Multidimensional Networking Behavior in Germany and China

Abstract

Prior research has so far established the multidimensional nature of networking behavior in Western but not Asian working populations. Thus, the purpose of the present study was to assess the cross-cultural measurement equivalence and predictive validity of an established multidimensional networking scale across a German and a Chinese sample of individuals employed by the same multinational company (total N = 248). Multigroup confirmatory factor analyses demonstrated configural and metric, but not scalar measurement invariance, indicating that cross-cultural comparisons of associations between the six theoretically anticipated networking dimensions and other constructs but not of mean scores of networking are meaningful. Path analytic findings showed that some but not all networking dimensions significantly predicted objective career success (i.e., salary and promotions, assessed 2 months later). Relative weight analyses indicated that all intraorganizational (i.e., building, maintaining, and using contacts), but no extraorganizational networking behaviors explained nontrivial variance in Chinese employees’ salary and promotions. In the German group, a largely opposite pattern was found. Interestingly, the relative importance of building and of using internal contacts was significantly greater in China than in Germany. We conclude that the functional but not necessarily the structural facet of a multidimensional conceptualization of networking behavior exhibits meaningful cross-cultural equivalence.

Keywords

networking career success longitudinal study cross-cultural measurement invariance relative weight analysis

Integrating the growing body of interdisciplinary research on networking behavior, Gibson, Hardy, and Buckley (2014) proposed a consensus definition of networking as “a form of goal-directed behavior, both inside and outside of an organization, focused on creating, cultivating, and utilizing interpersonal relationships” (p. 150). This complex multidimensional notion of networking (see also Porter & Woo, 2015) echoes recent advances in the networking literature that have relied on a multidimensional scale (Wolff & Moser, 2006; Wolff, Schneider-Rahm, & Forret, 2011) to begin to map out the differential effects that functional and structural facets of networking behavior exert on individual-level career outcomes (McCallum, Forret, & Wolff, 2014; Wolff & Moser, 2009, 2010).

So far, however, it has remained undetermined whether a multifaceted conceptualization of networking comprising multiple dimensions of engaging in informal relationships at work (i.e., building, maintaining, or using contacts either within or outside one’s organization; Gibson, Hardy, & Buckley, 2014; Wolff & Moser, 2006) can be readily applied to cultures other than Western ones. Implicitly, contemporary career models (e.g., Lent & Brown, 2013) describe adaptive career behaviors such as networking as universal, implying their cross-cultural equivalence in globalized occupational environments. However, this assumption has not received much scholarly attention, thwarting efforts to advance a culturally more informed understanding of the nature of networking behavior. The few studies conducted so far do not allow firm conclusions as they rely on ad hoc choices of items or scales (Rasdi, Garavan, & Ismail, 2013; Wok & Hashim, 2014) and only one study compared networking using samples from two cultures, albeit it did not examine whether cross-cultural comparisons were warranted (Y. H. Cheung, Dougherty, & Herndon, 2011).

Thus, the purpose of the present article is to investigate the measurement invariance of a multidimensional networking scale, so far validated in German- and English-speaking working populations (Gervorkian, 2011; Wolff & Moser, 2006; Wolff et al., 2011), across a Western (i.e., German) and Asian (i.e., Chinese) group of employees working for the same multinational company (MNC). This approach allows to largely rule out confounding effects of organization-specific variables (e.g., human resource [HR] practices, organizational culture, or socialization) so as to facilitate conclusions regarding the adequacy of a multifaceted construal of networking in collectivistic cultures such as China (Hofstede, 2003).

A further aim of this study is to examine and compare the predictive validity of the present networking scale in both cultural groups to derive a more comprehensive picture of the widely assumed role of networking as an important agentic vocational behavior (Gibson et al., 2014; Porter & Woo, 2015) across cultural boundaries. Specifically, we assessed relationships of networking dimensions with objective career success (i.e., salary and promotions) and pursued an exploratory approach to elucidate patterns of relative predictive importance within and between both cultural groups. Hence, our study aims at providing vocational scholars and practitioners with empirical guidance regarding the cross-cultural applicability of a multidimensional notion of networking behavior. Below, we develop our hypotheses regarding (a) the cross-cultural measurement invariance of networking behavior and (b) the association of networking dimensions with objective career success.

Networking as a Multifaceted Behavior

Networking refers to informal relationships (Wolff, Moser, & Grau, 2008) that individuals use to exchange work-related resources or favors such as task advice, information, or influence. These relationships represent a hybrid of universalistic and particularistic considerations, as individuals typically hold valuable resources due to their work role (the universalistic part) and are more inclined to informally provide these resources to those they know (the particularistic part). Trust and reciprocity are important mechanisms in this cooperative exchange, because trust facilitates this exchange of resources and the reciprocity norm secures that favors will be returned. Also, proficient “networkers” have more (nonredundant) contacts (Wolff & Moser, 2006) that provide the opportunity to tap into a variety of resources (Burt, 2004). As an important facilitator of career progress, networking behavior can be distinguished from the similar, but conceptually distinct construct of mentoring. Whereas mentoring refers to a single, specific relationship between a younger protégé and a senior mentor (Kram, 1985), networking denotes goal-directed interpersonal behaviors toward multiple informal contacts. Also, in contrast to networking, the exchange in mentoring relationships is typically asymmetric, such that the protégé receives career and psychosocial support, whereas the mentor attains loyalty and fulfillment of generative needs (Wolff et al., 2008). Supporting the notion that both constructs reflect related, yet distinct phenomena, empirical findings suggest that networking behavior serves as a mediator in the mentoring–career success relationship (Blickle, Witzki, & Schneider, 2009).

Accounting for calls to study career self-management strategies on more nuanced conceptual grounds (McArdle, Waters, Briscoe, & Hall, 2007), an increasingly influential stream of research on networking behavior has moved beyond a homogenous, unitary understanding of the construct toward a multifaceted notion capturing the distinct structural and functional manifestations of working individuals’ networking endeavors (Wolff et al., 2008). This theoretical shift has been acknowledged repeatedly in recent reviews of the pertaining literature, highlighting its scientific impact. For instance, Porter and Woo (2015) discuss how this “behavioral approach” provides incremental value to traditional organizational and managerial stances on networking by emphasizing “what people actually do when networking” (p. 1480) and that dimensions of networking behavior “may serve discrete functions that differentially relate to career outcomes” (p. 1480). As noted earlier, Gibson and colleagues (2014) have explicitly anchored their integrative theoretical model of networking around a multidimensional conceptualization of the construct, thereby considering both the structural (extra- vs. intraorganizational setting) and functional (building, maintaining, or using informal professional relationships) facets of networking. Paving the way for research in this vein, Wolff and colleagues have developed a multidimensional networking measure and empirically demonstrated its reliability and construct validity in German- and English-speaking employee samples (Gervorkian, 2011; Wolff & Moser, 2006; Wolff & Kim, 2012; Wolff et al., 2011).

Networking in Western and Non-Western Cultures

To our knowledge, three studies have examined networking behaviors in an Asian context. Utilizing samples from Hong Kong and the United States, Y. H. Cheung, Dougherty, and Herndon (2011) used items from three of Forret and Dougherty’s (2001) five networking dimensions (i.e., socializing, increasing internal visibility, and professional activities), stating that the other dimensions (i.e., maintaining external contacts and participating in community activities) were uncommon in the Asian context. In Malaysia, Rasdi, Garavan, and Ismail (2013) used items from all dimensions of Forret and Dougherty’s (2001) networking measure, and Wok and Hashim (2014) examined internal and external networking.

Although these studies provide valuable insights into networking in an Asian context, we argue that conclusions concerning cross-cultural comparability and generalizability are limited in two ways. First, these studies used ad hoc adaptations of a few items of some networking (sub)scales that do not sufficiently delineate the structural and functional components of networking behaviors. Thus, they do not provide full coverage of the consensual multidimensional definition of networking in the literature (Gibson et al., 2014). Second, we argue that a statistical examination of the cross-cultural measurement (non)invariance of the full range of networking dimensions is necessary to explicitly assess the cultural boundaries of a multidimensional conceptualization of networking behavior. As we will explain in the section on cross-cultural invariance, measurement equivalence is a crucial prerequisite for comparisons across cultures.

In light of this, our study contributes to the literature by applying a multidimensional networking scale in a Western and non-Western context, hence accounting for calls to explore the cross-cultural generalizability of vocational behavior (e.g., Heslin, 2005). We empirically assess the transferability of the scale (and thus, of a multifaceted notion of networking behavior) from a Western (i.e., German) to an Asian (i.e., Chinese) occupational context with a 2-fold approach. First, we conduct cross-group measurement invariance analyses to evaluate the extent of cross-cultural equivalence of the psychometric properties of the present scale. Second, we evaluate and cross-culturally compare the predictive validity of the multidimensional networking measure by examining its associations with objective career success (i.e., salary and promotions) via path- and relative weight analyses (RWAs).

Investigating whether the Western networking concept can be applied to the Chinese context requires a consideration of culture. Cultural norms and values shape affective and cognitive processes and also behavior. They therefore influence how people create, develop, and utilize interpersonal relationships at work. In terms of Hofstede’s (2003) dimensions of national culture, China and Germany differ in three key characteristics. The Chinese culture is characterized by (1) higher power distance, (2) lower individualism, and (3) lower uncertainty avoidance than the German culture. First, the dimension of power distance refers to an unequal distribution of prestige, wealth, and power that is more acceptable in China than in Germany. Thus, organizational hierarchies constitute higher boundaries in China, whereas direct and participative communication across hierarchies is more common in Germany. For example, Smith, Peterson, and Schwartz (2002) found that in a range of managerial situations, managers from high power distance societies thought that reliance on formal rules and procedures was more appropriate than managers from low power distance societies. As networking relies on informal contacts, networking across hierarchies might be harder and less common in China. Second, in individualistic cultures, cognition and behavior are governed by individuals’ goals, attitudes, and personal preferences, whereas members of collectivistic cultures place higher priority on group interests and harmonious relationships, which results in a stronger obligation to provide in-group members with favors. In numerous studies, Leung has shown that individuals place greater emphasis on the distinction between in- and out-groups in collectivistic cultures. For example, in collectivistic cultures, rewards are allocated in a more egalitarian manner in in-groups (Leung & Bond, 1984) and individuals behave more businesslike and exploitative toward out-groups (Leung & Iwawaki, 1988) and are less likely to interact and share their knowledge with out-groups (Lu, Leung, & Koch, 2006). Therefore, networking might have to take interests of in-groups into account, whereas individual goals may be more important for networking in Germany. Finally, Germany is also considered a culture with high uncertainty avoidance such that planning and rules are more valued than ambiguity and uncertainty, whereas people in the Chinese culture are more comfortable with uncertainty, as for example present in entrepreneurial contexts.

In sum, prior research has showed that networking per se is a relevant phenomenon in Asian working contexts, but a test of the cross-cultural applicability of a consensual multidimensional definition of networking is still pending. Models of cultural differences suggest that cultural values and norms may influence networking to the extent that, for example, organizational in-group networks may be more densely knit in China due to higher collectivism. Below, we hypothesize on the extent of cross-cultural measurement invariance of the present multidimensional networking scale.

Cross-Cultural Measurement Invariance

To overcome ethnocentric bias and to enhance the international theoretical and practical impact of vocational research, investigations of the equivalence of important psychological constructs across different cultural contexts are necessary (Gelfand, Erez, & Aycan, 2007; Shiraev & Levy, 2013). In this study, we use a cross-cultural sample of white-collar employees working for a global player in high-tech manufacturing to investigate the cultural boundaries of a multidimensional notion of networking behavior. We anticipate some degree of (i.e., configural and metric) cross-group invariance of the present multidimensional networking scale in light of the high degree of globalization that characterizes the work environments of our participants.

Measurement invariance is a statistical concept representing a continuum of increasing similarity. Using the confirmatory factor analysis (CFA) framework, scholars posit three general models that successively allow more meaningful comparisons across groups (Steinmetz, Schmidt, Tina-Booh, Wieczorek, & Schwartz, 2009; Milfont & Fisher, 2010). Cross-group configural invariance implies the same number of factors and the same pattern of fixed and nonfixed parameters of the respective measurement model. In this study, configural invariance indicates that members of each group conceptualize the theoretically anticipated dimensions of networking behavior similarly and that items belong to the same latent construct(s) in both cultures (e.g., asking for advice is a manifestation of using, but not building contacts).

Metric invariance refers to equally constrained factor loadings, reflecting a strict condition for construct comparability across groups. According to Steinmetz, Schmidt, Tina-Booh, Wieczorek, and Schwartz (2009), metric invariance constitutes “a prerequisite for inferring that the construct has the same meaning” (p. 603) and is necessary for meaningful quantitative comparisons. If metric invariance holds, relationships between constructs (e.g., the relationship between networking and career outcomes) can be compared meaningfully across cultures. On the basis of our considerations above, we assume the multidimensional networking scale will exhibit this level of cross-cultural measurement invariance in the present samples of white-collar employees in an MNC.

By contrast, we argue that it is unreasonable to expect scalar invariance, which refers to identical item means across cultures and represents a necessary condition to meaningfully compare scale means across cultures (Steinmetz et al., 2009). This is in line with findings from other scales assessing important vocational constructs such as career adaptability (Savickas & Porfeli, 2012). For example, slight differences in the meaning of scale anchor translations might result in mean score differences. Further, cultural differences, for example in the focus on in-group contacts and strong ties, might affect how often employees engage in specific networking behaviors. High collectivism may discourage Chinese individuals from engaging in networking behaviors across organizational boundaries, which might also give rise to item mean differences. In sum, we propose the following hypotheses regarding the invariance of the multidimensional networking scale across Germany and China.
Hypothesis 1a: The present multidimensional networking scale will exhibit configural invariance across groups of German and Chinese white-collar employees employed by the same MNC.

Hypothesis 1b: The present multidimensional networking scale will exhibit metric invariance across groups of German and Chinese white-collar employees employed by the same MNC.

Networking and Objective Career Success

Networking behavior is related to an array of important outcomes variables including visibility, power, career success, performance, and turnover (Gibson et al., 2014), but also well-being (Volmer & Wolff, 2016). The goal of this study is to explore the cross-cultural transferability of a multidimensional networking scale not only by examining its psychometric invariance but also its predictive validity. Regarding the choice for appropriate criterion variables, we opted for objective career success (i.e., salary and promotions). According to Heslin (2005), objective career success outcomes are defined as “directly observable, measurable, and verifiable by an impartial third party” (p. 114). Kraimer, Seibert, and Astrove (2015) recently reiterated this definition and specifically regard salary and promotions as extrinsic success outcomes due to their instrumental value. In light of this sentiment as well as research showing the meaningfulness of these outcome variables in a Chinese context (Bian, Huang, & Zhang, 2015; Gong & Chang, 2008; Tu, Forret, & Sullivan, 2006), we chose to assess these outcome variables.

Studies in Western and Asian cultures alike support a positive association between networking and objective career success (Ng & Feldman, 2014; Rasdi et al., 2013; Wok & Hashim, 2014). In their review of the networking literature, Gibson et al. (2014) identified two mechanisms thought to account for this relationship, namely, access to strategic information and social capital. The former may, for instance, facilitate salary negotiations and boost job performance. Social capital, defined as characteristics of social structures that facilitate certain actions (Coleman, 1988), captures a higher level construct than specific networking behaviors, because it refers to one’s overall individual network structure (Gibson et al., 2014). Social capital can increase a person’s visibility, reputation, or power, all of which are thought to affect career outcomes (Wolff et al., 2008).

As discussed earlier, a multidimensional notion of networking behavior entails the prospect to examine differential relations with relevant outcome variables, and research in this vein is slowly beginning to accumulate (e.g., McCallum et al., 2014; Wolff & Moser, 2010). However, theoretical advances that provide guidance regarding such differential predictions in light of cross-cultural differences have yet to be achieved. Hence, we pursue an exploratory approach to examine the pattern of associations between dimensions of networking behavior and objective career success in Germany and China in this study. The decision for this approach is also driven by the purpose of this study which centers on a preliminary examination of the cross-cultural equivalence of a consensual multifaceted conceptualization of networking behavior. For this purpose, we broadly hypothesize a positive link between networking and objective career success and conduct detailed exploratory analyses of the differential predictive power of networking dimensions within and between both cultural groups.
Hypothesis 2: Networking behavior will be positively related to objective career success (i.e., salary and promotions) in Germany and China.

Method

Procedure and Participants

We conducted a two-wave, web-based field study with a time lag of 2 months. Data were collected from Chinese and German white-collar employees who were employed by the same international manufacturing company, a global provider of semiconductors. With the help of two directors of the company’s HR department, potential participants were contacted via e-mail and invited to take part in a study on networking behavior. In order to ensure compliance, an anonymized feedback report was offered after study completion, outlining the purpose, theoretical background, and key results of the study. Data were available for 118 German and 130 Chinese employees at Time 1 (T1). After 2 months, participants completed the Time 2 (T2) questionnaire. Data were available for 107 German and 93 Chinese employees, resulting in a satisfactory response rate of 90.68% and 71.57% in the German and Chinese sample, respectively. Since data from one German participant could not be matched and had to be excluded from further analyses, the final German sample consisted of 106 participants with a mean age of 44.7 years (SD = 11.29). Of all, 37.7% of the German participants were female. Ten Chinese participants were excluded over the course of data matching. As a result, the final Chinese sample included 83 employees who, on average, were 37.2 years old (SD = 6.82). Of all, 56.6% of the Chinese participants were female. On average, German and Chinese participants reported an organizational tenure of 14.26 (SD = 10.50) and 10.05 years (SD = 6.22), respectively.

Measures

Following Banville, Desrosiers, and Genet-Volet’s (2000) recommendations for cross-cultural research, a double translation/back-translation procedure was used to translate the German scale into traditional Chinese. Chinese participants were additionally provided with the English version of the items (Wolff, Schneider-Rahm, & Forret, 2011). The German and Chinese version of the multidimensional networking scale can be obtained upon request.

Networking behavior

Employees’ habitual networking behavior was assessed with Wolff and Moser’s (2006) multidimensional 44-item measure assessing six subscales which result from crossing the theoretically derived functional and structural facets of networking. The functional facet denotes to building, maintaining, and using informal relationships at work, whereas the structural facet captures whether networking is pursued inside or outside of one’s organization (i.e., internal vs. external networking). On 4-point Likert-type scales (1 = never/very seldom and 4 = always/very often), participants indicated how often they engaged in specific networking behaviors. Sample items for the six subscales were “I use company events to make new contacts” (building internal contacts; 6 items, Cronbach’s α = .75/.84 in the German/Chinese sample, respectively), “I catch up with colleagues from other departments about what they are working on” (maintaining internal contacts; 8 items, α = .79/.85), “I use my contacts with colleagues in other departments in order to get confidential advice in business matters” (using internal contacts; 8 items, α = .80/.89), “I accept invitations to official functions or festivities out of professional interest” (building external contacts; 7 items, α = .81/.88), “I ask others to give my regards to business acquaintances outside of our company” (maintaining external contacts; 7 items, α = .80/.88), and “I exchange professional tips and hints with acquaintances from other organizations” (using external contacts; 8 items, α = .83/.93). Several studies provide evidence for the construct validity of this scale (e.g., Wolff & Moser, 2006; Wolff & Kim, 2012).

Objective career success

In line with extant career research (Ng, Eby, Sorensen, & Feldman, 2005) and contemporary typologies of career success (Kraimer, Seibert, & Astrove, 2015), we assessed salary and promotions as objective career success outcomes. Specifically, participants were provided with information on currency translation rates to indicate their monthly salary in US dollars (1 = up to US$999, 2 = US$1,000–US$1,999, 3 = US$2,000–US$2,999,…, 11 = US$10.000–US$10.999, and 12 = more than US$11.000). Participants reported how many promotions they had received since joining the company. A promotion was defined as “any upwards movement in the organizational hierarchy and/or any job change that involves substantial increase in responsibilities, power and salary” (Bozionelos & Wang, 2006, p. 1538).

Control variables

To rule out alternative explanations, several sociodemographic control variables were assessed. In particular, we controlled for participants’ age, sex (1 = female, 2 = male), education (1 = none, 2 = high school or equivalent, 3 = vocational/technical secondary school/Zhongzhuan, 4 = some college, 5 = university level degree, 6 = doctoral degree), and organizational tenure. Meta-analytic findings show positive associations of these variables with both salary and promotions (Ng et al., 2005).

Statistical Analyses

We used Mplus (Version 7.2; Muthén & Muthén, 2015) for all statistical analyses described below. Cross-cultural measurement invariance was assessed with two stages of CFAs. First, we examined the dimensionality of networking behavior separately in each cultural group, thereby testing the configural invariance of the multidimensional networking scale (G. W. Cheung & Rensvold, 1999; Milfont & Fisher, 2010). Second, multigroup CFA was used to further inspect measurement invariance across both samples. Following the approach pursued in previous research (Wolff & Moser, 2009), parcels were created according to the parceling scheme employed by Wolff and Moser (2006) in their validation study of the present networking scale. Thus, we ensured continuity and comparability with prior research by favoring a meso-level of (dis)aggregation in constructing our latent measurement models (Little, Cunningham, Shahar, & Widaman, 2002). To evaluate model fit, the comparative fit index (CFI), the Tucker–Lewis index (TLI), and the root mean square error of approximation (RMSEA) were assessed. Cutoff values were .90 for CFI and TLI and .08 for RMSEA (Bentler & Bonett, 1980; Byrne, 2010). To deal with missing values and nonnormality, robust maximum likelihood estimation was used. Latent factors were scaled by means of the marker variable method (Little, Preacher, Selig, & Card, 2007). Nested models were compared by conducting Satorra–Bentler (SB) scaled difference χ² tests (Bryant & Satorra, 2012; Satorra & Bentler, 2001). Further, we conducted multigroup path analysis with observed variables to investigate time-lagged associations of networking dimensions with both objective career success outcomes simultaneously in both cultural groups.

Results

Tables 1 and 2 show the means, standard deviations, and intercorrelations of study variables observed in the German and Chinese sample, respectively.

Table 1.
Summary of Intercorrelations, Means, and Standard Deviations for Study Variables as a Function of Cultural Group.

Variable 1 2 3 4 5 6 7 8 9 10 11 12 M SD

1. Age — .07 .21 .77 .32 .32 .29 −.02 −.03 −.03 .50 .47 37.2 6.82

2. Sex^a .23 — .22 .03 .06 .03 .01 .19 .17 −.01 .14* .05 1.43 0.50

3. Education^b −.16 .30 — .17 .10 .15 .14 .13 .01 .09 .13 .02 4.75 0.46

4. Organizational tenure .56 .20* −.13 — .34 .32 .36 .03 .05 .01 .52 .54** 10.05 6.22

5. Building internal contacts .03 .17 .24* −.05 — .75 .60 .18 .14 −.06 .65 .55 2.62 0.56

6. Maintaining internal contacts .28 .28 .25* .06 .30 — .69 .09 .18 .06 .57 .50 2.84 0.54

7. Using internal contacts −.03 .18 .26** −.07 .20* .50** — .12 .25* .16 .55 .62 2.68 0.66

8. Building external contacts .23* .37 .29 .22* .38 .42 .46 — .65 .53 .06 .02 1.88 0.44

9. Maintaining external contacts .27 .37** .25* .22* .29 .47 .55 .70 — .67** .04 .14 1.74 0.42

10. Using external contacts .20* .26 .15 .18 .26 .41 .52 .63 .77 — −.09 −.02 1.72 0.45

11. Salary .38 .46 .35 .30 .18 .43** .23* .48 .53 .49 — .76 4.06 2.31

12. Promotions .44 .38 .08 .36 .29 .31** .20* .50 .54 .43 .61 — 2.40 1.79

M 44.70 1.62 4.26 14.26 2.36 2.63 2.38 2.15 2.14 2.09 7.12 3.78

SD 11.28 0.49 1.02 10.50 0.50 0.48 0.49 0.66 0.64 0.79 0.74 2.58

Note. Intercorrelations for Chinese participants (n = 83) are presented above the diagonal, and intercorrelations for German participants (n = 106) are presented below the diagonal. Means and standard deviations for Chinese participants are presented in the vertical columns, and means and standard deviations for German participants are presented in the horizontal rows. Internal consistencies are reported in the Method section.

^aSex is coded as 1 = female, 2 = male. ^bEducation is coded as 1 = none, 2 = high school or equivalent, 3 = vocational/technical secondary school/Zhongzhuan, 4 = some college, 5 = university degree, 6 = doctoral degree.

*p < .05. **p < .01.

Table 2.
Fit of Latent Measurement Models to Assess Dimensionality of Networking Behavior in Germany and China.

Models by Sample SB-χ² df CFI TLI RMSEA

German sample

One-factor model 635.08 189 .69 .66 .143

Two-factor model^a (internal, external) 525.82 188 .77 .74 .125

Three-factor model^a (building, maintaining, using) 464.08 186 .81 .79 .113

Five-factor model^b 259.26 164 .92 .90 .078

Six-factor model^a 227.32 174 .97 .96 .049

Chinese sample

One-factor model 767.28 189 .48 .42 .186

Two-factor model^a (internal, external) 432.12 188 .78 .76 .121

Three-factor model^a (building, maintaining, using) 668.03 186 .57 .52 .170

Five-factor model^b 265.40 164 .90 .87 .089

Six-factor model^a 260.58 174 .92 .90 .077

Note. N _German = 106; N _Chinese = 83. All SB-χ² values were significant at p < .05. SB-χ² = Satorra–Bentler scaled chi-square; CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; NW = networking behavior.

^aLatent factors were correlated. ^bThe five-factor model incorporated the following latent dimensions: internal and external NW (i.e., two factors reflecting the structural facet of networking) and building, maintaining, and using contacts (i.e., three factors reflecting the functional facet of networking); factors were allowed to correlate within facets (e.g., internal and external NW were correlated), but not between facets (e.g., using contacts was neither correlated with internal and external NW). Thus, this model incorporated double loadings for each NW indicator.

Cross-Cultural Measurement Invariance

Configural invariance

Table 2 shows the model fit indices obtained for a series of competing latent measurement models tested separately in the German and Chinese subsample. The theoretically anticipated correlated six-factor model adequately fits the data in both cultural groups. Each of the competing models (i.e., a one-factor model, models in which either the structural or functional facets of networking were aggregated, and an alternative five-factor model) yielded fit indices below the cutoff values, indicating inferior model fit in comparison to the theoretical six-factor model (see Table 2). Hence, initial evidence for the cross-cultural configural invariance of multidimensional networking behavior was found, supporting Hypothesis 1a.

Metric invariance

Subsequently, cross-group invariance analyses proceeded with multigroup CFA. As a first step, the correlated six-factor model was specified simultaneously in both samples. This unconstrained model yielded good fit, SB-χ²(348) = 477.83, p < .001, CFI = .95, TLI = .93, RMSEA = .063, again confirming the configural invariance of multidimensional networking across cultural groups. Next, factor loadings of the same indicators (i.e., parcels) were constrained to be equal across groups to examine metric invariance. Results indicated that the metric model (see Table 3) did not differ significantly in fit from the previously specified unconstrained (i.e., configural) model ΔSB-χ²(15) = 11.16, p = .74. Thus, we found support for Hypothesis 1b.

Table 3.
Fit of Multigroup Measurement Models to Test Cross-Cultural Measurement Invariance of Networking Behavior.

Model SB-χ² df CFI TLI RMSEA (90% CI) ΔSB-χ² Δdf

Configural model 477.83* 348 .95 .93 .063 [.048, .076] — —

Metric model 488.16* 363 .95 .94 .060 [.046, .074] 11.16 15

Full scalar model 616.24* 384 .90 .89 .080 [.068, .092] 132.61* 21

Partial scalar model^a 565.40* 381 .92 .92 .072 [.059, .084] 80.53* 18

Note. N _German = 106; N _Chinese = 83. SB-χ² = Satorra–Bentler scaled chi-square; CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; CI = confidence interval.

^aThe partial scalar model was compared against the metric model; it incorporated relaxed equality constraints on three indicator intercepts in both groups.

***p < .001.

Scalar invariance

To assess the presence of scalar invariance, equality constraints were placed on all item intercepts in both cultural groups. This model fits the data significantly worse than the metric model, ΔSB-χ²(21) = 132.61, p < .001. An examination of the modification indices for the model specifying full scalar invariance revealed three substantial values. Testing a subsequent model in which the intercepts of the three respective indicators were freely estimated still indicated a significantly worse model fit compared to the metric model, ΔSB-χ²(18) = 80.53, p < .001. Thus, results showed the scalar noninvariance of the multidimensional networking scale across both groups.

Path Analysis

Table 4 shows the results from specifying a multivariate multigroup path analytic model in which control variables and networking dimensions predicted German and Chinese employees’ salary and promotions simultaneously. Positively predicting promotions, the only intraorganizational networking dimension significantly related to Germans’ objective career success was building internal contacts (β = .15, SE = 0.07, p = .049). Further, two different functional forms of German employees’ extraorganizational networking behavior significantly and differentially predicted objective career success: Maintaining external contacts positively predicted promotions (β = .32, SE = 0.14, p = .027), whereas using external contacts positively predicted salary (β = .25, SE = 0.12, p = .045). By contrast, extraorganizational networking emerged as entirely unrelated to objective career success in the Chinese group. Conversely, building and using internal contacts significantly predicted both salary (β = .38, SE = 0.10, p < .001 and β = .22, SE = 0.11, p = .043, respectively) and promotions (β = .25, SE = 0.12, p = .033 and β = .41, SE = 0.11, p < .001, respectively) among Chinese participants. In sum, path analytic results support Hypothesis 2.

Table 4.
Multigroup Path Analysis Predicting German and Chinese Individuals’ Objective Career Success From Dimensions of Networking Behavior.

Predictor German Group Chinese Group

Salary Promotions Salary Promotions

Control variables

Age .17* (0.08) .22* (0.10) .16 (0.11) .11 (0.11)

Sex .19* (0.08) .14 (0.09) .10 (0.08) .04 (0.08)

Education .26** (0.09) −.05 (0.12) −.03 (0.06) −.11 (0.08)

Organizational tenure .09 (0.08) .11 (0.11) .18 (0.11) .26* (0.11)

Networking behavior

Building internal contacts −.05 (0.07) .15* (0.07) .38*** (0.10) .25* (0.12)

Maintaining internal contacts .16 (0.10) .00 (0.08) .04 (0.10) −.08 (0.12)

Using internal contacts −.15 (0.10) −.07 (0.10) .22* (0.11) .41*** (0.11)

Building external contacts .04 (0.10) .14 (0.15) .02 (0.10) −.10 (0.11)

Maintaining external contacts .14 (0.14) .32* (0.14) −.06 (0.13) .12 (0.12)

Using external contacts .25* (0.12) .01 (0.11) −.06 (0.11) −.08 (0.12)

R ² .51* .45* .58* 56*

N 106 83

Note. Standardized robust maximum likelihood parameter estimates. Standard errors appear in parentheses. All estimates result from a single (multivariate, multigroup) path analytic model with observed variables.

*p < .05. ***p < .001.

RWA

Using Tonidandel and LeBreton’s (2015) online tool RWA-Web, we conducted RWA to investigate patterns of relative importance of our predictor variables for objective career success, both within and between both cultural groups. RWA is a powerful method to “decompose the total variance predicted in a regression model (R ²) into weights that accurately reflect the proportional contribution of [multiple, often correlated] predictor variables” (Tonidandel & LeBreton, 2015, p. 207), thus serving as a valuable supplement to traditional regression analysis. In light of the observed interrelations among subdimensions of networking behavior as well as control variables (i.e., age, sex, education, and organizational tenure), RWA also allowed us to appropriately address possible issues of multicollinearity (Tondidandel & LeBreton, 2011, 2015).

Within-group pattern of results

Tables 5 and 6 show the RWA results obtained for the criterion variables salary and promotions, respectively. In both tables, results from RWA conducted separately in the German and Chinese group are summarized. Multiple parameters are reported including (a) raw relative weights (RWs) as a measure of relative effect size, (b) rescaled RWs indicative of the percentage of the total explained variance in the criterion variable attributable to the respective predictor variable, (c) 95% confidence intervals (CIs) for individual raw RWs, and (d) CIs testing the statistical significance of RWs. All CIs were based on bootstrapping with 10,000 replications; bias corrected and accelerated CIs were used on grounds of their superior coverage accuracy, as recommended by Tonidandel, LeBreton, and Johnson (2009). Notably, the regular CIs around RWs (c) provide information on the sampling distribution of weight estimates, whereas meaningfully different CIs (d) were used to assess statistical significance (Tonidandel, LeBreton, & Johnson, 2009). Results demonstrate that weighted linear combinations of our predictor variables explained roughly half of the total variance in both objective career success outcomes in both cultural groups (R ² = .51–.57 and R ² = .45–.54 for salary and promotions, respectively), closely mirroring R ² values obtained in the previously conducted path analysis (see Table 4).

Table 5.
Results From Relative Weight Analyses Predicting Salary in the German and Chinese Sample.

Predictor Variable German Sample Chinese Sample

Relative Weight Regular CI Statistical Significance CI^a Relative Weight Regular CI Statistical Significance CI^a

Raw Rescaled (%) Raw Rescaled (%)

Age .059* 11.63 [0.02, 0.12] [0.01, 0.13] .086* 14.95 [0.03, 0.18] [0.02, 0.20]

Sex .080* 15.89 [0.02, 0.16] [0.02, 0.17] .012 2.10 [0.00, 0.06] [−0.01, 0.09]

Education .071* 14.14 [0.01, 0.16] [0.01, 0.17] .003 0.47 [0.00, 0.01] [−0.02, 0.03]

Tenure .036 7.04 [0.01, 0.09] [−0.00, 0.11] .093* 16.24 [0.04, 0.16] [0.03, 0.17]

Building internal contacts .010 1.04 [0.00, 0.01] [−0.02, 0.03] .172* 29.96 [0.10, 0.26] [0.10, 0.27]

Maintaining internal contacts .054* 10.67 [0.01, 0.13] [0.01, 0.15] .094* 16.39 [0.05, 0.14] [0.05, 0.16]

Using internal contacts .012 2.35 [0.00, 0.02] [−0.02, 0.03] .100* 17.37 [0.03, 0.18] [0.04, 0.19]

Building external contacts .046* 9.09 [0.02, 0.09] [0.01, 0.10] .003 0.54 [0.00, 0.00] [−0.03, 0.04]

Maintaining external contacts .067* 13.29 [0.03, 0.12] [0.02, 0.13] .003 0.56 [0.00, 0.00] [−0.03, 0.02]

Using external contacts .075* 14.86 [0.03, 0.14] [0.02, 0.15] .008 1.42 [0.00, 0.03] [−0.02, 0.06]

R ² .51 .57

Note. CI = 95% confidence interval. Rescaled relative weights indicate the relative contribution of each individual predictor variable to the total R ² (in %).

^aTo evaluate the statistical significance of relative weights, CIs were estimated following the approach described by Tonidandel, LeBreton, and Johnson (2009). A weight is considered significant, if this particular CI excludes zero.

Relative weights are statistically significant from zero (i.e., the statistical significance CI estimated for the respective relative weight excludes zero).

Table 6.
Results From Relative Weight Analyses Predicting Promotions in the German and Chinese Sample.

Predictor Variable German Sample Chinese Sample

Relative Weight Regular CI Statistical Significance CI^a Relative Weight Regular CI Statistical Significance CI^a

Raw Rescaled (%) Raw Rescaled (%)

Age .083 18.40 [0.03, 0.16] [0.03, 0.17] .075* 14.01 [0.03, 0.15] [0.02, 0.16]

Sex .049 10.89 [0.01, 0.12] [0.00, 0.13] .001 0.28 [0.00, 0.00] [−0.02, 0.04]

Education .004 0.92 [0.00, 0.00] [−0.04, 0.03] .006 1.11 [0.00, 0.03] [−0.01, 0.07]

Tenure .050 11.18 [0.01, 0.14] [0.00, 0.15] .119* 22.10 [0.05, 0.21] [0.05, 0.22]

Building internal contacts .036 8.04 [0.01, 0.10] [0.00, 0.12] .108* 20.15 [0.04, 0.19] [0.04, 0.20]

Maintaining internal contacts .018 4.10 [0.01, 0.05] [−0.00, 0.07] .065* 12.04 [0.02, 0.11] [0.02, 0.13]

Using internal contacts .010 2.25 [0.00, 0.02] [−0.01, 0.03] .142* 26.39 [0.06, 0.25] [0.06, 0.26]

Building external contacts .065* 14.49 [0.02, 0.16] [0.02, 0.17] .005 0.84 [0.00, 0.01] [−0.02, 0.04]

Maintaining external contacts .090* 20.02 [0.04, 0.17] [0.04, 0.17] .012 2.15 [0.00, 0.04] [−0.01, 0.06]

Using external contacts .044* 9.72 [0.02, 0.08] [0.01, 0.09] .005 0.93 [0.00, 0.01] [−0.02, 0.05]

R ² .45 .54

Note. CI = 95% confidence intervals. Rescaled relative weights indicate the relative contribution of each individual predictor variable to the total R ² (in %).

^aTo evaluate the statistical significance of relative weights, CIs were estimated following the approach described by Tonidandel, LeBreton, and Johnson (2009). A weight is considered significant, if this particular CI excludes zero.

Relative weights are statistically significant from zero (i.e., the statistical significance CI estimated for the respective relative weight excludes zero).

An examination of raw RWs indicates a largely opposite pattern for the German and Chinese group. In the Chinese sample, maintaining, building, and* using internal contacts explained a significant amount of variance in salary (RWs = .17, .09, and .10, respectively) and promotions (RWs = .11, .07, and .14), as all of the associated 95% CIs for tests of statistical significance excluded zero. Notably, maintaining internal contacts had emerged as unrelated to salary and promotions in our path analysis, whereas RWA showed that all dimensions of intraorganizational networking including the maintenance dimension significantly predicted objective career success. This “lack of concordance in the significance of the regression coefficients and the relative weights is not uncommon” (Tonidandel & Breton, 2015, p. 215) and in this case indicates that all dimensions of intraorganizational networking explain nontrivial variance in salary and promotions, but due to intercorrelations among the predictor variables maintaining internal contacts explains little unique, incremental variance. RWA results for the Chinese group further demonstrate that extraorganizational networking was unrelated to objective career success, in line with path analytic results.

By contrast, the pattern observed in the German group suggests that all functional dimensions of extraorganizational networking significantly predict variance in salary (RWs = .05, .07, and .08, for building, using, and maintaining external contacts, respectively) and promotions (RWs = .07, .09, and .04, respectively). Previous path analytic results had indicated that only using external contacts significantly predicted Germans’ salary and that only maintaining external contacts predicted promotions (see Table 4). Again, RWA results revealed that all subdimensions of extraorganizational networking explained nontrivial variance. Conversely, only the RW of maintaining internal contacts for salary was statistically significant (RW = .05), whereas none of the (other) intraorganizational networking behaviors significantly predicted objective career success (see Tables 5 and 6). Although the regression weight of building internal contacts predicting promotions had reached significance in the path analysis, the respective RW (.04) was not significant (only marginally) as the corresponding 95% CI included zero, [0.00, 0.12]. Briefly, RWA results supported Hypothesis 2 regarding our overall assumption of a positive association of networking behavior with objective career success.

Between-group comparisons

To assess whether the magnitude of the RW estimates for our predictor variables differ significantly between both cultural groups, group difference tests for RWs were conducted (Johnson, 2004; Tonidandel et al., 2009). If the resulting CIs estimated for each predictor variable exclude zero, the RW of the pertaining predictor differs significantly between groups (grouping variable: 0 = Chinese, 1 = German). We conducted this analysis separately for each career success outcome. With regard to dimensions of networking behavior, results suggest a greater relative importance of intraorganizational networking for objective career success in China. Specifically, the RW of using internal contacts was shown to be significantly higher for Chinese participants’ salary and promotions than for Germans’, 95% statistical significance CIs [0.03, 1.71] and [0.05, 2.45], respectively. Moreover, the RW of building internal contacts predicting salary was significantly higher in the Chinese group, 95% CI [0.10, 2.68].

Discussion

The purpose of this study was twofold. First, we examined the degree to which an established multidimensional networking scale, so far exclusively applied to Western working populations, exhibited cross-cultural measurement equivalence across a German and Chinese sample of working individuals employed by the same MNC. Results indicate the presence of configural and metric, but not scalar measurement invariance. Notably, mean scores of extraorganizational networking as well as correlations between intra- and extraorganizational networking were somewhat low in the Chinese group. A possible explanation for these findings may be the stronger distinction between in-groups and out-groups in China. Nonetheless, we cannot fully rule out a possibly decontextualized application of the concept of extraorganizational networking behavior among Chinese individuals. Second, we were interested in determining the predictive validity of the multidimensional networking scale in both cultural groups by assessing associations of networking behaviors with objective career success (i.e., salary and number of promotions, assessed 2 months later). In line with theoretical models of overall networking, specific networking dimensions were indeed significantly and positively related to self-reported objective career success, depending on which cultural group and which success outcome were considered. We provide an in-depth discussion of our findings below.

Theoretical Contributions

A multifaceted behavioral notion of networking behavior as posited by Wolff and colleagues (e.g., Wolff et al., 2008; Wolff & Moser, 2009) has been acknowledged as “the most comprehensive coverage of the networking construct space to date” (Porter & Woo, 2015, p. 1480). Yet, hardly any research has so far assessed the cross-cultural boundaries of this conceptualization (see Wolff et al., 2011, for an exception). Hence, our findings broaden the scope of the literature by demonstrating that networking behavior can not only be conceived of as functionally multifaceted in Western but also in an Asian (i.e., Chinese) context. Evidence for the cross-cultural configural and metric measurement invariance of Wolff and colleagues’ multidimensional networking scale (Wolff & Moser, 2006; Wolff et al., 2011) shows that German and Chinese employees conceptualize networking behavior equivalently and respond to the respective items similarly. Our finding of scalar noninvariance further implies that (latent) mean comparisons across both cultural groups are not advisable.

Moreover, our findings regarding the predictive power of dimensions of networking behavior for objective career success merit closer scrutiny. In sum, our results confirmed the assumption of a positive association of networking behavior(s) with objective career success, in line with previous research (Ng & Feldman, 2014; Rasdi et al., 2013; Wok & Hashim, 2014). Regarding German employees’ salary, we found that all networking behaviors except building and using internal contacts explained significant variance in salary. In comparison, Wolff and Moser (2009) found that, in a sample of German employees, all six networking dimensions concurrently explained significant nontrivial variance in salary. Our findings thus highlight the importance of external networking for salary attainment in Germany and suggest that effects of intraorganizational networking might depend on additional contextual variables. For example, historical differences (e.g., economic cycles) between studies or specifics of the MNC we examined might account for the slightly divergent findings.

In contrast, only intraorganizational, but not extraorganizational networking was significantly related to salary and promotions in the Chinese sample. Moreover, the relative importance of intraorganizational networking (i.e., building and using contacts) for predicting objective career success appeared to be greater in the Chinese group. We suggest that cultural differences in collectivism and power distance between China and Germany may be responsible for the differences in the importance of networking. Given the stronger focus on strong ties and in-group members in collectivistic societies like China, intraorganizational networking appears to be especially important for career success. Presuming that it is more difficult to network across hierarchies in high power distance cultures, overcoming this barrier by means of intraorganizational networking might yield even more success.

Another important implication of our findings is that theoretical models of networking should not only acknowledge the multidimensionality of networking behavior but also address the largely neglected issue of relative contributions of specific networking behaviors to the emergence of vocational outcomes. Do cultural, temporal (e.g., proximity patterns of mobility outcomes; Wolff & Moser, 2010), and contextual boundary conditions (e.g., HR practices, organizational culture) affect the relative importance of different forms of networking behavior for key outcomes? Thorough theorizing and more empirical insights into this question may greatly enhance the accuracy of our prediction regarding the differential effects of functional and structural facets of networking.

Limitations

Several study limitations need to be addressed. First, we modeled no change in our outcome variables because we did not opt for repeated assessments of salary and promotions across more frequent measurement occasions. The purpose of the current research, however, was not to elucidate the possible role of networking in long-term changes of objective career success. Rather, we were interested in establishing initial evidence for the predictive validity of the present networking scale in a Chinese sample of working individuals. A second limitation of our study was the narrow focus on objective success measures which we considered to be culturally invariant to the extent that salary and promotions are likely to be comparably valued and meaningful in both our subsamples. We call for more qualitative and quantitative efforts to determine whether and how cultural differences impact the conceptualization of self- and other-referent subjective career success outcomes (e.g., career satisfaction). Pertaining insights would allow future studies to assess multiple career success outcomes of networking on a cross-cultural level of analysis. Third, the correlative nature of our findings regarding associations between networking behaviors and career success implies that the possibility of reverse caution (e.g., more successful employees may engage in more frequent networking) cannot be ruled out. Multiwave, change-oriented investigations of the underlying developmental patterns and possible directions of causality are needed to elucidate this issue. Finally, it remains an open question whether our findings can be generalized to less globalized Chinese/Asian working environments.

Practical Implications

Our findings have implications for employees as well as organizations. The metric invariance of the networking construct implies that the concepts of building, maintaining, and using internal or external contacts is meaningful in both cultures and can thus be applied in strategic considerations of networking, for example, in career planning and counseling. Employees and practitioners should nonetheless be cognizant of the low prevalence of extraorganizational networking in China and the greater relative importance of building and using internal contacts for predicting objective career success. Trainings for expatriates, for example, should emphasize that learned contingencies between networking styles and career outcomes must be reconsidered and may require cultural adaptation. Further, globally acting organizations should be aware of possible cultural influences on networking behaviors and consider adjustments of HR instruments (e.g., incentives for specific networking styles).

Future Research

We call for more indigenous and cross-cultural research to unravel the nature of networking behavior in non-Western cultures. While our study provides initial insight into the cross-culturally invariant functional facets of networking inside one’s organization (i.e., building, maintaining, and using informal contacts), it is too early to derive a definitive answer as to whether networking is explicitly pursued as a universal strategy to self-direct one’s career (e.g., attain career goals, regulate career insecurity, cope with anticipated or experienced occupational transitions). Tied to this sentiment is the need for future research to thoroughly assess whether individual-level cultural differences (e.g., values, norms, beliefs) affect the enactment and functioning of networking behavior. Gibson et al.’s (2014) integrative networking theory organizes possible antecedents of (overall) networking behavior into three categories: organizational (e.g., organizational culture), job characteristics, and individual antecedents (e.g., personality, attitudinal variables, self-esteem), some of which may be subject to cultural differences and in turn affect the likelihood of engaging in specific forms of networking. Despite recent advances in the networking literature, however, findings on differential antecedents of structural and functional dimensions of networking are still scarce (see Wolff & Kim, 2012, for an exception), rendering it difficult to derive comprehensive explanations regarding our finding of low extraorganizational networking among Chinese employees. Consequently, we call for more exploratory field studies in this vein. A notable example for this approach is Kim’s (2013) semistandardized interview study of managers’ networking. This qualitative study adopted, at least partially, a multidimensional lens on networking behavior by distinguishing differential antecedents (enablers vs. constraints) and outcomes of building and maintaining (but not using) informal relationships. However, the distinction between intra- and extraorganizational networking was neglected. We nevertheless consider Kim’s (2013) methodology, perhaps complemented by (1) quantitative methods, (2) a rigorous cross-cultural approach, and (3) a complete (rather than abridged) multifaceted conceptualization of networking behavior, to be a promising avenue for future research.

Conclusion

Our study of white-collar workers employed by the same MNC in Germany and China indicated that the functional but not necessarily the structural facet of a consensual multidimensional conceptualization of networking behavior was applicable in a Chinese context. Specifically, the distinction between multiple functional dimensions of networking (i.e., building, maintaining, and using informal professional relationships) was cross-culturally applicable, whereas the irrelevance of extraorganizational networking behaviors for predicting objective career success in China suggests cultural differences in the impact of networking behavior. A complex pattern of differential predictive power of networking dimensions for salary and promotions emerged within and between cultural groups. A key cross-cultural finding was the greater magnitude of relative effect sizes observed for intraorganizational networking (i.e., the using and building dimensions) in China compared to Germany. We hope to stimulate future inquiry into the cultural, temporal, and contextual boundary conditions of the dimensionality and consequences of networking behavior.

Variable	1	2	3	4	5	6	7	8	9	10	11	12	M	SD
1. Age	—	.07	.21	.77**	.32**	.32**	.29**	−.02	−.03	−.03	.50**	.47**	37.2	6.82
2. Sex^a	.23*	—	.22*	.03	.06	.03	.01	.19	.17	−.01	.14*	.05	1.43	0.50
3. Education^b	−.16	.30**	—	.17	.10	.15	.14	.13	.01	.09	.13	.02	4.75	0.46
4. Organizational tenure	.56**	.20*	−.13	—	.34**	.32**	.36**	.03	.05	.01	.52**	.54**	10.05	6.22
5. Building internal contacts	.03	.17	.24*	−.05	—	.75**	.60**	.18	.14	−.06	.65**	.55**	2.62	0.56
6. Maintaining internal contacts	.28**	.28**	.25*	.06	.30**	—	.69**	.09	.18	.06	.57**	.50**	2.84	0.54
7. Using internal contacts	−.03	.18	.26**	−.07	.20*	.50**	—	.12	.25*	.16	.55**	.62**	2.68	0.66
8. Building external contacts	.23*	.37**	.29**	.22*	.38**	.42**	.46**	—	.65**	.53**	.06	.02	1.88	0.44
9. Maintaining external contacts	.27**	.37**	.25*	.22*	.29**	.47**	.55**	.70**	—	.67**	.04	.14	1.74	0.42
10. Using external contacts	.20*	.26**	.15	.18	.26**	.41**	.52**	.63**	.77**	—	−.09	−.02	1.72	0.45
11. Salary	.38**	.46**	.35**	.30**	.18	.43**	.23*	.48**	.53**	.49**	—	.76**	4.06	2.31
12. Promotions	.44**	.38**	.08	.36**	.29**	.31**	.20*	.50**	.54**	.43**	.61**	—	2.40	1.79
M	44.70	1.62	4.26	14.26	2.36	2.63	2.38	2.15	2.14	2.09	7.12	3.78
SD	11.28	0.49	1.02	10.50	0.50	0.48	0.49	0.66	0.64	0.79	0.74	2.58

Models by Sample	SB-χ²	df	CFI	TLI	RMSEA
German sample
One-factor model	635.08	189	.69	.66	.143
Two-factor model^a (internal, external)	525.82	188	.77	.74	.125
Three-factor model^a (building, maintaining, using)	464.08	186	.81	.79	.113
Five-factor model^b	259.26	164	.92	.90	.078
Six-factor model^a	227.32	174	.97	.96	.049
Chinese sample
One-factor model	767.28	189	.48	.42	.186
Two-factor model^a (internal, external)	432.12	188	.78	.76	.121
Three-factor model^a (building, maintaining, using)	668.03	186	.57	.52	.170
Five-factor model^b	265.40	164	.90	.87	.089
Six-factor model^a	260.58	174	.92	.90	.077

Model	SB-χ²	df	CFI	TLI	RMSEA (90% CI)	ΔSB-χ²	Δdf
Configural model	477.83***	348	.95	.93	.063 [.048, .076]	—	—
Metric model	488.16***	363	.95	.94	.060 [.046, .074]	11.16	15
Full scalar model	616.24***	384	.90	.89	.080 [.068, .092]	132.61***	21
Partial scalar model^a	565.40***	381	.92	.92	.072 [.059, .084]	80.53***	18

Predictor	German Group	Chinese Group
Control variables
Age	.17* (0.08)	.22* (0.10)	.16 (0.11)	.11 (0.11)
Sex	.19* (0.08)	.14 (0.09)	.10 (0.08)	.04 (0.08)
Education	.26** (0.09)	−.05 (0.12)	−.03 (0.06)	−.11 (0.08)
Organizational tenure	.09 (0.08)	.11 (0.11)	.18 (0.11)	.26* (0.11)
Networking behavior
Building internal contacts	−.05 (0.07)	.15* (0.07)	.38*** (0.10)	.25* (0.12)
Maintaining internal contacts	.16 (0.10)	.00 (0.08)	.04 (0.10)	−.08 (0.12)
Using internal contacts	−.15 (0.10)	−.07 (0.10)	.22* (0.11)	.41*** (0.11)
Building external contacts	.04 (0.10)	.14 (0.15)	.02 (0.10)	−.10 (0.11)
Maintaining external contacts	.14 (0.14)	.32* (0.14)	−.06 (0.13)	.12 (0.12)
Using external contacts	.25* (0.12)	.01 (0.11)	−.06 (0.11)	−.08 (0.12)
R ²	.51***	.45***	.58***	56***
N	106	83

Predictor Variable	German Sample	Chinese Sample
Age	.059*	11.63	[0.02, 0.12]	[0.01, 0.13]	.086*	14.95	[0.03, 0.18]	[0.02, 0.20]
Sex	.080*	15.89	[0.02, 0.16]	[0.02, 0.17]	.012	2.10	[0.00, 0.06]	[−0.01, 0.09]
Education	.071*	14.14	[0.01, 0.16]	[0.01, 0.17]	.003	0.47	[0.00, 0.01]	[−0.02, 0.03]
Tenure	.036	7.04	[0.01, 0.09]	[−0.00, 0.11]	.093*	16.24	[0.04, 0.16]	[0.03, 0.17]
Building internal contacts	.010	1.04	[0.00, 0.01]	[−0.02, 0.03]	.172*	29.96	[0.10, 0.26]	[0.10, 0.27]
Maintaining internal contacts	.054*	10.67	[0.01, 0.13]	[0.01, 0.15]	.094*	16.39	[0.05, 0.14]	[0.05, 0.16]
Using internal contacts	.012	2.35	[0.00, 0.02]	[−0.02, 0.03]	.100*	17.37	[0.03, 0.18]	[0.04, 0.19]
Building external contacts	.046*	9.09	[0.02, 0.09]	[0.01, 0.10]	.003	0.54	[0.00, 0.00]	[−0.03, 0.04]
Maintaining external contacts	.067*	13.29	[0.03, 0.12]	[0.02, 0.13]	.003	0.56	[0.00, 0.00]	[−0.03, 0.02]
Using external contacts	.075*	14.86	[0.03, 0.14]	[0.02, 0.15]	.008	1.42	[0.00, 0.03]	[−0.02, 0.06]
R ²	.51	.57

Predictor Variable	German Sample	Chinese Sample
Age	.083*	18.40	[0.03, 0.16]	[0.03, 0.17]	.075*	14.01	[0.03, 0.15]	[0.02, 0.16]
Sex	.049	10.89	[0.01, 0.12]	[0.00, 0.13]	.001	0.28	[0.00, 0.00]	[−0.02, 0.04]
Education	.004	0.92	[0.00, 0.00]	[−0.04, 0.03]	.006	1.11	[0.00, 0.03]	[−0.01, 0.07]
Tenure	.050	11.18	[0.01, 0.14]	[0.00, 0.15]	.119*	22.10	[0.05, 0.21]	[0.05, 0.22]
Building internal contacts	.036	8.04	[0.01, 0.10]	[0.00, 0.12]	.108*	20.15	[0.04, 0.19]	[0.04, 0.20]
Maintaining internal contacts	.018	4.10	[0.01, 0.05]	[−0.00, 0.07]	.065*	12.04	[0.02, 0.11]	[0.02, 0.13]
Using internal contacts	.010	2.25	[0.00, 0.02]	[−0.01, 0.03]	.142*	26.39	[0.06, 0.25]	[0.06, 0.26]
Building external contacts	.065*	14.49	[0.02, 0.16]	[0.02, 0.17]	.005	0.84	[0.00, 0.01]	[−0.02, 0.04]
Maintaining external contacts	.090*	20.02	[0.04, 0.17]	[0.04, 0.17]	.012	2.15	[0.00, 0.04]	[−0.01, 0.06]
Using external contacts	.044*	9.72	[0.02, 0.08]	[0.01, 0.09]	.005	0.93	[0.00, 0.01]	[−0.02, 0.05]
R ²	.45	.54

Footnotes

Authors’ Note

Judith Volmer and Maximilian Orth shared first authorship.

Acknowledgment

We gratefully acknowledge Julia Paul for her crucial help with data collection.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

References

Banville

Desrosiers

Genet-Volet

(2000). Translating questionnaires and inventories using a cross-cultural translation technique. Journal of Teaching in Physical Education, 19, 374–387.

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588–606.

Bozionelos

Wang

(2006). The relationship of mentoring and network resources with career success in the Chinese organizational environment. The International Journal of Human Resource Management, 17, 1531–1546. doi:10.1080/09585190600878345

Bian

Huang

Zhang

(2015). Information and favoritism: The network effect on wage income in China. Social Networks, 40, 129–138. doi:10.1016/j.socnet.2014.09.003

Blickle

Witzki

A. H.

Schneider

P. B.

(2009). Self-initiated mentoring and career success: A predictive field study. Journal of Vocational Behavior, 74, 94–101. doi:10.1016/j.jvb.2008.10.008

Bryant

F. B.

Satorra

(2012). Principles and practice of scaled difference chi-square testing. Structural Equation Modeling, 19, 372–398. doi:10.1080/10705511.2012.687671

Burt

R. S.

(2004). Structural holes and good ideas. American Journal of Sociology, 110, 349–399. doi:10.1086/421787

Byrne

B. M.

(2010). Structural equation modelling with AMOS: Basic concepts, applications, and programming (2nd ed.). New York, NY: Routledge.

Cheung

G. W.

Rensvold

R. B.

(1999). Testing factorial invariance across groups: A reconceptualization and proposed new method. Journal of Management, 25, 1–27. doi:10.1177/014920639902500101

10.

Cheung

Y. H.

Dougherty

T. W.

Herndon

N. C.

(2011). Personality, networking behaviors and network resources of employees in the U.S. and Hong Kong. Paper presented at the Academy of Management Conference, San Antonio, TX.

11.

Coleman

J. S.

(1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95–S120.

12.

Forret

M. L.

Dougherty

T. W.

(2001). Correlates of networking behavior for managerial and professional employees. Group & Organization Management, 26, 283–311. doi:10.1177/1059601101263004

13.

Gelfand

M. J.

Erez

Aycan

(2007). Cross-cultural organizational behavior. Annual Review of Psychology, 58, 479–514. doi:10.1146/annurev.psych.58.110405.085559

14.

Gervorkian

(2011). Networking as a mediator between proactive personality, performance perceptions, and career success (Unpublished Dissertation). Alliant International University, San Diego, CA.

15.

Gibson

Hardy

J. H.

III Buckley

M. R.

(2014). Understanding the role of networking in organizations. The Career Development International, 19, 146–161. doi:10.1108/CDI-09-2013-0111

16.

Gong

Chang

(2008). Institutional antecedents and performance consequences of employment security and career advancement practices: Evidence from the People’s Republic of China. Human Resource Management, 47, 33–48.

17.

Heslin

P. A.

(2005). Conceptualizing and evaluating career success. Journal of Organizational Behavior, 26, 113–136. doi:10.1002/job.270

18.

Hofstede

G. H.

(2003). Culture’s consequences (2nd ed.). Thousand Oaks, CA: Sage.

19.

Johnson

J. W.

(2004). Factors affecting relative weights: The influence of sampling and measurement error. Organizational Research Methods, 7, 283–299. doi:10.1177/1094428104266018

20.

Kim

(2013). Networking enablers, constraints and dynamics: A qualitative analysis. The Career Development International, 18, 120–138. doi:10.1108/CDI-04-2012-0051

21.

Kraimer

M. L.

Seibert

S. E.

Astrove

S. L.

(2015). Consequences of high LMX: Career mobility and success. In Bauer

T. N.

Erdogan

(Eds.), Oxford handbook of leader–member exchange (pp. 241–259). Oxford, England: Oxford University Press.

22.

Kram

K. E.

(1985). Mentoring at work: Developmental relationships in organizational life. Glenview, IL: Scott Foresman.

23.

Lent

R. W.

Brown

S. D.

(2013). Social cognitive model of career self-management: Toward a unifying view of adaptive career behavior across the life span. Journal of Counseling Psychology, 60, 557–568. doi:10.1037/a003344

24.

Leung

Bond

M. H.

(1984). The impact of cultural collectivism on reward allocation. Journal of Personality and Social Psychology, 47, 793–804.

25.

Leung

Iwawaki

(1988). Cultural collectivism and distributive behavior. Journal of Cross-Cultural Psychology, 19, 35–49.

26.

Little

T. D.

Cunningham

W. A.

Shahar

Widaman

K. F.

(2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9(2), 151–173. doi:10.1207/S15328007SEM0902_1

27.

Little

T. D.

Preacher

K. J.

Selig

J. P.

Card

N. A.

(2007). New developments in latent variable panel analyses of longitudinal data. International Journal of Behavioral Development, 31, 357–365. doi:10.1177/0165025407077757

28.

Leung

Koch

P. T.

(2006). Managerial knowledge sharing: The role of individual, interpersonal, and organizational factors. Management and Organization Review, 2, 15–41.

29.

McArdle

Waters

Briscoe

J. P.

Hall

(2007). Employability during unemployment: Adaptability, career identity and human and social capital. Journal of Vocational Behavior, 71, 247–264. doi:10.1016/j.jvb.2007.06.003

30.

McCallum

S. Y.

Forret

M. L.

Wolff

(2014). Internal and external networking behavior: An investigation of relationships with affective, continuance, and normative commitment. The Career Development International, 19, 595–614. doi:10.1108/CDI-08-2013-0101

31.

Milfont

T. L.

Fischer

(2010). Testing measurement invariance across groups: Applications in cross-cultural research. International Journal of Psychological Research, 3, 111–121.

32.

Muthén

L. K.

Muthén

B. O.

(2015). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

33.

T. H.

Eby

L. T.

Sorensen

K. L.

Feldman

D. C.

(2005). Predictors of objective and subjective career success. A meta-analysis. Personnel Psychology, 58, 367–408. doi:10.1111/j.1744-6570.2005.00515.x

34.

T. W. H.

Feldman

D. C.

(2014). A conservation of resources perspective on career hurdles and salary attainment. Journal of Vocational Behavior, 85, 156–168. doi:10.1016/j.jvb.2014.05.008

35.

Porter

C. M.

Woo

S. E.

(2015). Untangling the networking phenomenon: A dynamic psychological perspective on how and why people network. Journal of Management, 41, 1477–1500. doi:10.1177/0149206315582247

36.

Rasdi

R. M.

Garavan

T. N.

Ismail

(2013). Networking behaviours and managers’ career success in the Malaysian public service: The moderating effect of gender. Personnel Review, 42, 684–703. doi:10.1108/PR-07-2010-0117

37.

Satorra

Bentler

P. M.

(2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507–514. doi:10.1007/BF02296192

38.

Shiraev

E. B.

Levy

D. A.

(2013). Cross-cultural psychology: Critical thinking and contemporary applications (5th ed.). Boston, MA: Pearson.

39.

Steinmetz

Schmidt

Tina-Booh

Wieczorek

Schwartz

S. H.

(2009). Testing measurement invariance using multigroup CFA: Differences between educational groups in human values measurement. Quality & Quantity: International Journal of Methodology, 43, 599–616. doi:10.1007/s11135-007-9143-x

40.

Tonidandel

LeBreton

J. M.

(2011). Relative importance analysis: A useful supplement to regression analysis. Journal of Business and Psychology, 26, 1–9. doi:10.1007/s10869-010-9204-3

41.

Tonidandel

LeBreton

J. M.

(2015). RWA Web: A free, comprehensive, web-based, and user-friendly tool for relative weight analyses. Journal of Business and Psychology, 30, 207–216. doi:10.1007/s10869-014-9351-z

42.

Tonidandel

LeBreton

J. M.

Johnson

J. W.

(2009). Determining the statistical significance of relative weights. Psychological Methods, 14, 387–399. doi:10.1037/a0017735

43.

H. S.

Forret

M. L.

Sullivan

S. E.

(2006). Careers in a non-Western context: An exploratory empirical investigation of factors related to the career success of Chinese managers. The Career Development International, 11, 580–593. doi:10.1108/13620430610713454

44.

Savickas

M. L.

Porfeli

E. J.

(2012). Career Adapt-Abilities Scale: Construction, reliability, and measurement equivalence across 13 countries. Journal of Vocational Behavior, 80, 661–673. doi:10.1016/j.jvb.2012.01.011

45.

Smith

P. B.

Peterson

M. F.

Schwartz

S. H.

(2002). Cultural values, sources of guidance, and their relevance to managerial behavior. Journal of Cross-Cultural Psychology, 33, 188–208.

46.

Volmer

Wolff

H.-G.

(2016). The short-term consequences of networking: A daily diary study. Manuscript submitted for publication.

47.

Wok

Hashim

(2014). Communication networks, organizational contacts and communication power in grooming professionals for career success. Malaysian Journal of Communication, 30, 219–242.

48.

Wolff

H.-G.

Kim

(2012). The relationship between networking behaviors and the Big Five personality dimensions. The Career Development International, 17, 43–66. doi:10.1108/13620431211201328

49.

Wolff

H.-G.

Moser

(2006). Entwicklung und Validierung einer Networkingskala [Development and validation of a networking scale]. Diagnostica, 52, 161–180. doi:10.1026/0012-1924.52.4.161

50.

Wolff

H.-G.

Moser

(2009). Effects of networking on career success: A longitudinal study. Journal of Applied Psychology, 94, 196–206. doi:10.1037/a0013350

51.

Wolff

H.-G.

Moser

(2010). Do specific types of networking predict specific mobility outcomes? A two-year prospective study. Journal of Vocational Behavior, 77, 238–245. doi:10.1016/j.jvb.2010.03.001

52.

Wolff

H.-G.

Moser

Grau

(2008). Networking: Theoretical foundations and construct validity. In Deller

(Ed.), Readings in applied organizational behavior from the Lüneburg symposium—Personality at work (pp. 101–118). Mering, Germany: Rainer Hampp.

53.

Wolff

H.-G.

Schneider-Rahm

C. I.

Forret

M. L.

(2011). Adaptation of a German multidimensional networking scale into English. European Journal of Psychological Assessment, 27, 244–250. doi:10.1027/1015-5759/a000070

Predictor	German Group		Chinese Group
Predictor	Salary	Promotions	Salary	Promotions
Control variables
Age	.17* (0.08)	.22* (0.10)	.16 (0.11)	.11 (0.11)
Sex	.19* (0.08)	.14 (0.09)	.10 (0.08)	.04 (0.08)
Education	.26** (0.09)	−.05 (0.12)	−.03 (0.06)	−.11 (0.08)
Organizational tenure	.09 (0.08)	.11 (0.11)	.18 (0.11)	.26* (0.11)
Networking behavior
Building internal contacts	−.05 (0.07)	.15* (0.07)	.38*** (0.10)	.25* (0.12)
Maintaining internal contacts	.16 (0.10)	.00 (0.08)	.04 (0.10)	−.08 (0.12)
Using internal contacts	−.15 (0.10)	−.07 (0.10)	.22* (0.11)	.41*** (0.11)
Building external contacts	.04 (0.10)	.14 (0.15)	.02 (0.10)	−.10 (0.11)
Maintaining external contacts	.14 (0.14)	.32* (0.14)	−.06 (0.13)	.12 (0.12)
Using external contacts	.25* (0.12)	.01 (0.11)	−.06 (0.11)	−.08 (0.12)
R ²	.51***	.45***	.58***	56***
N	106		83

Predictor Variable	German Sample				Chinese Sample
	Relative Weight		Regular CI	Statistical Significance CI^a	Relative Weight		Regular CI	Statistical Significance CI^a
	Raw	Rescaled (%)	Regular CI	Statistical Significance CI^a	Raw	Rescaled (%)	Regular CI	Statistical Significance CI^a
Age	.059*	11.63	[0.02, 0.12]	[0.01, 0.13]	.086*	14.95	[0.03, 0.18]	[0.02, 0.20]
Sex	.080*	15.89	[0.02, 0.16]	[0.02, 0.17]	.012	2.10	[0.00, 0.06]	[−0.01, 0.09]
Education	.071*	14.14	[0.01, 0.16]	[0.01, 0.17]	.003	0.47	[0.00, 0.01]	[−0.02, 0.03]
Tenure	.036	7.04	[0.01, 0.09]	[−0.00, 0.11]	.093*	16.24	[0.04, 0.16]	[0.03, 0.17]
Building internal contacts	.010	1.04	[0.00, 0.01]	[−0.02, 0.03]	.172*	29.96	[0.10, 0.26]	[0.10, 0.27]
Maintaining internal contacts	.054*	10.67	[0.01, 0.13]	[0.01, 0.15]	.094*	16.39	[0.05, 0.14]	[0.05, 0.16]
Using internal contacts	.012	2.35	[0.00, 0.02]	[−0.02, 0.03]	.100*	17.37	[0.03, 0.18]	[0.04, 0.19]
Building external contacts	.046*	9.09	[0.02, 0.09]	[0.01, 0.10]	.003	0.54	[0.00, 0.00]	[−0.03, 0.04]
Maintaining external contacts	.067*	13.29	[0.03, 0.12]	[0.02, 0.13]	.003	0.56	[0.00, 0.00]	[−0.03, 0.02]
Using external contacts	.075*	14.86	[0.03, 0.14]	[0.02, 0.15]	.008	1.42	[0.00, 0.03]	[−0.02, 0.06]
R ²	.51				.57