Are Men More Rational than Women when Purchasing Cars ? A Comparison of Cultural Influences on Product Judgment in Brazil and the United States of America

The main goal of the study was to identify and to compare the type of judgment used in Brazil and in the United States of America relative to the purchase of automobiles and the relevance of Human Values in the evaluation used. Brazilians (N=542) and North Americans (N=449) filled out the List of Values (LOV) (KAHLE; KENNEDY, 1988) and the Judgment and Meaning Scale (ALLEN, 1997, 2000). The results suggested that in Brazil the judgment for purchasing a car is predominantly affective, while in the United States it is predominantly piecemeal. In both countries women scored higher in piecemeal judgment for car purchase. The values of “Excitement” and “being well respected” are related to affective judgment in both countries while “Personal Achievement” is related to piecemeal judgment in Brazil and the United States. This research contributes to the advancement of studies in Consumer Behavior by analyzing the role of human values in the type of judgment and meaning men and women use to buy cars in Brazil and the US. Furthermore, as practical implications, the results may help company managers in the automobile industry in their decision-making processes in order to better understand how to meet client’s specific needs in order to develop a long-term and profitable relationship.


INTRODUCTION
ver the past 40 years, organizations have experienced profound changes.A greater number of international products increased rapidly and considerably to match a competitive and global market (AXINN; MATTHYSSENS, 2002).Waters (2001) defines globalization as the direct consequence of the expansion of Western culture across the planet via settlement, colonization and cultural replication of the integration of markets on a worldwide scale and it could eventually means worldwide standards or practices for product quality, pricing, service, and design.
The consumption of physical products and services is part of the modern human being's daily routine and is studied by many researchers in different areas.Hunt (1991) points out that consumer behavior is characterized by the open use of different methods and theories, originating from very different sources such as Economics (DEATON; MUELLBAUER, 1980;DUESENBERRY, 1967), Sociology (BOURDIEU;PASSERON 1979;RIESMAN 1964;RIBEIRO, 2008) and Psychology (KAHLE; KENNEDY, 1988;ALLEN;TORRES, 2006).
Most studies in the consumer behavior area show how demographic variables can segment the market, dividing groups according to age, gender, income and other demographic variables (MOWEN;MINOR, 2001).However, one aspect that was not taken into account in consumer behavior studies is cultural influence.Studies show that different cultures may present different purchasing behavior (ALLEN;TORRES, 2006, TORRES;PÉREZ-NEBRA, 2007) even if economic and social factors are controlled or disregarded; influences of culture may divide segments of the market considered equal looking through the demographic approach only.
Studies that analyzed differences in evaluation, purchase and consumer behavior in general consider gender as the grouping variable and they are usually conducted within the same country (e.g.HAAS, 1979, PUTREVU, 2001, MITCHELL;WALSH, 2004).If internationalization of products and brands is taken into consideration then a gap is noticed.
Due to the production in industry scale and market research that is often done in the country of origin, the product design doesn't take into consideration cultural aspects.Hence, differences between segments appear to be similar using demographic variables only.
The studies of Hofstede (1980;1983;1984;1991)  and thus they behave unevenly.Therefore, culture can influence consumer behavior, altering the importance or interpretation of some intangible or tangible characteristics of a product or service.As pointed by Hofstede (1991), culture is learned and not genetically inherited, so personal experience and the environment are the main contributing factors in shaping consumer behavior.
In this context, this research intends to fill the literature gap regarding the cultural influence in shaping consumer behavior by analyzing the role of human values in the type of judgment and meaning men and women use to buy cars in both Brazil and US.The automobile was chosen because is a product that requires some evaluation before being purchased by the consumer, and its characteristics usually follows the conception from the country where it was designed.This misconception can lead a successful product or brand to a failure in a specific market.Cultural elements indeed alter the way the consumer perceives the product, requiring the companies who are attempting to enter in a new market or to launch a new product to develop and communicate their values and benefits in a manner that the market understands it (ALLEN;TORRES, 2006).
To meet the goal of this study, first, a literature review about culture and the constructs human values and types of judgement that affect consumer behavior.After, the method used is described, including sampling, scales and procedures of data collection and data analysis.
Finally, the results are presented and discussed and final remarks are made, pointing the research limitations and its practical implications as well as highlighting directions for future research.

THEORETICAL BACKGROUND
This section introduces the literature review of the research constructs including culture and consumer behavior, especially regarding human values and judgment and meaning.

CULTURE
Culture can be defined anthropologically since it's the field where more complex and diverse meanings are assigned to the term.Culture of Civilization in an ethnographic broad sense is the "complex role that includes knowledge, beliefs, art, moral, laws, customs and other capabilities and skills acquired by a man as a member of a society" (TYLOR, 1871, p.1).
For Keesing (1974) there are two strands when talking about culture.The first one considers it an evolutionary process.Culture is shaped and disseminated by society based divides culture into three idealistic theories that are defined as a cognitive system, a structural system and a symbolic system.
According to Goodenoogh (1957), culture is a cognitive system that emphasizes learning and knowledge.A person needs to learn or believe in the prevailing culture in order to live in an acceptable manner within their society.The author stresses that culture is neither a material phenomenon nor formed by things, people, behavior and emotions but by the organization of all of these factors.It's the way people think, the models they follow, how they relate and how they interpret (GOODENOOGH, 1957).
In Levi-Strauss' (1986) definition culture as a structural system is defined as the accumulation of the creations of the mind.What occurs in an individual's life shapes his thoughts that form cultural patterns.On the other hand, as stated by Keesing (1974), culture as a symbolic system attaches high importance to symbols and meanings.The rules and categories are not disrupted by behavior; they exist on a separate plane.In an attempt to integrate these two visions, cognitive and symbolic, Kluckhohn (1962) divided the concept of culture into two parts: the first referring to the objective elements (for example, crafts produced by social groups) and the second, reflecting its subjective elements (such as, the values, beliefs and social norms of these groups).
The definition of culture elaborated by Geertz (1973;2009) takes in consideration several aspects that range from the complete lifestyle of a population to the sharing of how they feel, think and believe.Associated to that concept, Geertz (2009) also adds the social legacy absorbed by an individual in the groups he lives with, including the abstract behavior that is understood as a system of shared learning that leads to similar orientations do recurring problems, or a learned behavior shaped by the normative social influence.Concluding, Geertz (2009) suggests that culture expresses techniques that adjust an individual to the external environment or to live with others.
When culture is analyzed from the point of view of its multiple concepts, some theoretical differences arise naturally.Noriega, Carvajal and Grubits (2009) understand culture as the practices of meaning that in the same time affects the production, reproduction and transformation of the material and symbolic systems where people live in.The definition of culture from Dupuis (2008) explains the inseparable bond between models, values, symbols and behavior that together establish the cultural configurations of people.Dupuis (2008) arguments that the concept of culture is constituted by the interaction of elements as economy, administration, practices and social representations of a social group.
D' Iribarne (1983) defines culture as a resource or a starting point for individuals to make relationships and cooperations with others.
Torres and Allen (2009) suggest that the basic cultural values influence the way people consume by imposing limits on human behavior, implying that these values directly impact consumer behavior.Arnould, Prince and Zinkhan (2003) indicates a relationship between culture and consumption, arguing that cultural consistency is also maintained in the consumption of products that reinforces that same culture.Hence it follows that consumer behavior presents an intrinsic link with population's culture, reflecting cognitive aspects and symbols within the goods and services that are consumed.
2.2 CONSUMER BEHAVIOR Cachon and Swinney (2009) suggest that research on consumer behavior are tools that help managers trace their objectives, understand the meanings and messages expressed by the ownership of products, and identify the wishes and expectations of the clients in each specific market.This can be observed in the description of Griskevicius and Kenrick (2013, p. 2) that defines consumer behavior as "How different people allocate their limited resources in different circumstances can how this can tell us a great deal about which needs people prioritize.It can provide insight into consumer preferences and decision processes".From a theoretical point of view, the purpose of understanding consumer behavior has been achieved through the development of analytical models that intend to portray the real world in which decision processes of purchasing products and their corresponding variables are present.
Studies such as those of Markin (1979), Kassarijan (1981), McAlister andPessemier (1982), andFerber (1984), organize many explanatory variables of different analytical models of consumer purchasing behavior.Such models attempt to predict the preferences among the alternatives, which are systematically related to psychological characteristics (STRAUGHAN; ROBERTS, 1999).
In the context of consumer behavior, according to Silverberg, Backman and Backman (1996), psychological characteristics are more complex than demographic ones.Eckman, Kotsiopulos e Bickle (1997) describe that psychological characterization adds activities,  , 1997).Baltas (2003) points out that the combination of psychological variables with demographic elements allows studies to go beyond identifying consumers by bringing an understanding the reasons why purchases are made.
White and Dahl ( 2007) describe that psychological characteristics, which are used in segmentation studies, such as the identification of target markets, may be fundamental tools to get to know consumers and their demographic background well.Researchers must use scrutiny to capture the state of mind of each consumer in order to identify the characteristics of their target audience.Baltas and Argouslidis (2007) explain the efforts of companies that attempt to position a product for their market and to make it suitable for an existing consumption pattern.Therefore, defining the characteristics of a product to a new market means aligning tangible and intangible elements of the product to the psychological characteristics of the target market.Kant (1724Kant ( -1804Kant ( /2002) ) apud Goergen (2005) stated that values are the prioritization of a norm and that it may or may not have a practical realization, but it provides truth, goodness and beauty to things that are subjected to choice.
Consequently, priorities of values will differ according to the environmental changes since people consider that this change in priorities represents a better way to live and express oneself (ROHAN, 2000).Rokeach (1973) SOLOMON, 2002).Allen (2000) indicates that even though values promote interests of individuals and social groups, motivate actions, and sometimes serve as a starting point for people to judge themselves and others, such judgments and preferences directly depend upon two processes namely abstraction and generalization.Rokeach (1973) andFeather (1982) suggest that the beliefs that an individual has about an object derive from the positive and negative experiences that occur with that object and the summary of the evaluations about these beliefs form a general perspective.
Given these findings, Allen (2000) proposes that the conclusions made by Rokeach (1973) andFeather (1982) regarding the beliefs about an object are the same as those shown in the expectancy theory model (FISHBEIN, 1967), which determines that each belief has an evaluation associated with it and that the combinations of beliefs and evaluations matched with other beliefs and evaluations pertaining to the object form an attitude towards it.Thus values determine the way people judge objects and form their beliefs about it, making relevant to understand how the relationship between values and judgment occur.

JUDGMENT AND MEANING
Lindberg, Garling and Montgomery (1989) argue that preferences between products derive from the value that the product's attribute emphasizes and how important consumers consider the emphasized value.Allen (2006) points out that this is the way consumers calculate the utility or preference for some products or brands by using a mathematical formula and then choose whichever one obtained the highest result.
However, Allen (2000) describes that such structure is restricted to tangible attributes of the product and can be considered limited, because it considers human values as indirect influence only in selection processes and only considers the utility value of products.Several studies analyzed the decision making process from another aspect, the hedonic aspect.
Hirschman and Holbrook (1982) affirm that when products are perceived symbolically, their physical characteristics are less valuable than their subjective representation that is attached to the product by the individual.On the other hand, Zajonc (1980) indicates that the sentimental association is immediate and precedes cognition excluding attention and processing elements that are connected to the identification of the object.He also argues that the very description of the object depends on its abstraction and meaning in the eyes of the individual that is describing it, thus "when we evaluate an object or an event, we are describing not so much  , 1980).
On the other hand, Lazarus (1982) defends that cognitive activity is a necessary prerequisite of emotion, since people must first comprehend the significance of what is happening in order to experience an emotion, (LAZARUS, 1984, p. 124).Nevertheless, Lazarus (1984) emphasizes that the preferences and the way that they are constituted reside in an uncertain zone between emotion and reason.Allen (2006) describes two types of judgment that derive from previous discussions, a piecemeal judgment, described in the studies of Fiske and Pavelchack (1986) as an evaluation in which "consumers evaluate products attribute-byattribute, and that the judgment of each product attribute independently contributes to the assessment of the product whole" (ALLEN, 2000, p. 2).The affective judgment, in turn, is derived from studies by Zajonc (1980) who says it evaluates the object as a whole, as opposed to comparing separate attributes, "the object is compared to a mental prototype, and if the two match, the affect associated with the category prototype is ascribed to the object in question" (ALLEN, 2006, p. 27).Allen and Torres (2006), on the other hand, argue that the type of judgment applied to a product derives from the meaning assigned to it by the individual, encompassing two categories, utilitarian or symbolic; being the utilitarian meaning represented by tangible and functional aspects and is directly related to the practical utility of the product, its performance and efficiency.In symbolic meaning, the attributes are intangible and represent an image or a symbolism of a product.Dittmar (2007, p. 34) defines symbol as "an entity that represents another entity and that can only have meaning to the extent that there is a shared understanding among the people that classify that symbol as real", thus, the symbolic meaning is linked to the group culture (DITTMAR, 1992).Allen (2006) summarizes the two types of judgment presented classifying them according to their characteristic as shown on

Source of benefit
The product's intrinsic qualities, means to an end, and ability to control de environment The use of the product as a vehicle for self-expression

Value relevance Low High
Product Attachment Weak Strong Source: adapted from Allen, M. W. Human values and product symbolism: Do consumers form product preference by comparing the human values symbolized by a product to the human values that they endorse?Journal of Applied Social Psychology, v. 32, n. 12, p. 2475Psychology, v. 32, n. 12, p. -501, 2006. .Therefore, culture can influence not only the values, but also individuals' perceptions concerning the characteristics that a product holds, enhancing or diminishing the importance of these characteristics.

METHOD
This section describes the research model, study sample, the instruments used and the procedures of data collection and data analysis.

RESEARCH MODEL
This research contemplates two distinctive models, one named model A that used Ttests and another one named model B that utilized stepwise regressions.
The model A, to note the similarity or dissimilarity between each type of judgment when evaluating cars comparing Brazil to The United States of America and the contrast between genders inside each country for each type of judgment as figure 1 shows.were enrolled in higher education or above, and of these individuals 79% and 55% were male.
The U.S. sample consisted of 461 individuals, with ages between 15 and 86 years (M = 42.2SD = 11.8),49.1% with income above the national average.Of these individuals 80% were attending a higher education or above and 52% of them were male.
To use multivariate analysis techniques such as T-tests and multiple linear regression, it is important to select a sample that has minimal statistical power greater than 0.80 (Cohen, 1992).Through the program GPower 3.1, we obtained the minimum sample of 114 subjects, with α = 0.05, ES medium and 9 predictors for the linear multiple regression and 64 subjects from each country considering α = 0.05, ES medium and power of 0.80 for the T-test.
Checking for outliers, using the Mahalanobis distance criterion (D =27.88; p<0.001), 57 subjects from the Brazilian sample and 35 from the United States of America sample were excluded.The percentage of missing data was lower than 5% as recommended by (HAIR JUNIOR et al. 2009;TABACHNICK;FIDELL, 2006) and was excluded by the Listwise Deletion method.The remaining sample from Brazil was 542 subjects and the remaining sample from the United States of America was 450 individuals, considerably more than the minimum needed.

INSTRUMENTS
To identify the values of respondents, the LOV scale (List of Values) (KAHLE; KENNEDY, 1988) was applied in which nine fundamental human values and one most important value are listed.Kamakura and Novak (1992) validated the value 9 dimensions using a constrained least-squares model, the same approach adopted by Schwartz and Bilsky (1992) to map their human values scale.To measure the type of judgment and the attribution of meaning to the product, the Meaning and Judgment Scale (ALLEN, 1997)  The chosen product for the survey was the automobile and in addition to these instruments, a list of demographics was presented, namely: sex, educational level, marital status, type of dwelling, number of dependents, age and income.

RESULTS
To attend the assumptions for multivariate analysis proposed by Tabachnick and Fidell (2006), Hair Junior et al. (2009) and Field (2009) the data were examined with no records of discrepancies in average and standard deviation.The assumptions for normality of the variables of the t-test and the linear multiple regressions were checked.Considering in particular the linear multiple regression and following the recommendations of Miles and Shevlin (2001) linearity, homoscedasticity and independence of the residuals were checked.Miles and Shevlin (2001) indicate that multicollinearity and singularity can be detected if the VIF (Variance Inflation Factor) is above 2.0 and tolerance close to zero.The highest variance inflation factor (VIF) among all the independent variables of the linear multiple regressions was 1.5 and the lower tolerance found was 0.754 indicating that these assumptions were not violated.As recommended by Hair et al. (2009), the test of equality of means between two independent samples indicated for group comparisons is the T-Test.So we compare the differences between North Americans and Brazilians regarding the predominant type of judgment (affective or piecemeal) in selecting a car.As a result of the T-Test for independent samples it was possible to observe differences in the type of judgment that prevails in each country, as shown in Table 4: The two factors presented significant differences for equal assumed variances, demonstrating that in Brazil affective judgment is predominant and its average is greater than the affective judgment in the United States for the automobile product.In the piecemeal judgment the results are reversed, its average is significantly higher for the sample collected in the United States resulting in a more affective judgment for Brazilians and a more rational one for North Americans regarding automobiles purchase.
In order to avoid type 1 error, caused by the influence of demographic variables (i.e. income, education, age, number of children) a multivariate analysis of covariance was performed (MANCOVA) for its sensitivity in terms of direction and size of the correlation between the dependent variables, and its ability to remove variance attributed to a variable that can influence the sample (HAIR JUNIOR et al., 2009).Gender has influenced the Brazilian sample, with n of 0,06 and power of 0,91.The other demographic variables did not achieve sig < 0,05 or power greater than 0,80, necessary conditions for it to figure as influential in the average effect for affective and piecemeal judgments (HAIR JUNIOR et al., 2009).Thus, a t-test for independent samples was performed only for the Brazilian sample using the piecemeal judgment factor and the sex as the grouping variable, as shown in Table 6.The t-test for independent samples suggests that for the Brazilian sample, women use in average the piecemeal judgment more predominantly than men.A t-test for independent samples considering the North American sample was then performed, in order to compare women and men regarding the predominant type of judgment used in terms of automobile purchase.Table 7 summarizes the results.

2. 3
HUMAN VALUES Values have been discussed and studied since Aristotle (2001), who defined them as what everyone desires and not what everyone should desire; in other words, values are the realization of what they consider to be important since human beings have the need to feel virtuously accomplished in what they consider natural, their reason and the way in which they blooms, is themselves (ARISTOTLE, 2001).Conversely, in the context of modern societies,

Figure 1 -
Figure 1 -Research model A The second model, named model B, tests in an exploratory manner how much each of the human values can predict both types of judgment in each country, Brazil and The United States of America, in this case piecemeal and affective judgment composed the model as dependent variables while the human values were tested as independent variables as figure 2 demonstrates.

Figure
Figure 2 -Research model B 3.2 SAMPLE The research was constituted by two non-probability convenience samples composed by Brazilians and North Americans.The American sample was collected online using MTurk in order to ensure the presence of a broad variety of industries located in the United States.Studies (BUHRMESTER; KWANG; GOSLIN, 2011, PAOLACCI; CHANDLER; IPEIROTIS, 2010) demonstrates that Mturk samples are more diverse and better than college Values (LOV) scale based on the 9 core values that individuals share independent of country location because they are based in the central dilemmas every human being has to face: needs of individuals as biological organisms, requisites of coordinated social interaction, and survival and welfare needs of groups(ROCKEACH, 1973).The 9 core values proposed by (DE MOOIJ, 2003;LINDQUIST;SIRGY, 2006)y learned values to help them rationalize about attitudes and behaviors that could otherwise be personally or socially unacceptable.Culturally learned values are hierarchically organized situational beliefs that serve to guide behavior and are internalized in the socialization process by the convergence of social institutions (e.g.family, school, friends).They constitute the core of personality and therefore are the basis of a self-concept(ROKEACH, 1973).Based onRokeach's (1973)structure of values,Kahle & Kennedy (1988)developed the List of Are Men More Rational than Women when Purchasing Cars?: A Comparison of Cultural Influences on Product Judgment in Brazil and the United States of America 78 BBR, Braz.Bus.Rev.(Engl.ed., Online), Vitória, v. 12, n. 6, Art. 4, p. 72 -99, nov.-dec.2015www.bbronline.com.br(tolead a pleasurable, happy life), Security (to be safe and protected from misfortune and attack), Self-Respect (to be proud of yourself and confident), Sense of Accomplishment (to succeed at what you want to do).Schwartz (2005) defines values as criteria that are used by people to evaluate actions, people and events.The author proposes a unification theory of the motivation and human behavior fields in order to develop a universal system of human values that contemplate the main values shared by all cultures (BLACKWELL; MINIARD; ENGEL, 2005, SCHWARTZ, 2005).According to the author, all people possess numerous common values that vary on levels of priority and importance from person to person.Because the values vary from individual to individual, from country to country and from culture to culture(DE MOOIJ, 2003), a detailed analysis of the values defended by a particular culture might demonstrate that certain approaches to product positioning, or even the product itself, are entirely unsuitable for a culture or a group of individuals, since the product may be in conflict with their values(DE MOOIJ, 2003;LINDQUIST;SIRGY, 2006).focusingon, recreate their advertisements and reposition their products in the market, communicating the values that are truly important to their customers in a more effective manner, and, it may even discourage a company from entering a country or a new culture (DE BBR, Braz.Bus.Rev. (Engl.ed., Online), Vitória, v. 12, n. 6, Art. 4, p. 72 -99, nov.-dec.2015 www.bbronline.com.brMOOIJ, 2003; BLACKWELL; MINIARD; ENGEL, 2005; LINDQUIST; SIRGY, 2006;

Table 1 -Types of Judgment that Affect Consumer Behavior
Are Men More Rational than Women when Purchasing Cars?: A Comparison of Cultural Influences on Product Judgment in Brazil and the United States of America 84 PROCEDURES The surveys were structured in Likert Scale and applied online using Mturk and in person, by the researchers.The respondents used, in average, 10 minutes to answer the survey completely.Using the Statistical Package for Social Sciences (SPSS) software, version 19.0 to analyze the data t-tests were conducted to analyze if the means for piecemeal judgment and affective judgment are different between Brazil and the United States of America.To ensure if there was influence of any demographic variable on the mean for each type of judgment on each country a MANCOVA test were applied.Demographic variables that influenced the mean of both types of judgment were tested using t-tests separately; only gender influenced the mean and was tested separately using t-tests.

Table 2
indicates the descriptive statistics and the correlations between all the studied variables.

Table 5 -MANCOVA Using the Demographics Variables
Table 5 presents the results of the MANCOVA for the Brazilian sample.Are Men More Rational than Women when Purchasing Cars?: A Comparison of Cultural Influences on Product Judgment in Brazil and the United States of America 88 BBR, Braz.Bus.Rev.(Engl.ed., Online),Vitória, v. 12, n. 6, Art. 4, p. 72 -99, nov.-dec.2015  www.bbronline.com.br