) is one of the correlates for individualism. While GDP is negatively associated with collectivism, it was expected that individual level of income would be negatively associated with collectivism. As expected, affluence was significantly negatively correlated with collectivism, r(45) = −.35, p < .01. In addition, when the average age was controlled, the correlation coefficient between the JCS and GDP decreased, r_p(44) = −.26, p < .05, yet it was still significant.

Correlate 2: Percentage of the tertiary economic sector is negatively associated with collectivism

Triandis (2001) attested that complex cultures, where individualists tend to live, are more likely to have service- and information-driven economies. Thus, the relationship between the JCS and this type of economic sector (Statistics Bureau, 2007b), particularly the tertiary sector (also known as the service- and information-driven industry), was examined. As expected, the percentage of tertiary sector in the total economy was inversely associated with the JCS, r(45) = −.45, p < .01. When age was controlled, the relationship between the percentage of the tertiary economic sector and the JCS increased, r_p(44) = −.50, p < .001, indicating that age has an influence on the relationship.

Correlate 3: Percentage of the primary economic sector is positively related with collectivism

In contrast to the second correlate, Vandello and Cohen (1999) investigated the relationship between farming practices and collectivism. They attest that agricultural farming practices require many individuals to cooperate together and to coordinate individuals’ time and efforts to effectively achieve greater productivity. Therefore, such cooperation and coordination may lead to collectivistic tendencies. Indeed, they found significant relationships between collectivism and historical farming practices, and thus concluded that agricultural farming practice is, in fact, a correlate of collectivism. In the present study, the percentage of the primary economic sector (including industries that collect natural resources, such as mining, forestry, farming, grazing, hunting and gathering, and fishing) was investigated as one correlate of collectivism (Statistics Bureau, 2007b). The underlying assumption is that such industries also require the cooperation and coordination of groups. As expected, the percentage of the primary economic sector was positively associated with the JCS, r(45) = .33, p < .05. When age was controlled, the relationship between the percentage of the primary economic sector was no longer significant, r_p(44) = .23, ns, indicating that age has significant influence on the relationship between the primary economic sector and the JCS.

Correlate 4: Greater degree of prefecture transiency is negatively associated with collectivism (Statistics Bureau, Ministry of Internal Affairs and Communications, 2009)

Triandis et al. (1988) claimed that one of the correlates of individualism is having rapid social mobility, given that transiency makes the control of groups less certain. Triandis (1989) further elaborated that transiency influences individualism–collectivism, using the concept of cultural looseness versus tightness. In tight (collectivistic) cultures, norms are easily retained and kept intact because tight cultures tend to be homogeneous. But in transient communities or prefectures, it may be difficult to retain group norms. In the present study, the degree of prefecture transiency was calculated by figuring the average rate of in- and out-migration from other prefectures for three different years (2000, 2005, and 2008). Higher scores represent greater prefecture transiency. The internal consistency of in- and out-migration for the 3 years was .98, which indicates that the in- and out-migration was very stable over that period. As expected, higher prefecture transiency was negatively associated with the JCS, r(45) = −.48, p < .001. When age was controlled in the relationship, the correlation coefficient decreased. However, it was still significant, r_p(44) = −.40, p < .01.

Correlate 5: Percentage of household Internet use is negatively correlated with collectivism. Triandis (2001) asserted that the more complex the culture, the more individualistic it is likely to be. As indices of complex culture, he listed the percent of population that is urban, the size of cities, and so forth. He also listed personal computer use per capita. Therefore, in the present study, the relationship between the percentage of household Internet use per prefecture (Statistics Bureau, 2007a) and the JCS was examined. As expected, they were significantly negatively associated with each other, r(45) = −.35, p < .01. When age was controlled, the relationship was still significant, r_p(44) = −.27, p < .05.

Correlate 6: Percentage of individuals who are age 10 or over and involved in volunteer services last year (Statistics Bureau, 2007c)

Numerous studies have investigated the effects of individualism–collectivism on volunteerism. The rationale of such investigations is that collectivists tend to emphasize the good of the whole and try to maintain relationships with the group, while individualists prioritize their personal goals over group goals. According to some researchers, collectivists tend to be more likely to volunteer in comparison to individualists (Hofstede, 2001; Wilson & Musick, 1997). However, other researchers found conflicting results. For instance, Kemmelmeier, Jambor, and Letner (2006) found individualism to be positively related to charitable giving and volunteerism. One recent study further demonstrated that individualists and collectivists did not differ in their willingness to volunteer; rather, they volunteered for different types of activities. For instance, collectivists were more involved and motivated to participate in volunteer work with collectivistic values, such as strengthening social ties, while individualists tended to participate in activities associated with individualistic values, such as career-related volunteer objectives (Finkelstein, 2010). Therefore, in the present study, the relationship between the JCS and the percentage of individuals age 10 or over who were involved in volunteer activities related to collectivist values was examined. The analyzed data specifically excluded activities related to individualist values, such as activities that support arts, recreation, educational causes, youth and women development programs as these were thought to reflect individuals’ self-actualization or self-promotion (Kemmelmeier et al., 2006). As expected, collectivistic volunteering and the JCS were significantly positively related to each other, r(45) = .56, p < .0001. Although age was controlled, there was no significant difference in such relationships, r_p(44) = .51, p < .0001. Because personal affluence is related to volunteerism (van de Vliert, Huang, & Levine, 2004), the average personal income per prefecture (Cabinet Office, 2010) was controlled. However, there was no crucial difference in the relationship between volunteerism and the JCS, r_p(44) = .55, p < .0001.

Discussion

The purpose of this study was to create a Japanese Collectivism Scale based on contemporary Japanese society and to validate it with individuals’ attitudes taken from the most current national data set. Furthermore, the author sought to test hypotheses in accordance with Vandello and Cohen’s (1999) study. As expected, the prefecture-level scale was significantly correlated with the individual level of collectivism, as reported by respondents on the national attitudinal survey. Furthermore, the scale was also significantly associated with the cultural syndromes of collectivistic correlates. As Bond (2002) suggested, exploring the importance of individual and extraindividual factors, socialization practices, and institutional processes is vital for the field of cross-cultural psychology because these factors vary in cross-cultural groups. The results of this study showed that some factors are universally influential on individualism–collectivism dimensions for both the United States and Japanese societies, while some are specifically influential for contemporary Japanese society. For instance, looking at three items of the JCS, these items were identical to those used in the American Collectivism Index, indicating that some living arrangements and divorce rates are vital indices for collectivism in both Japan and the United States. As for divorce rate, the causality of divorce may be different between Japan and the United States. In the contemporary Japanese society, the divorce rate has been drastically changing because of some factors, such as increasing young marriage, having an unstable job, and so forth. To date, however, there is no definitive way to determine causality of increased divorce in Japan. Therefore, such factors were not controlled to calculate the divorce rate. The other two items of the JCS are related to family structure (nuclear family and three generations living together). These indicators are particularly associated with Japan’s modernization and with Westernization influences after World War II. Indeed, Miyanaga (1993) reported that individualism has spread actively in Japan due to modernization and Westernization.

In the present investigation, age was controlled to investigate the correlates of collectivism, given that economic development in Japan drastically altered the regional distribution of age groups as young individuals moved towards large cities while elderly individuals stayed in the countryside. Yet the results demonstrated that age was not a crucial factor for the relationship between the JCS and the correlates, with one exception: the relationship between the JCS and the percentage of the primary economic sector as a correlate. When age was controlled, that relationship was no longer significant. This finding suggests that age significantly influences the relationship between the JCS and the percentage of the primary economic sector.

Another important finding from the present investigation is that the scale also showed the within-culture variation of collectivism in Japan. Although geographical factors were not tested, mapping the scores helps us to see the impact of geography on collectivism. The majority of collectivistic prefectures are in the northern part of Japan (the Tohoku district) and the central part of Japan (Chubu district). Indeed, among the top 10 most collectivistic prefectures, nine were from Tohoku or Chubu. On the other hand, the more individualistic prefectures are spread throughout Japan. However, there is a trend. That is, 8 out of 10 most individualistic prefectures include the government-appointed major cities (seirei shitei toshi), which are the most urbanized cities in Japan. Urbanization, which is closely linked to modernization and industrialization, generally shifts the population structure by moving the younger generations to big cities for better employment opportunities. Together, this trend suggests that urbanization may have greatly influenced regional variations of collectivism in Japan.

A prefecture whose low collectivism score does not follow the trend is Hokkaido. Geographically, Hokkaido is an isolated island, far away from the center of major Japanese cities. Yet Hokkaido is the second most individualistic prefecture. As the American West was the last remaining frontier in the United States, so Hokkaido was the last frontier in Japan. In the middle of the Meiji Restoration (1869–1882), the government sent pioneers to uncultivated land in Hokkaido to strengthen its economy (Tajima & Ito, 2009). Given that living in a frontier area is an important correlate of individualism (Triandis et al., 1988), Hokkaido may in fact be one of the most individualistic prefectures. Indeed, Kitayama et al. (2006) concluded that the Hokkaido pioneers’ motivations for settlement were deeply ingrained in their individualistic tendencies.

Some limitations of the present study should be noted. One of the limitations is that the JCS lacks content validity. The present study, in general, follows the conceptualization of Vandello and Cohen’s work, which investigated the variations of collectivism in the United States (1999). As a result, the study itself has its own limitation in terms of the content of the construct of collectivism as it applies to individuals in Japanese culture. Furthermore, the JCS does not include psychological measures of interpersonal relationship orientations that are sensitively evaluated for Japanese collectivism. In Japan, Yamaguchi (1994) has investigated some collectivism factors using an emic approach. He specifically suggested psychological attachment and fear of rejection as Japanese individuals’ collectivistic tendencies (Yamaguchi, 1994) by using empirical data. Yamaguchi argued that the collectivistic tendency of sacrificing one’s self-interest for the group is due to an expectation of reward from the group in the future; similarly, collectivistic individuals surrender their personal goals because of their expectations of punishment from group members. He further attested that lower need for uniqueness, higher external locus of control, and higher interpersonal sensitivity are all related to Japanese collectivistic tendencies. In the present study, such psychological measures were not investigated due to the lack of prefecture-based, large data that directly address Japanese collectivism. Rather, the JCS is developed mainly from the living arrangement or family structure. This is primarily because the current work focused on those items related to collectivism in cross-cultural work, particularly works that studied regional variations. As a result, however, the findings of this study reveal that living arrangements and divorce rate may prove to be cross-culturally common items of collectivism producing within-culture variations.

Another limitation of this study is that that the hypothesis testing of the relationships among correlates of collectivism heavily relied on Western literature. Thus far, the correlates of collectivism selected for the present study have been mainly explored among Western societies. It is urgent to examine the correlates of collectivism with more emic, or culturally informed, analysis suitable for the Japanese contemporary culture. Therefore, it is essential to examine the roles of common correlates of collectivism to test the reciprocal causality of antecedents and consequences with the JCS. As such, future research should examine the relationships using a cross-lagged effect model to demonstrate whether the selected correlates have actually preceded or followed the level of individualism–collectivism.

The within-culture variation of collectivism in Japan is vital information for researchers collecting data on cross-national differences. Such variation helps to explain the discrepancy between Oyserman et al.’s (2002b) conclusion—that Japan is not more collectivistic than America—and the general perception of Japan as a prototype of collectivism. Furthermore, Yamawaki, Parrot, and Spackman (2011) investigated the differing effects of individualism and collectivism on shame experience, in both within-culture and between-culture settings. As Oyserman et al. (2002a) suggested and the existing measures of individualism–collectivism seem to bear out, local phenomena, such as ecological factors, should be investigated and have a place in within-cultural psychology. The JCS is a useful tool to assess regional variations in collectivism in Japan and awaits further validation studies.

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Asahara

Konishi

Soyano

Davis

(1999). Long-term care for the elderly in Japan. Geriatric Nursing, 20, 23-27.

Bond

M. H.

(2002). Reclaiming the individual from Hofstede’s ecological analysis—A 20-year odyssey: Comment on Oyserman et al. Psychological Bulletin, 128, 73-77.

Cabinet Office. (2010). Heisei 19 nen Kenminn keizai keisansho [The 2007 annual report on prefectural-level accounts]. Tokyo: Cabinet Office of Japan.

Chan

D. K.-S.

(1994). Colindex: A refinement of three collectivism measures. In Kim

Triandis

H. C.

Kagitcibasi

Choi

S.- C.

Yoon

(Eds.), Individualism and collectivism: Theory, methods, and applications (pp. 200-210). Thousand Oaks, CA: SAGE.

Chen

Lee

S.- Y.

Stevenson

H. W.

(1995). Response style and cross-cultural comparisons of rating scales among East Asian and North American students. Psychological Science, 6, 170-175.

Dion

K. K.

Dion

K. L.

(1996). Cultural perspectives on romantic love. PersonalRelationships, 3, 5-17.

Early

C. P.

Gibson

C. B.

(1998). Taking stock in our progress on individualism–collectivism: 100 years of solidarity and community. Journal of Management, 24, 265-304.

Falicov

C. J.

(2001). The cultural meaning of money. American Behavioral Scientist, 45, 313-328.

Finkelstein

M. A.

(2010). Individualism/collectivism: Implications for the volunteer process. Social Behavior and Personality, 38, 445-452.

10.

Fiske

A. P.

(2002). Using individualism and collectivism to compare cultures—A critique of the validity and measurement of the constructs: Comment on Oyserman et al. Psychological Bulletin, 128, 78-88.

11.

Fiske

A. P.

Markus

H. R.

Kitayama

Nisbett

R. E.

(1998). The cultural matrix of social psychology. In Gilber

Fiske

Lindzey

(Eds.), The handbook of social psychology (4th ed., pp. 915-981). Boston: McGraw Hill.

12.

Gliem

(2003). Calculating, interpreting, and reporting Cronbach’s alpha reliability coefficient for Likert-type scales. Paper presented at the 2003 Midwest Research-to-Practice Conference in Adult, Continuing and Community Education, Columbus, OH.

13.

Heppner

P. P.

Heppner

M. J.

Lee

Wang

Y.- W.

Park

Wang

(2006). Development and validation of a collectivist coping styles inventory. Journal of Counseling Psychology, 53, 107-125.

14.

Hofstede

(1980). Culture’s consequences: International differences in work-related values. Beverly Hills, CA: SAGE.

15.

Hofstede

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

16.

Inglehart

(1990). Culture shift in advanced industrial society. Princeton, NJ: Princeton University Press.

17.

Ivey

A. E.

Ivey

M. B.

Simek-Morgan

(1997). Counseling and psychotherapy: A multicultural perspective (4th ed.). Boston: Allyn & Bacon.

18.

Kağitçibasi

Ç.

(1996). The autonomous-relational self: A new synthesis. European Psychologist, 1, 180-186.

19.

Kemmelmeier

Jambor

E. E.

Letner

(2006). Individualism and good works: Cultural variations in giving and volunteering across the United States. Journal of Cross-Cultural Psychology, 37, 327-344.

20.

Kitayama

(2002). Culture and basic psychological processes—Toward a system view of culture: Comment on Oyserman et al. Psychological Bulletin, 128, 89-96.

21.

Kitayama

Ishii

Imada

Takemura

Ramaswamy

(2006). Voluntary settlement and the spirit of independence: Evidence from Japan’s northern frontier. Journal of Personality and Social Psychology, 91, 369-384.

22.

Kumagai

(1983). Changing divorce in Japan. Journal of Family History, 8, 85-108.

23.

Miller

J. G.

(2002). Bringing culture to basic psychological theory—Beyond individualism and collectivism: Comment on Oyserman et al. Psychological Bulletin, 128, 97-109.

24.

Ministry of Health, Labour and Welfare. (2006). Heisei 17 nen Jinkoudoutai tokei [The 2005 vital statistics in Japan]. Tokyo: Ministry of Health, Labour and Welfare.

25.

Miyanaga

(1993). The creative edge: Emerging individualism in Japan. New Brunswick, NJ: Transaction Publisher.

26.

Naroll

(1983). The moral order. Beverly Hills, CA: SAGE.

27.

Osaka University of Commerce. (2002–2003, 2005). Japanese General Social Surveys: JGSS. Tokyo: Institute of Social Science at the University of Tokyo.

28.

Oyserman

Coon

H. M.

Kemmelmeier

(2002a). Cultural psychology, a new look: Reply to Bond (2002), Fiske (2002), Kitayama (2002), and Miller (2002). Psychological Bulletin, 128, 110-117.

29.

Oyserman

Coon

H. M.

Kemmelmeier

(2002b). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72.

30.

Pyke

K. D.

(1999). The micropolitics of care in relationships between aging parents and adult children: Individualism, collectivism, and power. Journal of Marriage and the Family, 61, 661-672.

31.

Pyke

K. D.

Bengtson

V. L.

(1996). Caring more or less: Individualistic and collectivist systems of family eldercare. Journal of Marriage and the Family, 58, 379-392.

32.

Schimmack

Oishi

Diener

E. D.

(2005). Individualism: A valid and important dimension of cultural differences between nations. Personality and Social Psychology Review, 9, 17-31.

33.

Singelis

T. M.

Triandis

H. C.

Bhawuk

Gelfand

M. J.

(1995). Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. Cross-Cultural Research: The Journal of Comparative Social Science, 29, 240-275.

34.

Statistics Bureau. (2006). Heisei 17 nen kokusei chousa dai ichiki kihon syukei kekka [The 2005 population census of Japan: Results of first basic complete tabulation]. Tokyo: Ministry of Internal Affairs and Communications.

35.

Statistics Bureau. (2007a). Heisei 18 nen shakai seikatsu kihon chosa—inta-netto no riyou [The 2006 basic study of living—Internet use]. Tokyo: Ministry of Internal Affairs and Communications.

36.

Statistics Bureau. (2007b). Heisei 18 nen jigyousho.kigyoutoukei chousa kekka [The 2006 Economic Census]. Tokyo: Ministry of Internal Affairs and Communications.

37.

Statistics Bureau. (2007c). Heisei 18 nen nen shakai seikatsu kihon chosa—borantia-katsudousharitu [The 2006 basic study of living—Volunteer activity]. Tokyo: Ministry of Internal Affairs and Communications.

38.

Tajima

Ito

(2009). The historical development and current status of the higher agricultural education system in Japan. Journal of Developments in Sustainable Agriculture, 4, 29-40.

39.

Tanaka

(2007). Graves and families in Japan: Continuity and change. The History of the Family, 12, 178-188.

40.

Tata

S. P.

Leong

F. T. L.

(1994). Individualism–collectivism, social-network orientation, and acculturation as predictors of attitudes toward seeking professional psychological help among Chinese Americans. Journal of Counseling Psychology, 41, 280-287.

41.

Toth

Kemmelmeier

(2009). Divorce attitudes around the world: Distinguishing the impact of culture on evaluations and attitude structure. Cross-Cultural Research, 43, 280-297.

42.

Triandis

H. C.

(1989). The self and social behavior in differing cultural contexts. Psychological Review, 96, 506-520.

43.

Triandis

H. C.

(2001). Individualism–collectivism and personality. Journal of Personality, 69, 907-924.

44.

Triandis

H. C.

Bontempo

Villareal

M. J.

Asai

Lucca

(1988). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54, 323-338.

45.

Triandis

H. C.

Leung

Villareal

Clark

F. L.

(1985). Allocentric vs. idiocentric tendencies: Convergent and discriminant validation. Journal of Research in Personality, 19, 395-415.

46.

Tsutsumi

(2002). Succession of stem families in rural Japan: Cases in Yamanashi Prefecture. International Journal of Japanese Sociology, 10, 69-79.

47.

Van de Vliert

Huang

Levine

R. V.

(2004). National wealth and thermal climate as predictors of motives for volunteer work. Journal of Cross-Cultural Psychology, 35, 62-73.

48.

Vandello

J. A.

Cohen

(1998). Collectivism, traditionalism, and domestic violence. Unpublished manuscript, University of Illinois at Urbana-Champaign.

49.

Vandello

J. A.

Cohen

(1999). Patterns of individualism and collectivism across the United States. Journal of Personality and Social Psychology, 77, 279-292.

50.

Wilson

Musick

(1997). Who cares? Toward an integrated theory of volunteer work. American Sociological Review, 62, 694-713.

51.

Yamaguchi

(1994). Empirical evidence on collectivism among the Japanese. In Kim

Triandis

H. C.

Kagitcibasi

Choi

S.- C.

Yoon

(Eds.), Individualism and collectivism: Theory, method, and applications (pp. 175-188). Newbury Park, CA: SAGE.

52.

Yamawaki

Parrot

G. W.

Spackman

M. P.

(2011). Responsibility and shame: A cross-cultural comparison of American and Japanese experiences of personal and vicarious shame. Manuscript in preparation, Brigham Young University.