Keywords

Negotiation, Cultures, Individualism, SEM

JEL Classifications

M14, M16

Introduction

In the 21st century, increased globalization and economic openness have helped to accelerate the volume growth of international business. According to Taiwan’s Ministry of Economic Affairs, major overseas trading partners include China, Japan and the United States. In 2015, the total value of Taiwan's international trade amounted to US$196.4 billion and US$24.5 billion worth of goods and services (representing 12.5 percent) was traded with the United States, ranking the US as Taiwan’s second largest trading partner. Both foreign investment and international trade continue to grow rapidly, thereby leading to an increased degree of interdependence among national economies, as well as to the increased globalization of companies. However, despite an enthusiasm for increased global interaction and economic exchange, a lack of understanding of cultural differences has been found to hinder the ability of firms to efficiently conduct business negotiations with different countries.

In view of the acceleration in international trade and investment, cross-cultural awareness is of crucial importance to the success of any business. With the resulting increase in the frequency of face-to-face negotiations, the nature of the strategies, styles and agreements used in negotiation are becoming increasingly important (Kumar et al., 2004). Successful negotiation requires not only the clear communication of the technical aspects of the exchange, but also involves the understanding of both parties of the context of the negotiation (Korobkin, 2000).

Wheeler (2006) reported that real-world negotiations are far more challenging than they might seem in theory, particularly in terms of the emotional demands made on the participants. Lee & Trim (2008) suggested that a shared organizational culture could help in the management of international partnerships, and that senior managers need to know the cultural traits of their international customers. Because of the considerable cultural differences that exist between geographical regions, as well as rapidly increasing globalization, multilateral negotiations are extremely important, and attract increasing research attention (Faure & Shakun, 1999). An understanding of the impact of culture on the style of negotiation is useful for all parties that are involved (Chang, 2003). The study of the effect of culture on international styles of negotiation is therefore key to the promotion of successful international negotiations.

The aims of the study described herein were to examine the hypotheses that culture influences the style of negotiation, and that nationality is a moderated variant that affects attitude to individualism and style of negotiation, and that styles of negotiation differ according to culture. The study also sought to identify the implications for international negotiations throughout the world of business, as well as for further research.

Literature Review

Culture

Culture is commonly defined to be a set of shared values and beliefs that characterize the behavior of groups in national, ethnic, moral and other related terms (Faure & Sjöstedt, 1993; Craig & Douglas, 2006; Adapa, 2008). Individual cultures can be revealed through the food, songs and stories that are exchanged with people outside that region (Parra, 2001). Schein (1997) added that culture is a pattern of shared basic assumptions that a group learns as it solves its problems of external adaptation and internal integration, which has been sufficiently successful to be considered valid, and are therefore taught to new members as the appropriate means to perceive, think and feel in relation to those problems. Simintiras & Thomas (1998) defined culture as a set of accepted values and norms that influence the ways in which people think, feel and behave. Because subcultures, cultures and super cultures merge and evolve, although being less bounded than they used to be, they have certainly become more porous and varied over time (Barbash & Taylor, 1997).

Cultural differences are important while conducting business abroad (Tu, 2015).With the goal of helping individuals to distinguish between the cultures of different countries, Hofstede (1980, 1994) formulated the theory of cultural dimensions. This theory identifies the major variables of cultural differences, which consist of power, uncertainty/avoidance, characteristics of individualism/collectivism, and masculinity/femininity. Hofstede proposed that cultural differences influence conduct, decision making and communication in business, and that collectivist and individualist values play a prominent role in the important areas of cross-cultural psychology, international management and religion (Hofstede, 1993; Kim et al., 1994; Triandis, 1995). Of Hofstede’s four cultural dimensions, it is the individualism/collectivism contrast that is most often employed in cross-cultural studies of negotiation (Bazerman et al., 2000).

Hall (1976) introduced the theory of high-low context cultures based on his studies about communication styles, and indicated that the context of communication significantly influences on the business negotiation. This theory stresses the influence of high-context cultures, defined as those which rely primarily on non-verbal/informal communication; and low-context cultures, which rely on verbal/formal communication (Simintiras & Thomas, 1998). Mintu-Wimsatt & Gassenheimer (2002) also indicated that the contexts of communication styles are embodied in high- and low-cultures. Therefore, the competing communicative styles of high- and low-context cultures that business negotiators rely on are often seen as hindrances to achieving beneficial outcomes (Fisher, 1983).

Negotiation

In the broadest sense, negotiation is a type of social interaction that involves the discussion of the issues concerned in order to reach an agreement, and satisfies all the parties that have different objectives or interests (Gulbro & Herbig, 1994; Foroughi, 1998; Manning & Robertson, 2003; Shakun, 2009). Negotiation is about/for those whose aim is to reach an agreement among two or more parties that have different objectives or interests (Fraser & Zarkada-Fraser, 2002). The process of negotiation that takes place between a buyer and seller is of great importance to both parties (Neslin & Greenhalgh, 1983; Federici-Nebbiosi, 2007), and the achievement of success through negotiation is considered to be one of the most challenging tasks that a business faces in terms of its communicative activities (Gilsdorf, 1997). However, the individual culture of each party determines their way of thinking, as well as their values, norms and behaviors (Simintiras & Thomas, 1998; Hung, 1998; Woo & Pru’homme, 1999; Chang, 2003).

Cross-Cultural Negotiation

Gulbro & Herbig (1994) stated that different cultures are associated with different styles of negotiation. These differences in style are the product of variances in means of communication, protocols, strategies of persuasion, and personal characteristics including accommodation, determination, flexibility and adaptation (Hung, 1998). Cross-cultural negotiations are made more complicated as a result of a range of factors, such as those relating to environment, language, ideology and customs (Mintu-Wimsatt & Gassenheimer, 2000; Hoffmann, 2001). Gulbro & Herbig (1995) stated that “when negotiating internationally, this translates into anticipating culturally related ideas that are most likely to be understood by a person of a given culture.” A number of authors have demonstrated that culture is one of the most important factors involved in cross- cultural negotiation (Hofstede, 1980; Gulbro & Hrbig, 1994; Schein, 1997; Salacuse, 2005). Because of the level of sophistication of the knowledge that is required to conduct these exchanges, many negotiators are unsuccessful in reaching agreements as a result of the challenges involved in overcoming cultural differences, as opposed to any economic or legal problems (Gulbro & Herbig, 1995).

Negotiation Styles

Jung (1973) indicated that there are two different ways, perceiving and processing functions, of perceiving information from inner or outer world into our psychic compass. Perceiving functions include senses and intuition, and processing functions contain thinking and feeling. Casse & Deol (1981) explained when a situation employs a sensing function, people focus on the facts, and try to be factual, objective, neutral, and as accurate as possible; when people use intuition function, they apply imagination, looking for possibilities and opportunities, and project into the future of the situation; when a situation applies thinking function, people process information through senses and intuition in a logical, neutral, objective analysis, systematic and scientific ways; and when people utilize feeling function, they process data relevance and importance by their value system. Casse & Deol (1981) went a step further to explain the four dominant physical functions determining the nature of four negotiation styles which are, sensing function is as factual; intuition function is as intuitive (as Jung’s perceiving functions); and thinking function is as analytical; and feeling function is as normative negotiation styles (as Jung’s processing functions).

The study that the research describes herein is based on the model that utilizes four styles of negotiation as dependent variables, which was developed and defined by Pierre & Surinder (1985) as follows: FA or Factual Style: people using the factual style are cool, collected, patient, down-to- earth, present-oriented, precise, realistic, able to document their statement, sticking to the facts that speak for themselves. IN or Intuitive style: this style is characterized by a charismatic tone, a holistic approach, a strong imagination, a tendency to jump from one subject to another, a lot of ups and downs, a fast pace, a deductive way to approach problems as well as a future orientation. NR or Normative Style: For those who use this style negotiating is basically bargaining. They judge, assess and evaluate the facts according to a set of personal values. They appeal to feelings, offer bargains, propose rewards and incentives. They look for compromises. AN or Analytical Style: The basic assumption that underlies this style is that “logic leads to the right conclusions”. These people for reasons of their own, analyze each situation in terms of cause and effect, put things into a logic order weight pros and cons, and use a sort of linear reasoning. They are unemotional and focus upon the relationship of parts.

Research Methodology

By means of a thorough examination of the styles of employees’ negotiation of public companies, the present research offers an understanding of the cultural differences between Taiwan (collectivism and high-context culture) and the US (individualism and low-context culture), and the effect of these on the styles of negotiation (perceiving and processing functions) concerned. The research model (Figure 1) and research hypotheses are as follows:

Research Hypotheses

H1: The cultural characteristic related to an individualist attitude is a direct path, and is a factor that significantly affects the style of negotiation employed.

H2: Nationality is a moderated variant between individualism and style of negotiation.

H3: In the United States, negotiators employ different styles of negotiation from those employed in Taiwan.

Instrumentation

A three-part Questionnaire For Cross-Cultural Negotiation Styles (QCNS) was adapted from Tu (2014a; 2014b); Farazmand, et al., (2012); Tu & Farazmand (2007); and modified here in order to measure the three research variables of individualism, style of negotiation and socio- demographic profile.

In the present study, the individualist characteristics that reflect cultural differences were the independent variables, and these four styles were the dependent variables used in the research model. In the questionnaire, five of the items were designed to examine these individualist characteristics, and three of the items were for each negotiation style by means of a five-point Likert scale. There were five possible responses for each statement: 5=“Always” (100% of the time); 4=“Often” (75%); 3=“Occasionally” (50%); 2=“Seldom” (25%); and 1=“Never” (0%). There were a total of 21 questions concerning the individualist characteristics and four styles of negotiation.

The model also took account of the socio-demographic characteristics of the participants, including gender, age, level of education, and region of birth. For the demographic variables, participants provided their own responses. These socio-demographic questions and the coding schemes used included: Region of birth (as a dichotomous variable): 1=The United States; 2=Taiwan). Gender (a dichotomous variable): 1=male; 2=female. Education (as a nominal variable): 1=High School Diploma or Equivalent; 2=Associate Degree; 3=bachelor degree; and 4=graduate degree. Age (as a nominal variable): 1=under 35; 2=35–45; 3=46–55; and 4=over 55.

The population chosen for the study was taken from public companies listed on the Taiwan Stock Exchange Corporation (TSEC) and the New York Stock Exchange (NYSE). The TSEC contained about 700 listed companies, which were classified into eight sectors. The NYSE contained about 3,000 listed companies, which were classified into 13 sectors. Any of the companies listed on these two exchanges could have been included in the study.

Data was collected using an online survey, with a hyperlink to the survey website that was provided on each emailed invitation. Versions of the survey in English and traditional Chinese characters were posted on the research website, and participants were directed to their preferred versions from the e-mail. Altogether, 350 invitations were e-mailed to the sales and purchase managers of public companies in each country. In order to avoid sampling errors, the population was collected from all different sectors of the stock markets in the two regions. A random sample was obtained from each of the sectors using stratified random sampling. This method was useful because it provided a small but carefully selected pool of data that can offer some insight into the general trends in the larger population. Overall, this method of obtaining a sample population is more accurate than purely random sampling. Furthermore, it allows the researcher to select a sample that accurately reflects the diverse sectors and characteristic patterns in the population of interest (Wallen & Fraenkel, 2001).

Methods of Data Analysis

Hair, et al., (2010) indicated that Structural Equation Modeling (SEM) had become a popular multivariate approach because it provided a means of assessing theories that were conceptually appealing. AMOS software (version 18.0), which includes an SEM package with maximum likelihood estimation, was used to test both the measurement and the structural models that related to the research hypotheses listed. The present research also made use of a number of criteria to determine the inclusion of items, and the goodness of fit, of the model. Hair et al., (2010) suggested a six-stage procedure for employing SEM, which the research also followed here.

Stevens (1996) stated that both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) can be used in practical research of this kind. Principal component analysis and the rotation method of varimax with Kaiser Normalization were employed to examine the underlying structure of the questionnaire. Anderson & Gerbing (1988) claimed that a two-step approach had a number of comparative strengths that permitted meaningful inferences to be made. The present study therefore made use of a two-step approach that included both EFA and CFA. First, EFA was used to pretest the questionnaire in order to reduce the items to a manageable and meaningful set of factors, and the reliability of the internal consistency was measured using Cronbach’s coefficient alpha. Reliability and validity were then examined using CFA, which provided a confirmatory test of how the measured variables fit logically and systematically into a theoretical model, and was a way of evaluating how well they represent a smaller number of constructs (Hair et al., 2010). Where the CFA offered an accurate measurement, the same data was tested using the SEM.

The primary purpose of these factor analyses was to assess the associations among the stated variables in terms of the correlations obtained, and to examine any underlying factors. Results of the Kaiser–Meyer–Olkin (KMO) test, and of Bartlett’s test, were obtained before performing the factor analyses. The KMO test indicated whether a sufficient number of items had been predicted by each construct, and Bartlett’s test indicated whether the items were sufficiently highly correlated to provide a reasonable basis for factor analysis. Cronbach’s coefficient alpha was used to analyze the variables related to the scales of each item, according to the average correlation of each item with every other item. Leech, et al., (2005) recommended that KMO values should be greater than 0.7, and Bartlett’s test should be significant. A factor loading of 0.50 or above was considered to be of practical significance (Hair et al., 2010). The lower limit for Cronbach’s coefficient alpha values was 0.7 (Leech et al., 2005).

The validity of the construct was measured using the convergent and discriminant validity. The convergent validity was used to determine whether scale items converged on a single construct during measurement (Steenkamp & Van Trijp, 1991). This was determined from the evaluation of the factor loadings (which must be at least 0.5), composite reliability (at least 0.6) and average extracted variance (at least 0.5) in the study (Fornell & Larcker, 1981; Hair et al., 2010). The discriminant validity was the extent to which a construct was truly distinct and unique, and this measure captures phenomena that other measures do not (Hair et al., 2010).

Hair et al., (2010) indicated that the goodness-of-fit of the overall model was indicated by how well it reproduces the observed covariance matrix among the indicator items. It can be classified into the following four categories:

• Chi-square measures including chi-square, degree of freedom (df) and probability.

• Measures of absolute fit, including The Goodness-Of-Fit Index (GFI), Root Mean Square Error of Approximation (RMSEA), Root Mean Square Residual (RMR), Standardized Root Mean Square Residual (SRMR) and normed chi-square.

• Incremental fit measures including the Normed Fit Index (NFI), the Tucker–Lewis Index (TLI), the Comparative Fit Index (CFI) and the Relative Noncentrality Index (RNI).

• Parsimony fit measures including the Adjusted Goodness-Of-Fit Index (AGFI) and the Parsimony Normed Fit Index (PNFI).

Chi-square (χ²) is a basic measurement of the differences between the observed and estimated covariance matrices (Hair et al., 2010). A smaller value of χ² is more desirable in that it supports the proposed theoretical model, but values of χ² also increase as the sample size increases. The p-value should be large and not statistically significant (p>0.05) between the two matrices (Jöreskog & Sörbom, 1992).

GFI was an early attempt to produce a fit statistic. The range of possible GFI values is between 0 and 1, and if the value is 0.90 or higher, the fit is considered to be good (Hair et al., 2010); however, MacCallum & Hong (1997) suggested that the GFI value could decrease to 0.80 in usage. RMSEA tries to correct for both the sample size and complexity of the model by including each in its computation. Steiger (1990) suggested that RMSEA values below 0.10 indicate a good fit, but Browne & Cudeck (1993); Hair, et al., (2010) argued that the value of RMSEA should be

0.08 or less. Hair, et al., (2010) indicated that RMR is problematic because it is related to the scale of the covariance. An alternative statistic is SRMR, which is useful for comparing the fit across models. Jöreskog & Sörbom (1992) indicated that an acceptable SRMR value would be 0.05 or less. The normed chi-square is given by χ²/df, and its value should be 3 or less to indicate a better fit between the observed and modeled values (Hair et al., 2010).

NFI is the ratio of the difference in the value of χ² between the fitted and null models, divided by the value of χ² for the null model (NFI=1 is a perfect model; Hair et al., 2010). Bentler (1992) suggested that the value of NFI should be 0.90 or above. The TLI is similar to the NFI, but compares the values of the normed chi-square and the specified model. A model with a good fit should have a TLI approaching unity (Hair et al., 2010). CFI is an improved version of NFI. It ranges between 0 and 1, with values above 0.90 being associated with a good fit (Gerbing & Anderson 1992; Hair et al., 2010). The RNI compares the observed fit that results from testing a specific model to that of a null model. A value of RNI less than 0.90 is associated with a poor fit (Hair et al., 2010).

AGFI takes into account different degrees of complexity in the model, and its value is usually lower than that of the GFI in complex models (Hair et al., 2010). MacCallum & Hong (1997) recommended that the value of AGFI should be 0.80 or higher to indicate a good fit. The PNFI adjusts the NFI by multiplying it by the parsimony ratio; high values represent a better fit (Hair et al., 2010). Wu (2009) indicated that the value of the PNFI should be 0.50 or above to indicate a good fit.

Data Analysis and Results

Demographic Result

The collection of data lasted for one month, during which time 700 e-mails of invitation were sent and 275 were returned. However, 42 of the questionnaires that were returned were incomplete or invalid. All questionnaires were coded for statistical analysis using the SPSS 14.0. From the 233 respondents, there were 115 (49.4%) usable questionnaires from Taiwan and 118 (50.6%) from the United States. 111 (47.6%) respondents were male and 122 (52.4%) were female.

19 (8.2%) respondents had a high school diploma or lower qualification, 81 (34.8%) held a Bachelor's degree and 133 (57.1%) had a graduate degree. 32 (13.7%) of the respondents were under 35 years old, 40 (17.2%) were between 35 and 45, 140 (60.1%) were between 46 and 55 and 21 (9.0%) were over 55. The demographic characteristics are presented in Table 1.

Table 1 Demographic Characteristics
Profile	Classification	Frequency	(%)
Gender	Male	111	47.6
	Female	122	52.4
	Total	233	100
Age	Under 35	32	13.7
	35-45	40	17.2
	46-55	140	60.1
	Above 55	21	9
	Total	233	100
Education Background	High school degree or below	19	8.2
	Bachelor’s degree	81	34.8
	Graduate degree	133	57.1
	Total	233	100
Religion	Christian	74	31.8
	Buddhist	84	36.1
	Other	75	32.2
	Total	233	100
Region	America	118	50.6
	Taiwan	115	49.4
	Total	233	100

Measurement Model and Scale Accuracy Testing

The five dimensions and 21 items of the QCNS scale were evaluated using EFA before conducting CFA. For the first-time EFA, all items of the factor loadings less than 0.80 or greater than
0.95 were deleted. For the second-time EFA, the KMO value of the variables used in the study was 0.78, indicating that the data from the results were sufficiently robust to allow EFA. The values of Bartlett’s test were χ²=2610, df=105 and p=0.000, which implied that all the items in this study were sufficient for research in social science and for factor analysis. The extraction and rotation sums of the squared loading of the total variance explained were 84.958%. Three items remained for each dimension which could therefore now be applied. The five dimensions of Cronbach’s coefficient alpha were between 0.83 and 0.94, which surpassed the criteria and indicated an internal reliability of the consistency of the instruments used in the present study that were appropriate for research in social science. As a result of EFA, five factors and 15 items were therefore derived to identify the QCNS construct.

The univariate normality of the skewness and kurtosis values and the multivariate normality were used to assess the normality. The most commonly used critical values of univariate normality are ±3 and ±10 (Kline 1998). In the study, all the values of skewness were between 0.833 and –0.899, and the values of peakedness lay between 2.672 and –1.999. The observed variables all had univariate normal distributions. The value of the Mardia statistic is for a multinormality measurement, and it is constructed on a test based on skewness and kurtosis. Bollen (1989) indicated that if the value of Mardia is smaller than p(p+2), p indicating the amount of observed variables, all dimensions are multinormality. In this study, the value of Mardia is 51.057, smaller than 15(15+2), indicating multivariate normality distribution.

The property of unidimensionality implies that a set of variables can be measured using one underlying construct (Gerbing & Anderson, 1988). For the collectivism/individualism, the value of χ² was 1.876 with two degrees of freedom, and the p-value associated with this result was 0.391. For the analytical negotiation, the value of χ² was 2.962 with two degrees of freedom, and the p-value associated with this result was 0.227. For the normative negotiation, the value of χ² was 4.385 with two degrees of freedom, and the p-value associated with this result was 0.112. For the factual negotiation, the value of χ² was 5.719 with two degrees of freedom, and the p-value associated with this result was 0.057. For the intuitive negotiation, the value of χ² was 1.345 with two degrees of freedom, and the p-value associated with this result was 0.511. The results showed that the assumption of the homogeneity of variances was not violated here, because Levene’s test rejected the hypothesis (p >0.05) that the means of the variances were unidimensional.

In the second-order CFA and structural models, all the factor loading estimates were higher than 0.70, and all the composite reliability (CR) values ranged from 0.83 to 0.94, and all the extracted average values of variance lay between 0.62 and 0.84. This evidence supports the convergent validity of the measurement model, as shown in Tables 2 and Table 3.

Table 2 Factor Loadings, Instrument Validity, and Instrument Reliability Measures
Variables	Items	EFA loadings	CFA loadings	Mean	SD	Cronbach’s Alpha (a)
IND	IND1	0.88	0.84	8.7	3.33	0.88
	IND2	0.89	0.87
	IND3	0.89	0.82
AN	AN1	0.94	0.95	10.3	1.72	0.94
	AN2	0.91	0.91
	AN3	0.92	0.89
NR	NR1	0.92	0.9	11.26	1.93	0.94
	NR2	0.94	0.94
	NR3	0.92	0.91
FA	FA1	0.81	0.7	13.12	1.49	0.83
	FA2	0.87	0.82
	FA3	0.88	0.84
IN	IN1	0.9	0.89	10.83	1.95	0.92
	IN2	0.86	0.86
	IN3	0.91	0.9

Table 3 Test of Composite Reliability, Convergent Validity, and Discriminant Validity
	CR	AVE	MSV	FA	IND	AN	NR	IN
FA	0.83	0.62	0.09	0.79
IND	0.88	0.72	0.13	0.01	0.85
AN	0.94	0.84	0.09	0.3	0.03	0.91
NR	0.94	0.84	0.13	-0.11	-0.08	-0.2	0.92
IN	0.92	0.79	0.13	-0.1	-0.36	-0.26	0.36	0.89

Bagozzi & Phillips (1982) stated that discriminant validity was supported if the number 1 is not included within the computed confidence interval. In the present research, a model was constructed for each of the 10 paired correlations of the latent variables. Then, the correlation was set between the two constructs to 1, and a 95 percent confidence interval was applied in order to apply a bootstrap. As a result, all values of paired correlations of the latent variables were from –0.504 to 0.441, the number 1 was not included with the upper and lower limits of the confidence interval, which indicated discriminant validity among the theoretical constructs.

In order to allow cross-cultural comparisons, a cross-validation strategy was used to assess the stability of the model. This involved the random splitting of all samples into a calibration sample and a validation sample. There are three ways to assess cross-validity, namely loose, moderate and tight replication strategies (MacCallum et al., 1994). For the loose replication strategy, the value of χ² was 13.296 with 10 degrees of freedom, and the p-value associated with this result was 0.208. For the moderate replication strategy, the value of χ² was 3.940 with four degrees of freedom, and the p-value associated with this result was 0.414. For the tight replication strategy, the value of χ² was 5.850 with five degrees of freedom, and the p-value associated with this result was 0.321. All the p-values showed no significant differences between the two split samples, which led to the measurement invariance.

The second-order CFAs were conducted using AMOS 18.0, and the model fits are reported in Table 4. The measurement model fit χ² was 300.36 with 175 degrees of freedom. The p-value associated with this result was 0.000. The value of the absolute fit index RMSEA was 0.06. This value seems low and was less than the guideline value of 0.08 for the model. The GFI had a value of 0.86, and RMR was 0.02. Moreover, the χ²/df was 1.72 and demonstrated an acceptable fit for the second- order CFA model.

Table 4 CFA And SEM Model Fit Indices of The Measurement Model
Indices	Criteria	SEM	CFA
χ2/df	<3	1.33	1.72
P-value	<0.05	0.022	0
Absolute fit measures
RMSEA	=0.08	0.04	0.06
GFI	>0.80	0.94	0.86
RMR	<0.05	0.03	0.02
Incremental fit measures
CFI	>0.90	0.99	0.95
NFI	>0.90	0.96	0.88
RFI	>0.90	0.95	0.86
TLI	>0.9	0.99	0.93
Parsimony fit measurement
AGFI	>.80	0.92	0.81
PNFI	>.50	0.77	0.73

For the incremental fit indices, the CFI is the most widely used. In the second-order CFA model, the CFI had a value of 0.95, which exceeded the CFI guidelines for a model of this complexity and sample size. The other incremental fit indices (NFI=0.88, RFI=0.86, and TLI=0.93) also exceeded the suggested cutoff values. The parsimony index of AGFI had a value of 0.81, and PNFI=0.73, which reflected a good model fit. The second-order CFA results suggested that the measurement model provided a reasonably good fit, and were therefore appropriate for the further examination of the model results.

In addition, the SEM model fits are also presented in Table 4. The overall model fit χ² was 113.33 with 85 degrees of freedom. The p-value associated with this result was 0.022. The p-value was significant using a type I error rate of 0.05; thus, the χ² goodness-of-fit statistic did not indicate that the observed covariance matrix matches the estimated covariance matrix within the sampling variance. According to previous research, a number of indices were available to evaluate model fits (Fornell & Larcker, 1981; Bentler, 1990 & 1992; Jöreskog & Sörbom, 1992), but no single index or standard was generally agreed; hence, multiple criteria should be used to evaluate the overall fit of the theoretical model (Bagozzi & Yi, 1988; Hatcher, 1994; Hair et al., 2010).

The value of RMSEA, an absolute fit index, was 0.04. This value was smaller than the guideline value of 0.08 for a model with 15 measured variables and a sample size of 236. Therefore, RMSEA supports the model fit. The value of GFI 0.94 was higher than the guideline value. RMR had a value of 0.03. Further, the χ²/df was 1.33 and suggested an acceptable fit for the structural model.

In the SEM model, the CFI had a value of 0.99, which exceeds the CFI guidelines for a model of this complexity and sample size. The other incremental fit indices (NFI=0.96, RFI=0.95, and TLI
= 0.99) also exceeded the suggested cutoff values. All the incremental fit indices presented an acceptable fit. The parsimony index of AGFI had a value of 0.92 and the PNFI was 0.77. Both indices were considered to represent a good model fit, given the acceptable critical value. The overall structural fit results of these analyses showed that the model provided a reasonable fit. In addition, the structural equation model is presented in Figure 2.

Research Hypotheses

H1: The cultural characteristic related to an individualist attitude is a direct path and is a factor that significantly affects the style of negotiation employed.

The unstandardized parameter estimates had a value of 0.08, and the value of the standardized parameter estimates was 0.37. The standard error was 0.03, and the t-value was significant (p=0.012*).

H2: Nationality is a moderated variant between individualism and style of negotiation.

As stated in the literature review, different nationalities demonstrate different preferences for styles of negotiation. The present research has established two country-based groups (Taiwan and the US) and used two models to look for significant differences between them. A multigroup SEM was used to test the moderation. The CFA measurement invariance was estimated at the start of the cross-validation. The value of χ² was 20.392 with one degree of freedom and a p-value=0.000, as shown in Table 5.

Table 5 Assuming Model Unconstrained to be Correct
Model	DF	CMIN	P	NFI	IFI	RFI	TLI
				Delta-1	Delta-2	rho-1	rho-2
Moderation	1	20.392	0	0.008	0.008	0.009	0.010

H: In the United States, negotiators employ different styles of negotiation to those utilized in Taiwan.

Hair et al., (2010) indicated that multiple group analysis provided a comprehensive framework for testing the differences between two or more samples of participants. The approach of measurement equivalence was employed, and multiple CFAs extended to separate samples in order to determine equivalence. For the United States, the estimated values of analytical, normative, factual, and intuitive styles were 0.419, 0.367, 0.289 and -0.809; and t-values were 5.675***, –4.213***, 4.580*** and –10.673***, respectively as shown in Table 6.

Table 6 Summary of H1 & H2 Hypotheses Analysis Results
Hypotheses	Path	B	Result	Model
H1	IND → NEG	0.37**	Supported	Default Model
H2a	IND → NEG	0.48**	Supported	US Model
H2b	IND → NEG	-0.35**	Supported	Taiwan Model

H3: There is a significant different between United States negotiation style and Taiwan negotiation style.

Hair et al., (2010) indicated that multiple group analysis provided a comprehensive framework for testing the differences between two or more participants. The approach of measurement equivalence was employed, and multiple CFAs extended to separate samples to determine equivalence. For the United States, the estimated values of analytical, normative, factual, and intuitive styles were 0.67**, -0.25 (NS), 0.57**, and 0.22 (NS); and for Taiwan 0.31(NS), -0.58**, -0.25 (NS) and -0.72** respectively as shown in Table 7.

Table 7 Summary of H3 Results - Estimates of us and Taiwan Styles of Negotiation
Variables	Estimate (US)	Estimate (TW)
Analytical style	0.67**	0.31(NS)
Normative style	-0.25 (NS)	-0.58**
Factual style	0.57**	-0.25 (NS)
Intuitive style	0.22 (NS)	-0.72**

Discussion and Conclusion

The aim of the study reported herein was to examine the assumption that culture influences the style of negotiation. The major findings indicated that the individualist attitude was a direct, positive and significant factor that affected the style of negotiation. The first Hypothesis (H1) was therefore supported, and the result was consistent with the findings of Oetzel & Ting-Toomeys (2003). For the Second Hypothesis (H2), the results showed that nationality was a moderated variant between individualism and style of negotiation, and this hypothesis was also therefore supported. For the Third Hypothesis (H3), the results showed that negotiators from Taiwan, as opposed to US negotiators, preferred to employ normative and intuitive styles of negotiation, whereas US negotiators employ factual and analytical styles more than negotiators from Taiwan.

Miller, et al., (1997) stated that Taiwan was characterized by a blend of different cultural influences, as a result of having been controlled by a number of different imperial powers, including Holland, Spain and Japan. Yoo, et al., (2006) stated that in recent decades, Taiwan has maintained close economic ties with the United States, but is culturally different.

Due to the different countries, cultures and races, nationalities are a moderating effect that changes the relationship between two related variables, which differed significantly between the individualism/collectivism attitude and negotiation styles. Taiwan belongs to an eastern culture, and is deeply influenced by Confucianism. America belongs to a western culture, and is mainly influenced by Christianity, among many other religious beliefs. Confucianism values long-term relationships, and hierarchy and family are important. Hassan, Dollard & Winefield (2010) announced that collectivist cultures (mostly eastern) tended to integrate work and family relationships, and work was as a way of supporting families. Tu (2007) indicated that education was a significant factor affecting negotiating styles. Confucian teachings emphasized modesty and obedience, and American culture respects student freedom, creativity, and independence (Yu, 2008).

Tu (2010) found that Taiwan was more collectivist in nature; this result was consistent with the findings of Hofstede (1991), which indicated that a highly individualist culture, such as the United States, is more person-centered, whereas in Taiwan, the organization is more important than the individual. Barry (2001) indicated that in individualist societies, each individual takes care of him/herself, in contrast to collectivist ones, in which groups of people take care of the individual. Those who live in collectivist cultures are typically more concerned with the group and social welfare, while those from individualist cultures tend to be more concerned with their own rights, benefits and outcomes (Hofstede, 1980). The values of collectivism emphasize the importance of the group. By contrast, individualist values place importance on individual development and expression, even at the expense of the collective (Triandis, 1990). Gulbro & Herbig (1999) claimed that high levels of collectivism resulted in more time being spent on indirect activities that were unrelated to communication, and high levels of individualism resulted in more time being spent on direct communication. Drnevich (2003) reported that negotiators from cultures that are characterized by a high degree of individualism might have difficulty in achieving synergistic or integrative outcomes during the process of negotiation.

An understanding of the differences and similarities between the cultures involved facilitates communication and increases the chance of success in negotiations. It is to be hoped that the findings of the present study will improve the general understanding of the styles of negotiation in Taiwan and the US, and help businesses to develop better strategies that may be of benefit in maintaining their competitive advantages. The researcher suggests that the negotiators still need to be trained in different skills, such as body language, strategies, anger management, international manners, and customs. A better knowledge of negotiation should, and will always, be helpful in understanding business and in realizing which negotiation styles are most appropriate for a particular country.

The study was limited to public companies listed on the TSEC and NYSE and the sales and purchase managers of those listed companies. The study was constrained by financial resources and time; therefore, it adopted only a quantitative research method, and only one factor of individualism was examined here. Although the SEM provided a good fit to the hypothesized model, future research would be able to use a different design to examine the causal relationships posited by the theories of negotiation. Alternatively, a comparison could be made of the differences and similarities among the styles of negotiation used in a number of different countries, such as Asia’s four little dragons or within regions of the European Union. Additionally, future studies possibly will employ a qualitative method, and the sampling plan could be expanded to include negotiators who are not sales and purchase managers.

Acknowledgement

We gratefully acknowledge MOST-Taiwan (Ministry of Science and Technology, Taiwan) for providing necessary funding to carry out this research (MOST-108-2410-H-468-002-MY2). The authors declare no conflict of interest (Corresponding Author: Massoud Moslehpour - writetodrm@gmail.com).

References

1. Adapa, S. (2008). Adoption of internet shopping: Cultural considerations in India and Australia. Journal of Internet Banking and Commerce, 13(2), 1-17.
2. Anderson, J.C., & Gerbing, D.W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411-423.
3. Bagozzi, R.P., & Yi, Y. (1988). On the evaluation of structure equations models. Academic of Marketing Science, 16(1), 76-94.
4. Bagozzi, R.P., & Phillips, L.W. (1982). Representing and testing organizational theories: A holistic construal. Administrative Science Quarterly, 27(3), 459–489.
5. Barbash, I., & Taylor, L. (1997). Cross-cultural filmmaking. Berkeley and Los Angeles, California: University of California Press.
6. Barry, S. (2001). Information sharing in global supply chain systems. Journal of Global Information Technology Management, 4(3), 27-50.
7. Bazerman, M.H., Curhan, J.R., Moore, D.A., & Valley, K.L. (2000). Negotiation. Annual Review of Psychology, 51, 279-314.
8. Bentler, P.M. (1992). On the fit of models to covariance and methodology to the bulletin. Psychological Bulletin, 112(3), 400-404.
9. Bentler, P.M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238-246.
10. Bollen, K.A. (1989). Structural equations with latent variables. Newbury Park, CA: Sage.
11. Browne, M.W., & Cudeck, R. (1993). Alternative ways of assessing model fit, in K.A. Bullen, J.S. Long (edition). Testing Structural Equation Models. Newbury Park, CA: Sage.
12. Casse, P. (1984). Training for the multicultural manager. Washington DC: SIETAR International.
13. Casse, P., & Deol, S. (1985). Managing intercultural negotiations: Guidelines for trainers and negotiators. Washingtom, DC: International Society for Intercultural Education.
14. Chang, L.C. (2003). An examination of cross-cultural negotiation: Using Hofstede framework. Journal of American Academy of Business, 2(2), 567-570.
15. Craig, C.S., & Douglas, S.P. (2006). Beyond national culture: Implications of cultural dynamics for consumer research. International Marketing Review, 23(3), 322–342.
16. Drnevich, P. (2003). The role of cultural distance in international negotiations. Purdue University - Center for International Business Education and Research.
17. Farazmand, F., Tu, Y.T., & Daneefard, H. (2012). A comparative study of negotiation styles: Afghanistan, Iran and Syria. Journal of International Business Research, 11(2), 61-72.
18. Faure, G.O., & Sjöstedt, G. (1993). Culture and negotiation: An introduction. In culture and negotiation: The resolution of water disputes, edited by G.O. Faure, J.Z. Rubin. Thousand Oaks, CA: Sage Publications.
19. Faure, G.O., & Shakun, M.F. (1999). Introduction to the special issue on intercultural negotiation. Group Decision and Negotiation, 8(3), 183-185.
20. Federici-Nebbiosi, S. (2007). The wisdom of sharing a frame. Psychoanalytic Dialogues, 17(1), 47-58. Fisher, G. (1983). International negotiation: Cross-cultural perception. The Humanist, 43(11), 14-18.
21. Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.
22. Foroughi, A. (1998). Minimizing negotiation process losses with computerized negotiation support systems. Journal of Applied Business Research, 14(4), 15-27.
23. Gerbing, D.W., & Anderson, J.C. (1992). Monte Carlo evaluations of goodness of fit indices for structural equation models. Sociological Methods Research, 21(2), 132-160.
24. Gerbing, D.W., & Anderson, J.C. (1988). An updated paradigm for scale development incorporating unidimensionality and its assessment. Journal of Marketing Research, 25(2), 186-192.
25. Gannon, M.J. (2001). Understanding global cultures. Newbury Park: Sage Publications, Inc.
26. Gilsdorf, J.W. (1997). Metacommunication effects on international business negotiation in China. Business Communication Quarterly, 60(2), 20-38.
27. Gulbro, R.D., & Herbig, P. (1994). The effect of external influences in the cross-cultural negotiation process. Journal of Strategic Change, 3, 329-340.
28. Gulbro, R.D., & Herbig, P. (1995). Cross-cultural negotiation behavioral differences: Demestic–focused versus worldwide-oriented firms. Cross Cultural Management, 2, 3-14.
29. Gulbro, R.D., & Herbig, P. (1999). Cultural differences encountered by firms when negotiating internationally. Industrial Management+Data Systems, 99(2), 47-56.
30. Hair, J.F., Black, W.C., Babin, B.J., & Anderson, R.E. (2010). Multivariate data analysis. Pearson, NJ: Pearson Education Inc.
31. Hall, E. (1976). Beyond culture. Garden City. NY: Anchor Press/Double Day.
32. Hassan, Z., Dollard, M.F., & Winefield, A.H. (2010). Work-family conflict in East vs Western countries. Cross Cultural Management, 17(1), 30-49.
33. Hoffmann, G. (2001). When scientists or engineers negotiate. Research Technology Management, 44(6), 13-16.
34. Hofstede, G. (1993). Cultural constraints in management theories. Academy of Management Executive, 7(1), 81-94.
35. Hofstede, G. (1991). Cultures and organizations: Software of the mind. McGraw-Hill, London.
36. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Beverly Hill, CA: Sage Publications.
37. Hofstede, G. (1994). Cultures and organizations: Software of the mind: Intercultural cooperation and its importance for survival. London: Harper Collins.
38. Hung, C.L. (1998). Canadian business pursuits in the PRC, Hong Kong and Taiwan, and Chinese perception of Canadians as business partners. Multinational Business Review, 6(1), 73-83.
39. Jöreskog, K.G., & Sörbom, D. (1992). LISREL: A guide to the program and applications, 3rd (edition). Chicago: Scientific Software International, Inc.
40. Jung, C.G. (1973). Jung’s psychology and its social meaning. New York; Anchor Books.
41. Kim, U., Triandis, H.C., Kagitcibasi, C., Choi, S.C., & Yoon, G. (editions) (1994). Individualism and collectivism: Theory, method, and applications. Newbury Park, CA: Sage.
42. Korobkin, R. (2000). A positive theory of legal negotiation. George Law Journal, 88(6), 1789-1732. Kline, R.B. (1998). Principles and practice of structural equation modeling. Guilford Press, New York.
43. Kumar, R., Markeset, T., & Kumar, U. (2004). Maintenance of machinery: Negotiating service contracts in business-to-business marketing. International Journal of Service Industry Management, 15(4), 400-421.
44. Leech, N.L., Barrett, K.C., & Morgan, G.A. (2005). SPSS for intermediate statistics use and interpretation. NJ: Mahwah. Lawrence Erlbaum.
45. Lee, Y.I., & Trim, R.J. (2008). The link between cultural value systems and strategic marketing: Unlocking the mindset of Japanese and South Korean managers. Cross Cultural Management, 15(1), 62-88.
46. MacCallum, R.C., & Hong, S. (1997). Power analysis in covariance structure modeling using. Multivariate Behavioral Research, 32, 193-210.
47. MacCallum, R.C., Roznowski, M., Mar, M., & Reith, J.V. (1994). Alternative strategies for cross-validation of covariance structure models. Multivariate Behavioral Research, 29, 1-32.
48. Manning, T., & Robertson, B. (2003). Influencing and negotiating skills: Some research and reflections-Part I: Influencing strategies and styles. Industrial and Commercial Training, 35(1), 11-15.
49. Miller, G., Yang, J., & Chen, M. (1997). Counseling Taiwan Chinese in America: Training issues for counselors. Counselor Education and Supervision, 37(1), 22-34.
50. Mintu-Wimsatt, A., & Gassenheimer, J.B. (2000). The moderating effects of cultural context in buyer-seller negotiation. The Journal of Personal Selling and Sales Management, 20(1), 1-9.
51. Mintu-Wimsatt, A., & Gassenheimer, J.B. (2002). The impact of demographic variables on negotiators' problem- solving approach: A two country study. Journal of Marketing Theory and Practice, 10(1), 23-35.
52. Neslin, S., & Greenhalgh, L. (1983). Nash's theory of co-operative games as a predictor of the outcomes of buyer-seller negotiations: An experiment in media purchasing. Journal of Marketing Research, 20, 368-379.
53. Oetzel, J.G., & Ting-toomey, S. (2003). Face concerns in interpersonal conflict: A cross-cultural empirical test of the face negotiation theory. Communication Research, 30(6), 599-624.
54. Parra, C. (2001). Political psychology: Cultural and cross-cultural foundations. New Zealand Journal of Psychology, 30(1), 40-43.
55. Salacuse, J.W. (2005). Negotiating: The top ten ways that culture can affect your negotiation. Ivey Business Journal, 69(4), 1-6.
56. Schein, E.H. (1997). Organizational culture and leadership. CA: Jossey-Bass Inc..
57. Shakun, M.F. (2009). Connectedness problem solving and negotiation. Group Decision and negotiation 18(2), 89-117.
58. Simintiras, A.C., & Thomas, A.H. (1998). Cross-cultural sales negotiations: A literature review and research propositions. International Marketing Review, 15(1), 10-28.
59. Steenkamp, J.B.E.M., & Van Trijp, H.C.M. (1991). The use of LISREL in validating marketing constructs. International Journal of Research in Marketing, 8(4), 283-299.
60. Steiger, J.H. (1990). Structural model evaluation and modification: An interval estimation approach. Multivariate Behavioral Resaerch, 25(2), 173-180.
61. Stevens, J. (1996). Applied multivariate statistics for the social science, (4th edition). NJ: Mahwah, Lawrence Erlbaum.
62. Triandis, H.C. (1995). Individualism and Collectivism. Boulder, CO: Westview.
63. Triandis, H.C. (1990). Cross-cultural studies of individualism-collectivism, In J. Berman (edition). Nebraska Symposium on Motivation, 1989 (41-133). Lincoln: University of Nebraska Press.
64. Tu, J.T. (2010). Cross-cultural comparisons between Taiwan and America. Interdisciplinary Journal of Contemporary Research in Business, 2(4), 116-129.
65. Tu, J.T. (2007). Impact of culture on International business negotiations: A cross-cultural comparison of Taiwan, Hong Kong and Mainland China. Dissertation, Lynn University.
66. Tu, J.T., & Farazmand, F. (2007). Impact of culture on International business negotiations: A cross-cultural comparison of Taiwan, Hong Kong and Mainland China. Paper presented at the Proceeding of International Conference on Business, Management and Economics (ICBME), June 13–17, in Yasar University, Izmir.
67. Tu, Y.T. (2014a). Trust affecting on negotiation styles. International Journal of Humanities and Social Science, 4(1), 259-267.
68. Tu, Y.T. (2014b). Cross-cultural equivalence and latent mean differences of the Negotiation Style Profile (NSP-12) in Taiwan and US managers. International Journal of Intercultural Relations, 43, 156-164.
69. Tu, Y.T. (2015). A cross-cultural comparison of BRIC on negotiation styles. The Anthropologist, 19(2), 457-467.
70. Wallen, N.E., & Fraenkel, J.R. (2001). Educational research: A guide to the process, 2d. NJ: Mahwah, Lawrence Erlbaum Associates, Inc.
71. Wheeler, M. (2006). Is teaching negotiation too easy, too hard, or both? Negotiation Journal 22(2), 187-198.
72. Woo, H.S., & Prud’homme, C. (1999). Cultural characteristics prevalent in the Chinese negotiation process. European Business Review, 99(5), 313-315.
73. Wu, M.L. (2009). Structural equation modeling: Amos operation and application. Taipei, Taiwan: Wu-Nan Book Inc,.
74. Yoo, D.K., Rao, S.S., & Hong, P. (2006). A comparative study on cultural differences and quality practices - Korea, USA, Mexico, and Taiwan. The International Journal of Quality & Reliability Management, 3(6), 607-624.
75. Yu, F. (2008). The expatriate experience: Teaching nursing across Eastern and Western cultures. Journal of Nursing Scholarship, 40(2), 184-186.