Developing a Parsimonious Measure of High-Low Context Communication Style

ABSTRACT - High- versus low-context communication style (HLCCS) refers to the extent to which individuals rely on contextual cues for meaning (high-contextBHC) as opposed to attending to the literal text alone (low-contextBLC). Seeking to develop a practical measurement, two studies were undertaken. The first study replicated 6 of 8 original factors identified by Gudykunst et al. (1996). A seventh factor combined two of the original eight. Four top loading items from each factor (28 total) were then subjected to confirmatory factor analysis (CFA) using a second sample to assess the reduced scale’s validity. While promising, the CFA results indicate a need for further refinement.



Citation:

Eugene S. Kim, Kawpong Polyorat, and Dana L. Alden (2003) ,"Developing a Parsimonious Measure of High-Low Context Communication Style", in E - European Advances in Consumer Research Volume 6, eds. Darach Turley and Stephen Brown, Provo, UT : Association for Consumer Research, Pages: 168-174.

European Advances in Consumer Research Volume 6, 2003      Pages 168-174

DEVELOPING A PARSIMONIOUS MEASURE OF HIGH-LOW CONTEXT COMMUNICATION STYLE

Eugene S. Kim, University of Hawaii at Manoa, USA

Kawpong Polyorat, University of Hawaii at Manoa, USA

Dana L. Alden, University of Hawaii at Manoa, USA

ABSTRACT -

High- versus low-context communication style (HLCCS) refers to the extent to which individuals rely on contextual cues for meaning (high-contextBHC) as opposed to attending to the literal text alone (low-contextBLC). Seeking to develop a practical measurement, two studies were undertaken. The first study replicated 6 of 8 original factors identified by Gudykunst et al. (1996). A seventh factor combined two of the original eight. Four top loading items from each factor (28 total) were then subjected to confirmatory factor analysis (CFA) using a second sample to assess the reduced scale’s validity. While promising, the CFA results indicate a need for further refinement.

INTRODUCTION

High-versus low-context communication content has been extensively examined in advertising as an example of differences in advertising strategy across countries (e.g., Callow and Schiffma 2001; Cho et al. 1999). Content analytic studies have consistently confirmed that advertisements from Western cultures employ more direct, verbal and/or printed text communication (low-contextBLC), while those in Eastern cultures rely more on indirect and often subtle cues "surrounding" the overall advertisement (high-contextBHC; cf., Cho et al.1999). Content analytic studies have made major contributions by documenting the prevalence of HC versus LC advertisements in different cultures. However, relatively little is known about the ways that individual differences on high versus low-context communication style (HLCCS) impact the relative effectiveness of advertising content within and across cultures. To date, three advertising studies involving high- versus low-context communication have employed an experimental approach (Callow and Schiffman 2001, forthcoming; Taylor, Miracle, and Wilson 1997). However, in these studies, the construct was not measured at the individual level. Rather, Korean, Filipino, and Spanish participants were assumed to employ a high-context communication style (HCCS) while their American counterparts were assumed to employ a low-context communication style (LCCS).

We believe that the scarcity of the experimental research using HLCCS stems in part from the lack of a valid, reliable and practical measure of the construct at the individual level. The absence of such a measure is surprising given that HLCCS along with individualism-collectivism are two of the most frequently cited differences in advertising content across cultures (Taylor et al.1997).

HIGH-VERSUS LOW-CONTEXT COMMUNICATION STYLE (HLCCS)

According to Hall (1976, p.91), "a high-context communication or message is one in which most of the information is either in the physical context or internalized in the person, while very little is in the coded, explicit, transmitted part of the message. A low-context communication is just the opposite; i.e., the mass of the message is vested in the explicit code." Certain cultures are hypothesized to encourage one or the other as the preferred style of communication. Thus, interpersonal communication in HC cultures is usually characterized as more indirect, nonverbal and ambiguous. In LC cultures, on the other hand, communication is relatively more direct, verbal, and unequivocal (Gudykunst and Matsumoto 1996). Hall (1976) identifies northern European, western European, and northern American countries as LC cultures and southern European, East Asian and Latin American countries as HC cultures. Despite widespread discussion of the construct, an extensive review of the literatures in marketing, advertising, speech communication and management revealed a limited number of studies that have attempted to develop individual level difference measures of HLCCS. These measures are now discussed.

EXISTING MEASURES OF HLCCS

Gudykunst et al.’s (1996) Measure

Gudykunst et al.(1996) developed an 80-item self-report scale using subjects from two HC countries (Korea and Japan) and two LC countries (the US and Australia). Exploratory factor analysis (EFA) indicated that the scale tapped eight dimensions of the construct. Six of these were hypothesized to reflect LCCS and two, HCCS (see Figure 1). Most of the dimensions are in line with Hall’s (1976) description of high versus low-context culture, for example, use of indirect/ambiguous communication in HC interactions. Interpersonal sensitivity also appears to have face validity as a HC dimension. Given the strong association of context-dependent cues with HC communication, interpersonal sensitivity appears far more important in HC interaction than it would n situations in which printed and verbal text were paramount (i.e., as in LC interactions).

Furthermore, use of feelings to guide communication behavior does seem less likely in HC cultures as HC communicators often must minimize display of feelings in order to maintain relationships (Gudykunst and Ting-Toomey 1988). As a result, high scores on this dimension seem likely to reflect preference for a LCCS. Dramatic communication is another dimension hypothesized to be LC. Here too, face validity is suggested by the observation that LC communicators more frequently view argument, persuasion and larger numbers of words as important to communication (De Mooij 1994). Finally, openness and preciseness are labeled as LC dimensions. This is justified by the reasoning that consumers in LC cultures tend to be more explicit and direct (Hall 1976). Whereas conversation in LC cultures is expected to provide neither more or less information than is required, verbal conversation in HC cultures often involves understatement (Gudykunst and Matsumoto 1996).

Despite such face valid EFA results, two dimensions from Gudykunst et al.’s (1996) factor analysis appear counterintuitive. First, Gudykunst et al. (1996) hypothesized that a high perceived ability to infer reflects LCCS. The researchers argue that on the surface, inferring others’ remarks is part of HC communication. However, the inference is said to be "taken-for-granted" in HC cultures. What is measured is the 'perceived ability’ to infer as opposed to what people actually do and because individuals in HC cultures take inference for granted, Gudykunst et al. (1996) argue that LC communicators will rate their "perceived ability to infer" more highly than will HC communicators. However, this argument is at odds with empirical findings reported by Holtgraves (1997) who reported that HC Korean scored significantly higher than LC Americans on interpretation of conversational indirectnessBa trait which presumably requires high levels of self-perceived inferential ability. Gudykunst et al.’s (1996) position is also at odds with that of Singelis and Brown (1995) who argue that the ability to intuitively understand others is a HC characteristic. As a consequence, the actual validity of this dimension as an indicator of LCCS is subject to question.

FIGURE 1

GUDYKUNST ET.AL.=S (1996) HLCCS

The other counterintuitive dimension is silence. In LC cultures, silence is space to be filled. In HC cultures, silence conveys meanings. Based on this functional difference, people in LC cultures would generally be assumed to experience more discomfort with silence because it interrupts the flow of conversation (Gudykunst and Matsumoto 1996). Therefore, LC communicators would be expected to hold more negative attitudes toward conversational silence. Gudykunst et al. (1996) argue the opposite. They suggest that because HC people tend to use silence to convey negative information, LC communicators will express relatively more positive attitudes toward silence. This position is supported by Hasegawa and Gudykunst (1998) who report that Japanese versus Americans have a more negative view of silence when communicating with strangers. Even, without solid evidence that HC communicators use silence predominantly to communicate negative information, the theoretical rationale employed by Gudykunst et al. (1996) appears questionable and is ripe for an additional study.

Kim, Pan, and Park’s (1998) Measure

Developed using samples from the U.S., China, and Korea, Kim, Pan, and Park’s (1998) 16-item scale does not directly measure preference for HLCCS. Rather, their scale is hypothesized to measure five correlated constructs: social orientation, responsibility, confrontation, communication/commitment, and dealing with new situations. As predicted, Chinese and Koreans (Americans) displayed characteristics consistent with HC (LC) communication. For example, Chinese and Koreans were more socially oriented, less confrontational, and more content with current ways of living than Americans. Although consistent with high- versus low-context (HLC) characteristics, the relationships between these constructs and actual communication style have not been directly tested. Rather than representing styles of communication per se, it appears more reasonable to view the five constructs as antecedents to or consequences of HLCCS.

Other Measures

There are other measures that are relevant to a discussion of HLCCS. For example, the Conversational Indirectness Scale or CIS (Holtgraves 1997) was developed to measure the extent to which individuals prefer to communicate indirectly. Although evidence suggests this scale is both reliable and valid (Holtgraves 1997), it taps only one dimension of HLC communication (indirectness). Based on the literature, HLC appears to be a multi-dimensional construct. As a consequence, CIS is unlikely to tap the richness of the HLC construct relative to approaches proposed by Gudykunst et al.(1996) and Kim et al.(1998).

Finally, Singelis and Brown’s (1995) measure uses HLC communication as an outcome variable through a set of four scenarios. Each scenario features three parts: (a) a short description of the people in the scenario, (b) a description of the setting of the scenario, and (c) a verbatim dialogue consisting of an exchange of a greeting, a request and a response. Six questions measuring reliance on context, attributions to context and receiver versus sender orientation follow each scenario. Unfortunately, the lengthy format in this approach limits its usefulness in experimental work in which subjects are typically exposed to multiple stimuli and measures. In addition, the scenarios, by nature, are context-specific. As a consequence, scenarios developed with student subjects might not be applicable for non-student subjects. With a similar line of reasoning, scenarios developed in a LC culture might not work as well for subjects from a HC culture.

After thorough review of the relevant measures, we have the following concerns regarding their practicality in experimental and survey settings. First, use of Singelis and Brown’s (1995) scenario measure may require excessive time and limit researcher’s ability to execute multi-factor/multi-measure designs or surveys in the field. Second, Holtgrave’s (1997) CIS approach taps only one of several eight dimensions associated with the HLC construct. Third, Kim et. al.’s (1998) scale measures HLCCS through five constructs that appear more appropriate as antecedents or consequences rather than direct indicators.

As noted earlier, Gudykunst et al.’s (1996) scale attempts to directly measure an individual’s HLCCS. In addition, its self-report format makes it practical for use in both experimental and survey settings. Even so, questions remain regarding the validity of some of the scale’s dimensions. Such questions, coupled with a need to reduce the current scale’s 80-item format, suggest possible benefits from further research prior to application. To this end, we conducted two studies.

STUDY 1

Study 1 seeks to determine whether or not Gudykunst et al.’s (1996) original eight factors that emerged following EFA are replicable. Assuming a reasonable EFA solution, Study 1 also provides a basis for selecting a reduced number of items for further testing.

Method

Two hundred and six junior and senior business students from a major Western university participated in this study. Extra credit was given to increase motivation. The original 80-item Gudykunst et al.’s (1996) scale was distributed to students as a take-home survey. One hundred and ninety-three usable questionnaires were returned. There were slightly more female than male students (107 vs. 86).

Results

The means of the 80 items ranged from 2.52 to 5.85 (from 7-point scale) with standard deviations from 1.04 to 1.75. The 80 items were factor analyzed using principal axis factor extraction with promax rotation since the HLC factors are thought to be correlated (Gudykunst et al. 1996). An imposed 8-factor EFA solution accounted for 62% of the total variance compared with 30% from the original study. Six factors which are similar to the original factors emerged. The first factor was comprised of items previously on the Dramatic Communication and Openness dimensions. Given the combination of constructs, we have re-labeled the factor as Self-Expressive Communication. The eighth factor was dropped due to interpretation difficulty. It contained only 3 items from Gudykunst et al.’s (1996) Openness and Preciseness factors.

To shorten the scale, four items with the highest factor loadings from each of the seven retained factors were selected. Altogether, 28 items remained in the shortened version. These 28 items are provided in the Appendix. The reliabilities of the 7 four-item factors ranged from .55 to .83. The four items in each factor were combined to compute sum scores. The sum score of each factor exhibited a high correlation with the original longer version, ranging from .81 to .93. This result provides preliminary evidence that the shortened version accurately represents the latent constructs from the full scale.

Factors from the original 80-item scale were replicated fairly well in this study. This confirmed the robustness of the proposed factors in the original scale. However using the original scale in the consumer research remains problematic due to its length. The shortened version appears promising but would benefit from further validity testing prior to use. To this end, we now present the results of Study 2.

STUDY 2

The primary objective of Study 2 is to assess the reliability and validity of the shortened version of Gudykunst et al.’s (1996) scale. Reliability is tested through an attempted replication of our initial EFA results using the reduced version of the scale. Validity is determined using two approaches. First, confirmatory factor analysis (CFA) is employed to discover how well the theoretical latent factor model fits the data. Second, relationships of the shortened scale with scales measuring related constructs are identified, including the Conversational Indirectness Scale (Holgraves 1997); Self-Construal Scale (Singelis 1994); and the Five Cultural Dimensions Scale, based on Hofstede (1990) and developed by Furrer, Liu, and Sudharshan (2000).

Conversational Indirectness (Holtgraves 1997) refers to the extent to which people express their meanings directly or indirectly and the extent to which they look for indirect meanings from others. As noted earlier, this construct is theoretically related to HLC because one distinction of HC versus LC is the use of indirect and direct communication. As a consequence we expected this construct to exhibit higher correlation with HC items in Gudykunst et al.’s scale than with LC items. The Conversational Indirectness Scale or CIS (Holtgraves 1997) is comprised of 19 items measuring two separate but related dimensions: conversational indirectness interpretation and production. The scale’s reliability and validity have been supported in previous research and it has been cross-culturally validated. For example, a series of studies by Holtgraves (1997) found that reliability ranged from .89 to .91. The scale’s convergent validity and discriminant validity have also been demonstrated with several other measures, including: need for cognition, self-monitoring, assertiveness and social desirability.

Another conceptually related construct is self-construal. Previous research found that the use of HC communication is related to interdependent self-construal while the use of LC communication is related to independent self-construal (Gudykunst et al. 1996; Singelis and Brown 1995). As result, it is expected that the nomological validity of the shortened scale will be demonstrated through the correlations with this construct. Self- construal was measured with Singelis (1994)’s self-construal scale (SCS). This scale measures two images of self. Both are conceptualized as reflecting an emphasis on connectedness and social relations (interdependent) or separateness and uniqueness of the individual (independent). The scale has been extensively used in consumer research (e.g., Aaker 2000).

Hofstede (1990) identifies five cultural dimensions; i.e., individualism (the extent to which ties between individuals are loose or tight); power distance (the extent to which the less powerful members expect and accept that power is distributed unequally); masculinity (the extent to which social gender roles are clearly distinct); uncertainty avoidance (the extent to which the members of a culture feel threatened by uncertain situations); and long-term orientation (the extent to which a society exhibits a pragmatic future-oriented perspective rather than a historic or short-term point of view).

We expect that individualism will correlate with HLC scores because this cultural dimension is related to self-construal (SC). However the magnitude of the correlation of individualism and HLCCS should be smaller than that between self-construal and HLC because SC is hypothesized to mediate the relationship (Gudykunst et al. 1996; Singelis and Brown 1995). In addition, HLCCS should exhibit low or no correlation with the other four cultural dimensions. Because Hofstede’s original research setting was the workplace, we employed Furrer et al.’s (2000) measures, which were adapted for use in a consumer research setting.

FIGURE 2

CONFIRMATORY FACTOR ANALYSIS RESULTS

Method

Procedures followed those used in Study 1, except that students completed not only the shortened 28-item (instead of the full version) scale but also CIS, SCS, and Hofstede’s cultural dimension scales. Two hundred and forty-six useable questionnaires were analyzed.

Results

EFA produced a 7-factor solution similar to that reported in Study 1 with similar reliabilities. This outcome further suggests that the shortened version reliably represents the full 80-item scale. To analyze the validity of the 7-factor solution, HC and LC items were subjected to CFA as two separate dimensions. The rationale for this approach was two-fold. First, preliminary analysis of the sum scores showed that the sum scores of HC items and LC items were uncorrelated (r=.02, p=n.s.). In addition, HLCCS has been found to be a consequence of independent-interdependent self-construals. Because these two SC dimensions are considered by Singelis (1994) as orthogonal, it seems reasonable to assume that HLCCS will exhibit a similar relationship.

Our CFA analysis produced mixed results. The indices in Table 1 suggest that the HC model in Figure 2 fits the data reasonably well, but the LC model falls short of achieving a satisfactory overall fit. In particular, for the LC model, the GFI and CFI measures were slightly lower than .9 , which are below recommended limits (Browne and Cudek 1993).

In addition, although the HC model achieved a relatively good overall fit, the two latent variables in the HC model exhibited an unexpected negative correlation (-.17) and some item loadings were less than ideal.

TABLE 1

FIT INDICES

TABLE 2

CORRELATION OF HLC WITH CIS, SCS AND HOFSTEDE=S CULTURAL DIMENSIONS

Relations with other scales

To assess nomological validity, correlations of HLC scores with other theoretically related scales were computed. Four items in each factor were combined to yield a sum score for each factor. Next, sum scores from the HC factors were combined and sum scores from LC factors were combined. As result, each respondent had two scores, one for HC and the other for LC communication. HLC correlations with the CIS, SCS, and Hofstede’s cultural dimensions are shown in Table 2.

The HC (LC) sum score was significantly related to interdependent (independent) self as predicted. However, the smaller but still significant correlation between LC and interdependent self was not expected. Furthermore, there was a significant correlation between LCCS and the CIS interpretation dimension but not CIS production. The first correlation seems counterintuitive inasmuch as one would not expect LC communicators to be strong on indirect interpretation. However, this result is consistent with Gudykunst et al.’s (1996) argument that LC communicators have higher perceived (not necessarily actual) ability to infer others’ message. As expected, HC exhibited significant correlations with both CIS interpretation and production dimensions.

Finally, correlations between HC and LC dimensions and Hofstede’s cultural dimensions were not significant. This outcome was expected for four of the dimensions and indicates that the reduced scale possesses good discriminant validity. In addition, the fact that individualism/collectivism was not significantly related to HC/LC does not necessarily suggest lack of convergent validity, as this measure is undoubtedly less sensitive than its individual difference level counterpart, self-construal, which was positively correlated with HC/LC as expected.

DISCUSSION

This research is an initial attempt to provide a HLC scale that can be used in either experimental or survey consumer research. The original 80-item Gudykunst et al.’s (1996) scale was first reduced to 28 items based on an EFA, which replicated the original study reasonably well. Both this study and a second study indicated that the shortened version reliably represents the original scale. Furthermore, the second study provides preliminary evidence of the shortened scale’s promise. However, Study 2’s findings also indicate the need for additional research before consumer scholars can confidently apply the more parsimonious scale (or the full scale, given the high correlations between the longer and shorter versions).

First, as previously discussed, not all CFA fit indices achieved ideal levels. In addition, three out of seven dimensions in Study 1 and four out of seven dimensions in Study 2 had alphas below the traditional cut-off of .70. Thus, certain items or even dimensions may need to be rephrased, removed or added. Using only the LC items may be considered, given the low alphas (<.60) of the two HC dimensions. Second, our respondents were students at an American university. A test of the shortened scale in other cultures using a mix of groups viewed as predominantly HC versus LC would strengthen the cross-cultural validity of the scale.

Finally, to illustrate the potential value of future research using this scale, we performed a second CFA using the data from Study 2. In this analysis, Factor 6, silence was dropped from the model. This was done for two reasons. First, the previous CFA had resulted in a negative correlation between silence and all the other factors. This counterintuitive outcome casts doubt on the construct’s face validity within the nomological net specified by Gudykunst et al.’s (1996) model since all factors are hypothesized to directly predict LC. In addition, as mentioned earlier, there are theoretical reasons for questioning the validity of silence as a LC factor. CFA results suggest that the 4-factor, 16-item LC model fits the data better (Chi-Square=192.56, df=98, GFI=.91, AGFI=.88, CFI=.89, 90% confidence level of RMSEA=.05-.07) than the 5-factor model. Results such as these suggest that one or more of Gudykunst et al.’s (1996) factors may be more problematic than others.

It is also possible that the less than desirable outcome regarding HLC scale development is due in part to the effects of postmodernity on general consumer experience. That is, multiple realities are rapidly developing and previously valid categories of cultural patterns, for example, HLCCS ay not be as relevant (Steenkamp, Batra and Alden 2003). A casual glance at magazines, whether they are from the West or the East, will reveal advertisements with charged images, but little or no text. These advertisements, which may have been previously categorized as indirect or HC, are more prevalent in this meaning-charged world, and consumers, whether they are culturally LC or HC, may be increasingly more adept at interpreting indirect meanings. New conceptualization of HLC may be required to develop a more ecologically valid scale. [We appreciate the comment from an anonymous reviewer for this direction.]

In sum, these studies have attempted to develop a valid yet parsimonious measure of HLC at the individual level. Given the wide scope of the HLC concept, and the complexity of its underlying dimensions, future research to better understand the construct and refine its measurement appears warranted.

APPENDIX

TWENTY-EIGHT ITEMS THAT WERE REMAINED IN THE PARSIMONIOUS VERSION OF GUDYKUNST ET AL.=S (1996) HLC SCALE

REFERENCES

Aaker, Jennifer L. (2000), "Accessibility or Diagnosticity? Disentangling the Influence of Culture on Persuasion Process and Attitude," Journal of Consumer Research, 26(March), 340-357.

Browne, Michael W. and Robert Cudeck (1993), "Alternative Ways of Assessing Model Fit, " in Testing Structural Equation Models, eds. Kenneth A. Bollen and J. Scott Long, Newbury Park, CA: Sage.

Callow, Michael A. and Leon G. Schiffman (2001), "Contextual Communication Styles and Their Effect on Visual Imagery in Print Advertising: A Cross-Cultural Study, in Advances in Consumer Research, Vol. 28, eds. Mary C. Gilly and Joan

Meyers-Levy, Valdosta, GA: Association for Consumer Research, 389.

Meyers-Levy (forthcoming), "Over-Reading into the Visual Print Ad: The Development of Weak Implicatures among Consumers from the United States, Spain, and the Philippines," in Advances in Consumer Research, Vol. 29, eds. Susan M.

Broniarczyk and Kent Nakamoto, Valdosta, GA: Association for Consumer Research.

Cho, Bongjin, Up Kwon, James W. Gentry, Sunkyu Jun, and Fredric Kropp (1999), "Cultural Values Reflected in Theme and Execution: A Comparative Study of U.S. and Korean Television Commercials," Journal of Advertising, 28 (4), 59-73.

De Mooij, Marieke K. (1994), Advertising Worldwide: concepts, theories and practice of international, multinational and global advertising, New York: Prentice-Hall.

Furrer, Olivier, Devanathan Sudharshan, and B. Liu (2000), "The Relationship between Culture and Service Quality Perceptions: Basis for International Market Segmentation and Resource Allocation," Journal of Service Research, 2 (4), 355-371.

Gudykunst, William B. and Yuko Matsumoto (1996), "Cross-Cultural Variability of Communication in Personal Relationships," in Communication in Personal Relationships across Cultures, eds. William B. Gudykunst, Stella Ting-Toomey, and Tsukasa Nishida, Thousand Oaks, CA: Sage, 19-55.

Gudykunst, William B., Yuko Matsumoto, Stella Ting-Toomey, Tsukasa Nishida, Kwangsu Kim, and Sam Heyman (1996), "The Influence of Cultural Individualism-Collectivism, Self Construals, and Individual Values on Communication Styles across Cultures," Human Communication Research, 22 (4), 510-543.

Gudykunst, William B. and Stella Ting-Toomey, with Elizabeth Chua (1988), Culture and Interpersonal Communication, Newbury Park, CA: Sage.

Hall, Edward T. (1976), Beyond Culture, New York: Doubleday.

Hasegawa, Tomohiro and William B. Gudykunst (1998), "Silence in Japan and The United States," Journal of Cross-Cultural Psychology, 29 (September), 668-683.

Holtgraves, Thomas (1997), "Styles of Language Use: Individual and Cultural Variability in Conversational Indirectness," Journal of Personality and Social Psychology, 73 (3), 624- 637.

Kim, Donghoon, Yigang Pan, and Heung Soo Park (1998), "High versus Low-Context Culture: A Comparison of Chinese, Korean, and American Cultures," Psychology & Marketing, 15 (6), 507-521.

Singelis, Theodore M. (1994), " The Measurement of Independent and Interdependent Self-Construals," Personality and Social Psychology Bulletin, 20, 580-591.

Singelis, Theodore M. and William J. Brown (1995), "Culture, Self, and Collectivist Communication: Linking Culture to Individual Behavior," Human Communication Research, 21 (30), 354-389.

Steenkamp, Jan-Benedict and Hans Baumgartner (1998), "Assessing Measurement Invariance in Cross National Consumer Research, "Journal of Consumer Research, 25 (June), 78-90.

Steenkamp, Jan-Benedict, Rajeev Batra and Dana L. Alden (2003), "How Perceived Brand Globalness Creates Brand Value," Journal of International Business Studies, 34, 53-65.

Taylor, Charles R., Gordon E. Miracle, and R. Dale Wilson (1997), "The Impact of Information Level of the Effectiveness of U.S. and Korean Television Commercials," Journal of Advertising, 26 (1), 2-15.

----------------------------------------

Authors

Eugene S. Kim, University of Hawaii at Manoa, USA
Kawpong Polyorat, University of Hawaii at Manoa, USA
Dana L. Alden, University of Hawaii at Manoa, USA



Volume

E - European Advances in Consumer Research Volume 6 | 2003



Share Proceeding

Featured papers

See More

Featured

The Best of Both Worlds: Androgyny in Consumer Choice

Niusha Jones, University of North Texas
Blair Kidwell, University of North Texas

Read More

Featured

Inside Out: Product Essence is Perceived to be Concentrated in the Center of a Group of Products

Kunter Gunasti, Washington State University, USA
Noah VanBergen, University of Cincinnati, USA
Caglar Irmak, University of Miami, USA

Read More

Featured

The Messy Satiation Effect: The Benefits of Eating Like a Pig

Kevin L. Sample, University of Georgia, USA
Kelly Haws, Vanderbilt University, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.