Culture and Alcohol Use: Historical and Sociocultural Themes From 75 Years of Alcohol Research

Abstract

Objective:

For the period of almost 75 years, we examined the literature for studies regarding the influences of culture on alcohol use and misuse.

Method:

This review is a chronology of research articles published from 1940 to 2013. From a structured literature search with select criteria, 38 articles were identified and 34 reviewed.

Results:

This analysis revealed a progression across this period of research from studies that began as descriptive ethnographic evaluations of one or more indigenous societies or cultural groups, evolving to studies using complex multivariate models to test cross-cultural effects in two or more cultural groups. Major findings across this period include the assertions that (a) a function of alcohol use may be to reduce anxiety, (b) certain cultural groups possess features of alcohol use that are not associated with negative consequences, (c) the disruptive effects of acculturative change and the stressors of new demands are associated with an increase in alcohol consumption, (d) cultural groups shape expectations about the effects of alcohol use and their definition of drunkenness, and (e) the hypothesized relationships of culture with alcohol use and misuse have been demonstrated in multivariate model analyses.

Conclusions:

Across this 75-year period, the early proposition that culture is an important and prominent correlate of alcohol use and misuse has persisted. Within the current era of alcohol studies, this proposition has been supported by multivariate model analyses. Thus, the proposition that culture might affect alcohol use remains prominent and is as relevant today as it was when it was first proposed nearly 75 years ago.

Throughout history, alcohol has influenced human thought and behavior (Horton, 1943; Jellinek, 1977); conversely, “problems related to the use of alcoholic beverages have plagued man from the beginning of recorded history” (Bacon, 1951, p. 446). Accordingly, cultural groups and entire societies have created various norms for defining the appropriate use of alcohol (Schmidt et al., 1999). In a quest for understanding relationships between the world views of diverse cultural groups and alcohol use and misuse, we created a chronology of alcohol-related articles from each of four research eras, spanning a period of almost 75 years.

During the early era of alcohol research in the United States, around the 1940s, the Yale Laboratory of Physiology (New Haven, CT)—which later evolved into the Yale, then Rutgers, Center of Alcohol Studies—became the center point for early scientific and sociocultural research on alcohol. Selden D. Bacon, the director of this center from 1950 though the mid-1970s, wrote an analytic review article (Bacon, 1951) that set the tone for attaining a scientific understanding of the many factors that influence alcohol use within American and other societies. The pressing need for this research was heralded in Bacon’s assertion that, “Nobody knows who uses alcoholic beverages, how many, how often, how much, where, with whom, under what conditions” (Bacon, 1951, p. 447). The center thus initiated three projects that generated early hypotheses and inspired several future studies. These three studies examined (a) alcohol use behavior and attitudes among college students, (b) alcohol use among Italians and Italian Americans (Lolli et al., 1952), and (c) alcohol use within Jewish cultures (Bacon, 1951; Snyder, 1955). The latter two lines of research surmised the presence of cultural features in these ethnic/religious groups that facilitated alcohol consumption without developing alcohol-related problems.

Generally, culture consists of a schema of beliefs, practices, values, and other worldviews maintained by a cultural group or society because these contribute to the group’s well-being and survival (Shiraev and Levy, 2010). As embodied by traditions and customs, this cultural schema is preserved and transmitted across generations. It has been noted that cultures “are human-made responses to the ecology within which societies exit; cultures are created as societies adapt to their contexts in order to meet the biological and social necessities of survival” (Matsumoto and Yoo, 2006, p. 240).

Research on culture and alcohol use can be organized in parallel with four developmental phases in cross-cultural psychology (Matsumoto and Yoo, 2006). These phases are as follows: Phase I—cross-cultural comparisons—the assessment of broad cultural differences between two distinct groups or nations; Phase II—an examination of the major dimensions of culture—the core cultural dimensions such as individualism versus collectivism, that allow standardized comparisons of cultural differences on these dimensions; Phase III—cultural studies—”rich descriptions of complex theoretical models of culture and self that predict and explain cultural differences” (Matsumoto and Yoo, 2006, p. 236). A fourth emerging phase is Phase IV—link studies—studies that “empirically link [using statistical model analyses] the observed differences in means or correlations among variables with the specific cultural sources that are hypothesized to account for these differences” (p. 236). This progression reflects advances in conceptual and methodological rigor that evolved from the broad between-group comparisons as conducted in Phase I, to the Phase IV studies that examine complex multivariate models.

In parallel with this progression, associations between culture and alcohol use can be examined within two domains: (a) international studies, which examine cultural effects across nations, and (b) domestic studies, which examine cultural effects among racial/ethnic or religious groups within a nation. Furthermore, studies within each domain can examine the cultural influences of (a) external “environmental cultural factors” including social policies, community norms, and cultural products in the environment such as dwellings, artwork and music; and (b) internal “subjective culture factors” (Triandis, 1972), which as related to alcohol use include beliefs about alcohol use, attitudes toward alcohol use, subjective norms (the person’s beliefs about what others think about alcohol use), and expectancies, anticipated effects resulting from the consumption of alcohol.

Method

We conducted literature searches using the EBSCO Host database, using the search term studies on alcohol as the source (the publication source). We then added a second search term to pursue one of two major approaches for identifying articles on culture and alcohol use. In the first approach to the source term studies on alcohol, we added the key word cross-cultural, which identified international studies. In a second approach to the source term studies on alcohol, we added the key word culture, which identified domestic cultural studies conducted primarily within the United States and with racial/ethnic and religious groups.

The studies generated by these searches were also screened for relevance to our proposed analyses. Studies of interest were selected with the following inclusionary criteria: (a) articles that explicitly mentioned “culture” or “cultural factors” as a central constructs when used as a predictor, moderator, mediator, or outcome variable, and not solely as a control variable; and (b) articles that used culture or a cultural factor as an independent (grouping) variable in the study design or literature review.

We then stratified the selected studies into one of four research eras. These are (a) the early era (1940–1959), (b) the early-middle era (1960–1974), (c) the late-middle era (1975–1989), and (d) the contemporary era (1990–2013).

We thus reviewed the selected articles for findings, thematic content, and conclusions that reflected each era’s prevailing approach to alcohol research. These sampling procedures identified a total of 38 eligible articles, from which we selected and were able to obtain 34 articles (Table 1). We also identified other relevant research articles to provide greater depth of analysis for understanding the historical, temporal, social, and other contexts linking culture with alcohol use.

Table 1.

Chronology of major studies that examined culture and alcohol use (1940–2013)

I. The First, “Early Era” (1940–1959)
	1.	2.	3.	4.	5
	a. Investigatorsb. Location	a. Aimsb. Sample	*Study Design or Method	* Sociocultural Factors Related to Alcohol Use	*Major Findings
1	a. Shalloo (1941)b. United States	a. To analyze important factors in the etiology of alcoholism, asking the question, “Precisely what is alcoholism?”b. Not applicable	*An analytic review	*“Culture” as a broad construct	*“Culture” as a whole is a most important factor in the etiology of alcoholism, as compared with marital status, occupation, or nationality.
2	a. Horton (1943)b. “Primitive” (indigenous) societies worldwide	a. To identify and describe the functions of alcohol use in diverse societiesb. 56 indigenous societies	*An analytic review	*Anxiety (in several forms), gender, religious orthodoxy, acculturation, subsistence insecurity, alcoholic aggression, sorcery (beliefs in super-natural forces)	*Hypothesized that a major motive of alcohol use is to reduce various forms of anxiety
3	a. Riley and Marden (1947)b. United States	a. To conduct an analysis of data from a survey by the National Opinion Research Center. This survey was described as a first database of patterns of alcohol consumption in the United States.b. U.S. residents ages 15and older	*Details of survey methodology were not specified.*Group comparisons by levels of specific socio-cultural variables	*Culture, race, gender, nationality, religion, economic and educational status, urbanicity of the community	*Alcohol use rates were greater in men versus women, urban versus rural, higher educated and higher income respondents versus those with lower education and income.
4	a. Devereux (1948)b. Mojave Indian tribe in California	a. Study of the dynamics of alcoholism among Mojave Indiansb. 13 cases of Mojave community residents	*A psychoanalytic case analysis of incidents involving alcohol use and misuse among Mojave men and women	*Culture of the Mojave Indian community as a whole	*Mojave Indians differ from other tribes in their absence of anxiety and aggression on consuming alcohol.
5	a. Bacon (1951)b. Not applicable	a. To conduct a critical analysis of prior studies of Jewish culture and alcohol useb. Not applicable	*An analytic essay and critique	*Jewish culture as a whole	*Prior studies had drawn inaccurate conclusions about the influences of Jewish culture on alcohol use based on the use of weak research methods.
6	a.Lolli et al. (1952)b. Rome, Italy, and New Haven, CT	a. To examine the eating and drinking practices of Italians and Italian Americans, to understand the physiological and psychological aspects of alcohol useb. 26 Italians from Rome, Italy and 26 Italian Americans from New Haven, CT	*A cross-cultural comparative group study and case analysis	*Alcohol consumption during meals as compared with its consumption in other situations	*Lower rates of alcohol abuse and dependence when alcohol is consumed within the context of meals and a family setting versus other contexts
7	a. Jackson and Connor (1953)b. Skid Row, Seattle, WA	a. To examine features of the Skid Road culture for insights into the motivations and behaviors of this type of alcoholicb. Residents of the Skid Road community	*An ethnographic study of the subcultural groups from the Skid Road community	*Skid Road has several alcoholic subcultural groups emerging from features of members’ personalities and affiliations.	*There are several distinct types of alcoholics within Skid Road, and not all are solitary drinkers, social isolates, or irretrievable derelicts. Understanding these diverse types and their lifeways can be useful for incorporating them into relevant treatment and rehabilitation programs.
8	a. Snyder (1955)b. New Haven, CT	a. To study the role of religious involvement in alcohol useb. Jewish males	*A comparative study of religious orientation and alcohol use	*Orthodox and other conservative beliefs as protective against alcohol problems	*Most conservative religious orientation is associated with lowest levels of alcohol intoxication.
9	a. Berreman (1956)b. Alaska	a. To study the role of emotional expressiveness after alcohol useb. Residents of a small Aleutian village in Alaska	*An ethnographic study	*The Aleut culture as a whole	*Aleut express little emotion when sober, but much emotion after alcohol use.
10	a. Sayers (1956)b. Three rural farming villages in Southern Colombia	a. To examine ritual drinking and nonritual drinking.b. Residents of these three communities	*An ethnographic study of 3 communities	*Community cultures as a whole as general influences on alcohol use	*Ritual drinking was protective against alcohol misuse. Support for the general hypothesis that a community’s higher level of anxiety is associated with greater alcohol use
11	a. Mangin (1957)b. Vicos, a small indigenous community in the Peru Andes mountain region	a. To describe the permissive cultural norms and customs favoring alcohol consumption in the indigenous community of Vicosb. Community residents from Vicos	*An ethnographic study of drinking customs in Vicos	*Overall community culture as an influence on alcohol use	*This community has integrated alcohol use into many daily activities and especially in fiestas and local celebrations. The community takes a permissive attitude toward alcoholuse and drunkenness.
12	a. Lemert (1958)b. British Columbia, Canada	a. To describe core cultural aspects of alcohol use among the Salish Indian tribesb. Residents from 3 Salish Indian communities	*An ethnographic study of 3 Salish communities	*The influence of community norms and values on alcohol use and expressive behaviors, both prosocial and antisocial	*Alcohol consumption is normative and creates problems. Nonetheless, alcohol consumption is well integrated into many aspects of community life, and community residents take an accepting attitude toward these problems.
II. The Second, “Early Middle Era” (1960–1974)
	1	2	3	4	5
	a. Investigatorsb. Location	a. Aimsb. Sample	*Study Design or Method	*Sociocultural Factors Related to Alcohol Use	*Major Findings
1	a. Bacon et al. (1965)b. Preliterate societies around the world	a. To determine if there are relations between societal alcohol use (3 measures) and cultural variables concerned with interpersonal dependency and achievementb. 110 preliterate societies from around the world (subset used in quantitative analyses)	*Correlational design in which society was the unit of analysis; cultural variables correlated with 3 alcohol use variables (ceremonial drinking, general consumption, frequency of drunkenness)	*Indulgence of dependence, and pressures toward achievement	*Indulgence of dependence was negatively related to frequency of drunkenness and general alcohol consumption; pressures toward achievement were positively related to frequency of drunkenness.
2	a. Child et al. (1965)b. Preliterate societies around the world	a. To determine gender differences in alcohol use across many societies; how widespread is male-dominated alcohol use?b. 139 preliterate societies from around the world (although not all could be included in ratings)	*Societies classified as: “men drink more,” “women drink more,” and “men and women drink equally”	*Gender differences	*Gender differences in 53 societies (men drank more than women), no clear evidence of gender differences in 36 societies. Gender differences found for those societies that used alcohol in an aboriginal manner
3	a. Simmons (1968)b. Lunahuana, Peru	a. To study the sociocultural context of alcohol useb. The community of Lunahuana, Peru	*Anthropological study of the village and its outlying barrios	*Comprehensive examination of all aspects of community life that included alcohol use	*Numerous celebrations require consumption of alcohol, strong sense of obligation to consume alcohol in social situations; adult male-dominated consumption, belief that drinking is important throughout life and is healthful when it is not excessive.
4	a. Adler and Goleman (1969)b. Cultures around the world	a. To determine the functional equivalence of gambling and alcohol misuse (symptom substitution). Societies in which gambling is prevalent will have low rates of alcohol abuse and dependence, and vice versa.b. 117 societies	*Correlational design in which society was the unit of analysis. Prevalence of gambling and alcohol abuse was determined for each society.	*Gambling should be inversely related to alcohol abuse because gambling should satisfy the same psychological needs that are met by misusing alcohol.	*The statistical analysis supported the hypothesis that gambling and alcohol misuse were inversely related at the societal level of analysis.
5	a. Madsen and Madsen (1969)b. San Francisco Tecospa and Tepepan, Mexico	a. To explain Mexican drinking behavior through an understanding of assumptions regarding “identity, community, and prestige”b. 2 Mexican villages	*Anthropological study of 2 villages	*Cultural assumptions concerned with identity, community, and prestige	*Drinking strengthens social relationships, Indian identity (Indian village of San Francisco Tecospa), drinking associated with manliness, prestige lost with severe intoxication (mestizo village of Tepepan).
6	a. Negrete(1973)b. Montreal (Canada)	a. To determine if there is a relation between cultural affiliation and the psychosocial adjustment of alcoholicsb. 91 alcoholics: 34 Anglo-Protestants, 28 Anglo-Catholics, and 29 Franco-Catholics	*Between-group comparisons of Anglo-Protestants, Anglo-Catholics, and Franco-Catholics on indicators of psychosocial adjustment	*Ethnic/religious identity and affiliation	*Anglo cultural affiliation was associated with greater marital disruption and more arrests compared with Franco affiliation.
7	a. Park and Whitehead (1973)b. U.S. (Massachusetts) and Finland	a. To compare American and Finnish alcoholics on developmental sequencing of alcoholism; compare both sequences to the progression established by Jellinekb. 806 male Finnish alcoholics, 148 male U.S. alcoholics	*Nationality group comparisons on the chronological ordering of alcohol abuse and dependence experiences; comparisons to Jellinek’s sequencing	*Nationality (Finland and United States)	*Chronological ordering of alcohol adverse events correlated .70 for United States and Finns, .74 for United States and Jellinek, and .55 for Finns and Jellinek.
8	a. Bacon (1974)b. Preliterate societies around the world	a. To determine if there is a relation between drunkenness in a society and that society’s restrictions on interpersonal dependencyb. 53 preliterate societies	*Correlational design in which society was the unit of analysis; a composite measure of dependency restriction and achievement emphasis was correlated with frequency of drunkenness.	*A society’s restrictions on interpersonal dependency and promotion of achievement during adolescent development	*The composite variable of restricted dependency and achievement promotion was correlated r = .45 with societal measures of drunkenness frequency.
III.	The Third, “Late Middle Era” (1975-1989)
	1	2	3	4	5
	a. Investigatorsb. Location	a. Aimsb. Sample	* Study Design or Method	* Sociocultural Factors Related to Alcohol Use	* Major Findings
1	a.Paredes (1975)b. Pre-Columbian Mexico; the Aztec civilization	a. To identify cultural factors related to alcohol use and dependenceb. The Aztec civilization	*A historical analysis of alcohol use, and sociocultural controls over alcohol use	*Sociocultural factors in the control of alcohol use and misuse	*A narrow set of permissible practices for the celebratory use of alcohol (practices dictated by religious leaders); these also define harsh punishments for unsanctioned intoxication, thus establishing a relatively strict control over alcohol abuse.
2	a. Jellinek (1977)b. United States, Babylonia, Mexico, Island of Java (Indonesia), Greece, British Isles, Germany	a. To examine cultural and historic symbolic connections between alcohol and “life-giving” substances (water, milk, and blood). Discussed alcohol consumption as a positive part of cultureb. Not applicable	*An analytic review	*Symbolism of alcohol consumption and fertility, strength, religion*Social relations are eased by drinking (e.g., bonding).	*Within many cultures, alcohol serves a utilitarian function (bonding between different groups, easing social interaction), and a symbolic function (use in religious rituals representing other substances such as blood, while also representing induction into adulthood).*As a result of the positive cultural connotation of alcohol consumption, culture generally fails to illustrate the detrimental effects of excessive alcohol use.
3	a. Johnson and Matre (1978)b. Houston, TX	a. Drinking patterns in Mexican American and Anglo neighborhoods. Studied anomie and alcohol use within Mexican American and Anglo neighborhoodsb. 109 adults from a Mexican American neighborhood and 73 adults from an Anglo American neighborhood	*A culture group comparison study	*Mexican Americans were more likely to agree that a “real man” is one that can “hold his liquor.”*More Mexican American men than Anglo American men expressed approval for getting drunk at a party or excessive drinking at home.*More Mexican American women than Anglo American women drank at a friend’s request and considered social situations as proper for alcohol consumption.	*No relationship between anomie and alcohol consumption among Mexican Americans*Anomie distinguished light drinkers from heavy drinkers, but not between drinkers and nondrinkers.*Mexican Americans have standards that are lenient toward men’s drinking yet are also very restrictive for women.*Anglo Americans had uniform standards for alcohol use between men and women.
4	a. Stratton et al. (1978)b. Oklahoma	a. To use public records from the State of Oklahoma, regarding alcohol-related consequences(cause of death, arrests) to evaluate tribal variations in alcohol problems	*Between-region comparisons. Eastern tribes ad traditions of farming, business; Western tribes had nomadic hunter, horse-man traditions.	*Tribal affiliations; historical lifestyle traditions associated with hunting as a contemporary influence on problem drinking	*Western tribal regions had more alcohol-related deaths than Eastern tribal regions.
5	a. Kandel and Sudit (1982)b. Israel and the United States	a. To compare Israeli and American alcohol use behaviorsb. 464 Israeli urban adults; U.S. national survey data on drug use were used for comparisons.	*Between-country comparisons using epidemiological data	*Nationality (Israel vs. United States)	*Urban-dwelling Israeli adults had a lower prevalence of alcohol consumption and drank less often and in smaller quantities than did Americans.
6	a.Weisner et al. (1984)b. United States	a. To examine drinking levels and styles in an urban American Indian population. Examined differences among abstainers, moderate drinkers, and heavy drinkers in American Indians living in Los Angeles, CAb. N = 155	*A group comparative study	*Siouan-speaking Navajo, tribes of eastern Oklahoma origin and indigenous California tribes were included as in this study.	*Heavy drinkers were more likely to have had heavy-drinking models in the family of origin, to be men, and to score high on psycho-physiological measures of stress.*Socioeconomic status and traditionalism found to be weaker predictors of drinking level
7	a. Christiansen and Teahan (1987)b. Dublin (Ireland), Detroit, MI	a. To compare Irish and American youth on drinking behavior and alcohol expectanciesb. 168 Irish adolescents, 168 American adolescents, with samples matched on age and gender	*Between-sample (“nation”) comparisons	*Nation of residence	*Irish adolescents reported less social and problematic drinking than American adolescents.*Irish adolescents had lower alcohol-related expectancies for social, cognitive/motor, and sexual benefits than did American adolescents.
8	a. Caetano and Medina Mora (1988)b. United States (nationwide) and Mexican state of Michoacán	a. To determine the relation between acculturation and alcohol use; to compare U.S. and Mexican sample on alcohol useb. 949 Mexican-heritage adults in the United States and 1,191 Mexicans living in Michoacán	*Household surveys between-nation comparisons on quantity-frequency of alcohol use; within-U.S. analyses of acculturation’s relation to quantity-frequency of alcohol use	*Nation of residence, level of acculturation	*Mexican American men and women drink more than men and women in Michoacán, Mexico.*In logistic regression analyses, greater acculturation was related to heavy drinking for Mexican American women but not for Mexican American men.*Acculturation has a stronger relation to drinking for Mexican American women than it has for Mexican American men.
IV. The Fourth, “Contemporary Era” (1990–2013)
	1	2	3	4	5
	a. Investigatorsb. Location	a. Aimsb. Sample	*Study Design or Method	* Sociocultural Factors Related to Alcohol Use	*Major Findings
1	a. Schmidt et al. (1999)b. 9 distinct national/cultural groups: Greece, India, South Korea, Mexico, Nigeria, Romania, Spain, Turkey and United States	a. To assess the applicability of international classifications of alcohol dependence criteria. Compared data on the meaning of alcohol dependence in 9 different countriesb. Purposive and quota sampling were used to obtain 20 key informants (actual range from 18 to 25) per each of the 9 countries.	*A cross-cultural study using in-depth individual interviews with key informants*A qualitative analysis, to conduct a cross-cultural comparison of the meaning of alcohol dependence using 4 alcohol dependence criteria	*The local culture and its interpretations of alcohol and drug dependence criteria and related concepts	*Cultural differences exist across “drinking cultures” in the conceptualization and meaning of alcohol problems and dependence.*Descriptions of dependence symptoms were similar regarding symptom severity from sites that shared norms around drinking and drunkenness.*Descriptions of physical alcohol dependence criteria varied across sites as much as those of the more subjective psychological dependence symptoms.*Future research should consider cultural variations found in the symptom levels of alcohol dependence and withdrawal descriptions as found in this study.
2	a. Lindman et al. (2000)b. Belgium, Finland, France, Italy, Poland, Panama, Spain, and United States	a. To compare expectations from alcohol-induced positive affect across national culturesb. A total sample of N = 1,008 university students (Belgium n = 108, Finland n = 106, France n = 98, Italy n = 83, Panama n = 159, Poland n = 121, Spain n = 207, and United States n = 126)	*Conducted a multigroup latent covariance structural model analysis*Cross-cultural comparative self-report survey in 8 countries*Compared each national/cultural group on “alcohol-induced positive affect,” and on other model parameters, to examine measurement equivalence and cross-cultural differences	*Key sociocultural variables included: feelings, behaviors, interpersonal processes and moral evaluations as related to alcohol consumption, and to acts committed during an intoxicated state.	*U.S. respondents had significantly more positive expectancies about drinking than those reported in most other countries.*Subjects from Spain and Panama held the least alcohol-induced positive expectancies, and these were significantly different from those reported in all other countries.*Multigroup latent covariance structure analysis was useful for addressing critical problems in comparative cross-cultural research.*The expectation of greater positive affect as associated with drinking may be moderated by contextual factors and by cultural traditions, such that positive affect is less likely attributable to the direct pharmacological effects of alcohol than previously believed.
3	a. Venner and Miller (2001)b. United States	a. To study the progression of alcohol problems in a Navajo sample. Tested the cross-cultural applicability of E. M. Jellinek’s progression of alcohol problems in a Navajo sampleb. N = 99	*A correlational analysis to examine model fit	*Jellinek s progression of alcohol problems is used widely in treatment, but it is unclear whether it applies to other cultures. This study specifically studied whether it applies to a Navajo culture.	*”Of the 46 events established by Jellinek, the order of progression for the Navajo sample was modestly correlated with that for Jellinek’s white men.”*The correlation of Jellinek’s sample with the Navajo men was slightly larger than that of the whole sample.*The comparison with Navajo women resulted in a near zero correlation.*Jellinek’s model might not be applicable to Navajo culture.
4	a. Schmid et al. (2003)b. A total of 22 countries: France, Greece, Germany, Belgium, Austria, Switzerland, Poland, Czech Republic, Slovak Republic, Hungary, Russia, Estonia, Latvia, Lithuania, Canada, United States, Ireland, Greenland, Finland, Sweden, Norway and Denmark	a. To study number of occasions of drunkenness among 15-year-old students in 22 countries. Examined the association between drunkenness and the frequency of alcohol consumption, and in preferences for alcoholic beverages. Also examined the effects of country characteristics and drunkennessb. This study used a sub-sample from a larger study; for this sample: (n = 10,951) males and (n = 11,451) females who consume alcohol, ages 15 and older in 22 countries, including the United States.	*A hierarchical generalized linear model analysis to examine cross-national similarities and differences in drunkenness	*Differences between countries in occasions of adolescent drunkenness*The distribution of adolescent beverage preference for beer, wine, and spirits varied among different countries.*The link between a specific beverage preference among adolescents and drunkenness was independent of the frequency of alcohol consumption.	*Beer was the preferred beverage choice for males in every country; females preferred wine in 12 of 22 countries.*In Greenland, Denmark, Hungary, and Canada 1/3 of the alcohol intake was attributed to spirits.*The association between drunkenness and the consumption of spirits was positive.*Lower numbers of drunkenness occasions were found among southern European countries.*Higher occasions of drunkenness were found for Scandinavia, the Baltic countries, and Russia.*Geographic location is an important country-level association for drunkenness and its predictors.*There is a high prevalence of drinking among young people in many countries, and preventive action should be taken by creating interventions to change this behavior.
5	a. Venner and Feldstein (2006)b. United States	a. Alcohol dependence and remission events for a Native American sample. Construct a combined sequence of both problem emergence and recovery efforts with Native Americans.b. N = 44	*A correlational analysis to examine model fit	*The sequences of alcohol-related behavior across gender and culture*Sequences of change efforts across gender and culture	*“There was greater concordance between this intertribal sample and Jellinek’s white male sample than between this sample and a Mission Indian sample.” The findings reveal “both cross-cultural and intra-cultural variation.”
6	a. Link (2008)b. Germany and United States	a. Cross-cultural study of drinking among German and American adolescents comparing differences in the national/cultural groups, taking into account protective and risk factorsb. German youth n = 11,043, American youth n = 16,300	*Conducted a series of negative binomial regression model analyses to test cross-cultural models of counts data involving the frequency of drunkenness	*Cross-national differences in legal age for drinking alcohol, in perceived availability of alcohol, and in perceived risks from alcohol consumption	*American adolescents were more likely than German adolescents to be influenced by deviant peers.*Both German and American adolescents who consumed alcohol had similar frequency of getting drunk, despite the differences in laws that purport to protect adolescents from alcohol consumption in the United States.

Results

Studies and findings from the first, “early era”: 1940–1959

During the early era of alcohol research (1940–1959), considerable interest emerged regarding the role of culture in the etiology of mental disorders and personality problems, and including the etiology of alcoholism (Shalloo, 1941). During this era, several descriptive ethnographic studies were conducted. One notable exception is the study by Riley and Marden (1947) that reported results from a national survey of alcohol use among American male and female respondents ages 15 and older. As an overview of alcohol use by Americans in the late 1940s, this survey revealed that a greater proportion of males (75%) relative to females (56%) consumed alcohol, with a greater proportion of urban residents (77%) doing so relative to rural residents (46%). Rates of alcohol consumption increased with greater levels of education and socioeconomic status. Also, changes in gender roles prompted by the influences of World War II on the American workforce revealed increases in alcohol consumption, with greater increases occurring among women than among men. Regarding abstinence from alcohol in relation to religious identification, Protestants exhibited the highest rate of abstinence (41%), with a lower rate among Catholics (21%), and the lowest rate among Jews (13%).

Shalloo (1941) examined major factors in the etiology of alcoholism, asking the question, “Precisely what is alcoholism, and when may a person be said to be an alcoholic from the cultural standpoint?” (p. 464). He accorded a greater importance to “cultural learning” as a contributor to alcoholism relative to several demographic variables, concluding that alcoholism is the end product of a complex process of social and cultural learning. In a comprehensive analytic review to identify the functions of alcohol use within diverse societies, Horton (1943) examined the effects of psychological factors (conceived to be universal) and cultural factors (conceived to be culturally specific) for several “primitive societies” (i.e., indigenous cultural groups). Based on ethnographic evidence from 56 of these societies, Horton hypothesized that a major function of alcohol use was to reduce various types of anxiety.

Studies of special populations.

A series of studies of special populations were conducted to examine remarkable cultural factors that might promote or diminish the adverse consequences of alcohol consumption. Toward this aim, Bacon (1951) wrote an analytic essay and critique that lauded the core attitudes and alcohol use behaviors of the Jewish culture, whose members possessed a world view that made this culture “dramatically superior to the majority of patterns in our [American] society” (Bacon, 1951, p. 449). Shortly thereafter, Snyder (1955) examined religious orthodoxy and sobriety among Jewish adult men and college students from New Haven to assess the working hypothesis that participation in traditional Jewish rituals and related obligations to the group would be associated with moderate alcohol consumption and sobriety.

Similarly, a small, cross-national study that compared 26 Italians in Rome with 26 Italian Americans in New Haven, CT, concluded that alcohol consumption within the context of meals and family was protective against the misuse of alcohol (Lolli et al., 1952). Among all 52 Italian and Italian American study participants, despite initiating their wine consumption during their youth, “no evidence was discovered that wine had affected their physical condition or their emotional equilibrium unfavorably” (Lolli et al., 1952, p. 46).

Another ethnographic study, of alcoholics from Skid Road in Seattle, WA, examined relationship patterns for distinct subgroups of alcoholics (Jackson and Connor, 1953). A Skid Road culture existed wherein distinct subgroups of alcoholics created their own subgroup norms regarding group membership, which centered around drinking and sharing alcohol. These investigators concluded that “the Skid Road culture contributes to the persistence of alcoholism and raises barriers to successful rehabilitation” (Jackson and Connor, 1953, p. 485).

Studies of indigenous societies.

During this early era, a series of ethnographic studies also examined the alcohol use practices of indigenous societies. From a psychoanalytic framework, Devereux (1948) examined the Mohave Indian society, which had cultural norms that discouraged fighting and encouraged emotional self-control and restraint. These cultural values were associated with low rates of aggressive or violent behavior, even when under the influence of alcohol. This investigator noted the remarkable absence of anxiety and aggression among the Mohave Indians even when consuming alcohol, concluding that unlike some other Indian tribes, the Mohave Indians were able to evade the adverse consequences of alcohol use.

Another ethnographic study (Berreman, 1956) examined the emotional expressiveness of native residents of a small remote Aleutian Indian community in Alaska where “everybody gets drunk” (p. 506), and where “drinking is done with the intent of becoming drunk” (p. 507). Nonetheless, within this village, alcohol dependence did not appear to be a problem because drunkenness was regarded as an opportunity to release suppressed tensions. Although villagers often expressed little emotion, when intoxicated with alcohol they became emotionally expressive, indulging in behaviors otherwise regarded as reprehensible.

Another study compared three communities in rural Colombia that differed in their patterns of ritual alcohol consumption (Sayers, 1956), hypothesizing that alcohol consumption would be associated with greater levels of community-level anxiety. This three-community study compared the most indigenous reservation-dwelling Coconuco Indians with the more acculturated and Westernized pueblo-dwelling Coconuco mestizos (mixed ancestry) and with the partially acculturated Zarzal Indians who lived within a community in transition. Residents from these two latter communities did less ritual drinking (i.e., drinking small quantities at cultural ceremonies) and more often drank to relieve distress, followed by more episodes of drunkenness. In those two communities, the stress of acculturative transition from traditional to Westernized modernistic life ways was implicated as a source of greater alcohol use.

Ceremonial and convivial forms of drinking were also studied among the residents of Vicos, a small Andean mountain village (Mangin, 1957), where alcohol consumption was well integrated into daily life. This community exhibited permissive norms toward alcohol use. When coupled with their casual social and occupational roles, drunkenness neither created problems nor interfered with their social roles and obligations. The Salish Indian tribes also exhibited permissive drinking norms, although tribal members acknowledged the problems created by drunkenness. Nonetheless, they expressed a “fatalistic acceptance” of the negative consequences of drunkenness, “with little thought given to the matter” (p. 102). This and other studies of indigenous societies also alluded to the destructive effects of colonialization and its imposed Westernized cultural change that potentially contributed to cultural confusion and erratic patterns of alcohol consumption. Others have described this process and its creation of historical loss or trauma (Whitbeck et al., 2004), an effect that may have persisted for generations.

Summary of the early era: 1940–1959.

Studies and analytic reviews from this first early era (1940–1959) examined aspects of alcohol use and misuse in rich, descriptive ethnographic studies. From these, one emerging theme was that alcohol use, whether ritualized or convivial, functioned to reduce anxiety and to release inhibitions, thus examining variations of Horton’s (1943) hypothesis. A second emerging theme involved the social integration of alcohol use in the form of alcohol-use customs or traditions, including permissive attitudes toward drinking. A third and related theme involved the stress of acculturative change among indigenous cultural groups, which was associated with heavier and erratic alcohol consumption. These indigenous societies recognized that drunkenness could be disruptive to family and community yet often did not define drunkenness as malicious or as antisocial conduct.

Studies and findings from the second, early-middle era (1960–1974)

The second era continued the analysis of formative hypotheses on the function of alcohol use among diverse indigenous societies or cultures. In these studies, the entire society was typically used as the unit of analysis (Bacon et al., 1965). In 1965, a special supplement to the Quarterly Journal of Studies on Alcohol introduced a series of articles based on a study of 139 preliterate societies (Supplement No. 3, A Cross-Cultural Study of Drinking). One of the articles evaluated the dependency-conflict hypothesis, which proposed that high levels of alcohol use and drunkenness are associated with societies that restrict interpersonal dependency and emphasize achievement and independence (Bacon et al., 1965). Investigators found good support for this hypothesis, support that became even stronger when data were reanalyzed several years later (Bacon, 1974) with more sensitive quantitative methods.

Another article in this special supplement explored gender differences in alcohol use (Child et al., 1965) and reported greater use by men than by women in 53 societies, no differences in 36 societies, and indeterminate gender differences in 50 societies (primarily because of insufficient information). Gender differences were observed in societies that practiced “aboriginal drinking.” Society was the unit of analysis in a study that examined the inverse relation of alcohol use and gambling in 117 societies (Adler and Goleman, 1969). This study was partially grounded in psychoanalytic theory, arguing that gambling and alcohol use met the same psychodevelopmental needs, and thus they should compensate for each other, such that societies in which gambling was prevalent should exhibit low rates of alcohol abuse and dependence and vice versa. That hypothesis was largely confirmed.

In-depth studies of alcohol use in village cultures were conducted in Mexico (Madsen and Madsen, 1969) and Peru (Simmons, 1968), and both studies reported that alcohol consumption appeared to be used to build social solidarity and a sense of community, particularly among men from these traditional cultures. Drinking to build and confirm social bonds also came with a sense of obligation, because declining invitations to drink was perceived as rejection. Madsen and Madsen (1969) conducted a study of an acculturating village where harmonious alcohol consumption was associated with displays of machismo, an exaggerated manliness involving competition and aggressiveness. A few years earlier, Madsen (1964) wrote about Mexican-heritage men living near the Texas–Mexico border and how marginalization—a rejection of their own Mexican cultural identity, coupled with their failure to be accepted by Anglo Americans—could lead such persons, the “agringados” (overly Americanized individuals), to experience distress and shame and turn to alcohol to cope with this sense of failure and marginalization. This perspective echoes an earlier account of indigenous villagers who struggled with conflicts involved in “straddling two cultures” (Berreman, 1956, p. 509). The central point of Madsen’s study was captured in the quote, “Caught between two culture worlds and accepted by neither, the agringado frequently seeks to capture what he has lost and what he has failed to gain in the twilight zone of intoxication” (p. 357).

Advancing beyond simple group comparisons, an emerging methodological approach examined cultural groups’ fit to a hypothetical model. For example, Park and Whitehead (1973) evaluated how well Americans and Finns fit Jellinek’s model of progressive stages in the development of alcoholism. In his research, Jellinek (1946, 1952) identified 46 events in the chronological development of alcohol problems. Park and Whitehead correlated the chronological sequences found in Jellinek’s sample with those found for their samples of American and Finnish male alcoholics. They concluded that for Americans and Finns, alcoholism develops in similar progressive stages, noting also that Jellinek’s sequence exhibited a greater correspondence for American alcoholics relative to Finnish alcoholics. Notably, three decades later, Venner and Miller (2001) and Venner and Feldstein (2006) examined Jellinek’s model as applicable to Navajo Indians and found a moderate correspondence.

By the end of this period, important efforts were made to organize and summarize findings from several within-culture and between-culture studies, in a book edited by Pittman and Snyder (1962). This book contained 35 chapters written by leading anthropological and sociological alcohol researchers of the day. A reviewer of that volume (Mangin, 1964) captured the book’s meta-message—throughout time, culture has shown a stronger influence than biological factors on alcohol use and misuse. Also, a literature review by Mandelbaum (1965) was heralded by Heath (1987) as a milestone paper of that time that clearly synthesized the available literature on key issues. Mandelbaum’s review examined similarities and differences in alcohol use across highly diverse cultures and how changes in alcohol use practices might reflect broader societal changes.

Summary of the early-middle era: 1960–1974.

Articles published during this second era described in-depth group comparisons of drinking practices. Studies that treated “society” as the unit of analysis examined more than 100 different cultures. A novel methodological approach emerged in this era that compared Jellinek’s model of the temporal progression toward developing alcoholism with the developmental sequences observed for two cultural groups. Anthropological research was also prominent and at times framed alcohol research in accord with psychoanalytic theory (e.g., Adler and Goleman, 1969; Bacon, 1974). Another prominent theme regarded alcohol as a means of coping with modernistic pressures involving achievement and acculturation conflicts, as antecedents of social marginalization that resulted from being caught between two cultures.

Studies and findings from the third, late-middle era: 1975–1989

Anthropological–historical research appeared in this period. Paredes (1975) examined alcohol use within the Aztec civilization, reporting that the Aztecs established a narrow set of permissible practices for the celebratory use of alcohol (practices dictated by religious leaders). This involved ancient versions of social and injunctive norms regarding the appropriate use of alcohol (Shmulewitz et al., 2012). These norms imposed strict controls over alcohol use, coupled with harsh punishments for unsanctioned intoxication.

A unique historical analysis by Jellinek (1977) published 14 years after his death examined symbolic aspects of alcohol use. Contemporary examples of alcohol as a symbol include adolescent drinking to symbolize the transition into adulthood and the notion that “holding one’s liquor like a man” (Jellinek, 1977, p. 862) establishes the adolescent as a virile male. Jellinek’s essay offered historical accounts of alcohol’s symbolic equivalence to certain vital fluids such as blood, water, and milk, adding that social drinking has long been symbolic of solidarity and a sense of community.

Culture group comparative studies.

A study of American Indian tribes in Oklahoma (Stratton et al., 1978) attributed the higher rates of alcohol-related deaths in western counties to the culture of tribes that exercised historical roles as hunters, in contrast to the stable farmer–merchant roles of tribes that lived in eastern counties. A study of drinking patterns of American Indians from various tribes who resided in the Los Angeles area (Weisner et al., 1984) identified the strongest predictors of heavy alcohol consumption as male gender; younger age; during childhood, having a family role model who consumed alcohol; and undergoing high levels of psychological stress. In this era, other studies compared the drinking practices of American and Irish adolescents (Christianson and Teahan, 1987), Israeli and American urban adults (Kandel and Sudit, 1982), and religious groups living in Montreal (Negrete, 1973).

One comparative group study examined how drinking practices may change when individuals migrate from a native culture to reside within a new host culture (i.e., acculturative adaptation when immigrating to the United States). Caetano and Medina Mora (1988) compared Mexicans living in the state of Michoacán, Mexico, with a nationally representative sample of Mexican-heritage individuals living in the United States. In general, Mexican American men drank more frequently than men living in Michoacán, although the men in Michoacán were more likely to engage in heavy episodic drinking; that is, consuming five or more drinks on one occasion. By contrast, compared with women in Michoacán, Mexican American women were less likely to be abstinent and more likely also to engage in such heavy drinking episodes.

Religious orientation and alcohol use.

Religious orthodoxy was again identified as a factor in the relatively low consumption of alcohol by Israelis. As an extension of the influence of religious practices, Kandel and Sudit (1982) reported that urban-dwelling Israeli adults exhibited a lower prevalence of alcohol consumption and drank less often and in smaller quantities than did Americans, findings that supported the hypothesis introduced earlier by Snyder (1955) that greater religious orthodoxy among Jewish people would be associated with a lower consumption of alcohol. The effect of religious beliefs on abstinence and temperance was also examined in a study of differences between Catholics and Protestants who lived in Montreal (Negrete, 1973).

Despite variations in alcohol use sanctions even within broad religion categories (e.g., Christianity), there is general agreement that those who endorse a religious affiliation are less likely to misuse alcohol and other substances than those who have no religious affiliation (Gorsuch, 1995). There has been considerable interest in understanding the potential protective mechanisms relating to religiosity and how that understanding might inform substance-related interventions (Miller, 1998). The U.S. National Alcohol Survey (Michalak et al., 2007) identified complex interactions and differences in the predictors of (a) alcohol use versus abstinence and (b) heavy alcohol use versus moderate use. Regarding frequent heavy drinking, beyond gender (being male) and age (the youngest group), religious predictors of heavy versus moderate alcohol consumption were (a) proscription—a religion’s discouragement of alcohol use, and (b) religiosity—the importance afforded to religion. Also, considerable variations existed by denomination (e.g., Muslims and Mormons, who eschewed alcohol use, as opposed to Catholics and Jews, who accepted it). These investigators thus added that “religious variables are a powerful factor in abstention” (p. 279).

Culture as moderator.

Besides operating as an independent variable or as a predictor variable, culture has been also studied as a moderator of the effect of a precursor variable on alcohol use. As one illustration, Johnson and Matre (1978) examined the effect of anomie as a predictor of alcohol use—the outcome variable, as moderated by the variable of ethnic community (i.e., Mexican American vs. “Anglo” neighborhoods in Houston, Texas). These investigators stated that anomie occurs when a person “expresses feelings of rejection, helplessness, hopelessness, or lack of gratification” (p. 895). In this moderator model analysis, anomie was predictive of heavy alcohol use in Anglo neighborhoods (r = .47) but not in Mexican American neighborhoods (r = .15, n.s.). Thus, moderation by cultural group was indicated by the differential relationship of anomie to alcohol consumption, as observed for Anglo Americans but not for Mexican Americans. Johnson and Matre (1978) speculated that “anomie” did not capture a meaningful aspect of deviance for Mexican Americans, as it did for Anglo Americans. Similarly, in a cross-cultural study of 26 countries that examined how patterns of alcohol consumption related to negative consequences, the variable of “nation/culture” operated as a moderator of effect, whereby large differences emerged across countries regarding negative consequences resulting from alcohol consumption (Graham et al., 2011).

An emerging issue in cross-cultural group comparative studies was the need to ascertain a construct’s measurement equivalence across two distinct cultural groups as an essential prerequisite for conducting valid cross-cultural comparisons (Knight and Hill, 1998; Widaman and Reise, 1997). In such analyses, a vexing problem involves establishing when an apparently significant moderation effect constitutes evidence that (a) a measured construct does not have the same predictive validity across cultures (i.e., the cultural nonequivalence of that construct), or (b) that it truly captures meaningful cultural differences in the mechanisms that produce a particular behavioral outcome, such as alcohol abuse or dependence (i.e., a significant moderator effect; Barrera et al., 1999). Worldwide, a diversity of cultural differences have now been demonstrated regarding the concept and meaning of heavy drinking and its perceived detrimental effects (Graham et al., 2011).

Summary of the late-middle era: 1975–1989.

In contrast with studies from the previous early-middle era, research conducted within this third late-middle era shifted away from ethnographic studies of small-village indigenous societies and showed a waning emphasis on psychoanalytic theory. By contrast, an emerging methodological theme in these comparative culture group studies involved a testing of statistical models that examined the potential effects of cultural factors, either as predictors or as moderators of alcohol use or misuse. Thus, epidemiological and between-culture comparative quantitative analyses grew in prominence as researchers developed and tested new and more complex cultural models.

Studies and findings from the fourth, contemporary era: 1990–2013

In a study to improve the measurement of diagnostic criteria for alcohol dependence, Schmidt and colleagues (1999) examined relevant symptoms and criteria across nine national/cultural groups. These investigators noted that cultures potentially shape expectancies about the anticipated effects of alcohol consumption (Schmidt et al., 1999) and differ also in how they define serious drinking and drunkenness. This study revealed significant differences not only for subjective indicators of psychological dependence but surprisingly also for the more objective symptoms of physical dependence. From this study, Schmidt and colleagues proposed four criteria to explain observed cultural differences: (a) threshold of severity—cultural group differences regarding the point at which a symptom is considered serious, (b) problemization of dependence symptoms—variations in how a symptom is defined to be problematic, (c) the causal ordering and sequence of symptoms—variations in the “causal” sequence of events and attributions in describing the etiology of alcohol-related problem behaviors, and (d) culture-specific manifestations—for a particular cultural group unique interpretations regarding alcohol use symptoms.

Multivariate cross-cultural models of adolescent alcohol use.

A study of cross-cultural differences in the perceived reinforcing effect of alcohol used stacked latent variable structural equation model analyses, one for each of eight national/ cultural groups (Lindman et al., 2000). These investigators concluded that the positive reinforcing effects of alcohol are partly determined by culture and contextual variables, and that these sociocultural effects are independent of the frequency of alcohol consumption and of the pharmacological effects of alcohol itself.

In this contemporary era, cultural research began to more closely examine alcohol use in adolescent populations. Schmid and colleagues (2003) conducted a hierarchical generalized linear model analysis of cross-national similarities and differences in alcohol-related intoxication among 15-year-old students who already started consuming alcohol. These models examined differences in intercepts across structural models for each of 22 countries, which revealed significant variations across countries in the occurrence of drunkenness. Also, significant variations in slopes across these models revealed differences across countries in the strength of association between specific predictors and the occurrence of drunkenness. These analyses revealed that a higher frequency of overall alcohol consumption and greater consumption of distilled spirits were both associated with greater drunkenness. Also, for both genders, fewer occasions of drunkenness were observed for southern European countries, as contrasted with Russia, Scandinavian, and Baltic countries. Investigators interpreted these differences by geographical location as a proxy for the existence of different “drinking cultures” while demonstrating empirically that “cultural differences in alcohol use exist.” (p. 650).

In a cross-cultural study of alcohol consumption among German and American adolescents, Link (2008) examined the predictors of adolescent drunkenness. Analyses revealed that the relationship of peer alcohol use to a youth’s own frequency of drunkenness was stronger among American youths versus German youths. Similarly, the association between perceptions of risk resulting from heavy alcohol use on own frequency of drunkenness was stronger for American youths compared with German youths (Link, 2008). Link emphasized the importance of examining alcohol etiology and problem behaviors within a specific cultural context.

Summary of the contemporary era: 1990–2013.

One emerging theme from contemporary era studies involves the use of model-driven analysis of the antecedents of alcohol use and misuse within and across adolescent populations. As an extension of similar research initiated in the prior era, contemporary era studies incorporated more reliable and valid measurements of cultural constructs and examined more complex multivariate models using more advanced multivariate statistical analyses. These produced more potent confirmatory methods, which demonstrated empirically that “cultural differences in alcohol use exist” (Schmidt et al., 2003).

Discussion

Role of cultural factors in defining acceptable alcohol use

Across these four eras in the study of culture and alcohol use, an originally proposed meta theme was that culture influences alcohol use. That theme has not diminished in appeal or prominence across this span of 75 years. An extension of this perspective is that some cultural groups possess culture-based cognitive schemas and interpersonal relationship structures that permit alcohol use without a significant loss of control, whereas other cultural groups possess different psychological and relationship structures that promote a loss of control over alcohol use, leading to alcohol-related problems. For instance, the Mojave Indians were found to consume alcohol without problems or disruptions (Devereux, 1948), whereas the life activities of Skid Road alcoholics revolved almost entirely around alcohol dependence and associated struggles to meet conventional roles and obligations (Jackson and Connor, 1953). Regarding environmental influences, the transition from a casual and informal indigenous society to a regimented Westernized, modernistic society appears to have introduced stressors that promote greater alcohol consumption as one means of coping.

Enhanced methodological and statistical methods

Across this span of 75 years, studies of culture and alcohol use have transitioned from early descriptive ethnographic studies of entire societies to contemporary studies that emphasize measurement and the use of complex multivariate models. These advanced methodologies have included (a) confirmatory factor analyses to test the cultural equivalence of one or more cultural constructs (Brown and Moore, 2012; Lindman et al., 2000), (b) multigroup structural equation model analyses (Lindman et al., 2000), (c) hierarchical generalized linear model analyses (Schmid et al, 2003), and (d) negative binomial regression model analyses (Cohen et al., 2003; Link, 2008).

Conclusion

In summary, during the first era of alcohol studies almost 75 years ago, Shalloo (1941) set forth the proposition that culture is an important and prominent determinant of alcohol use and misuse. This proposition has persisted across the 75-year period in which studies of culture and alcohol use have been conducted. In today’s contemporary era, this proposition persists and has been demonstrated empirically in multivariate model analyses (Lindeman et al., 2000; Link, 2008; Schmid et al., 2003). It remains a prominent and potent proposition that is as relevant today as it was when it was first proposed nearly 75 years ago.