Socioemotional wealth (SEW) across borders: Integrating national context into SEW research

Socioemotional wealth (SEW) across borders: Integrating national context into SEW research Valeriano Sanchez-Famoso a,*,1 , Cristina Cruz b , Mohamed Mazen Batterjee b , Jorge-Humberto Mejia-Morelos c,2 , Luis Cisneros d,3 , Nhu Tuyen LE e,4 a University of the Basque Country UPV/EHU, Spain b IE Business School –IE University, IE Tower, Paseo de la Castellana, 259, Madrid 28046, Spain c Entrepreneurship &Innovation Department, Observatory of Entrepreneurship La Sph` ere, HEC Montreal 3000, Chem. de la Cˆ ote-Sainte-Catherine, Montr´ eal, QC H3T 2A7, Canada d Entrepreneurship &Innovation Department, Observatory of Entrepreneurship La Sph` ere, Business Families Center, HEC Montreal 3000, Chem. de la Cˆ ote-SainteCatherine, Montr´ eal, QC H3T 2A7, Canada e Grenoble Ecole de Management, 12 Rue Pierre Semard, Grenoble 38000, France ARTICLE INFO Keywords: SEW FIBER scale Context research Measurement invariance Cross-country comparison ABSTRACT This study addresses the challenges associated with integrating the national context into socioemotional wealth (SEW) research and highlights the consequences of overlooking contextual variations. We emphasize two critical issues: inadequate testing of SEW assumptions and threats to the construct validity of SEW measurement. We recommend that cultural and institutional aspects of the national context should be incorporated to understand how family owners prioritize SEW dimensions, and how their willingness trades off current SEW wealth for prospective financial gains. We also conduct an exploratory study measuring the FIBER scale in Canada, Mexico, Saudi Arabia, Spain, and Vietnam. We survey 1464 family owners to enhance SEW construct validity by probing the cross-country measurement invariance of the FIBER scale. Furthermore, we conduct comparative research to investigate how cultural and institutional aspects shape the FIBER dimensions across national contexts. 1. Introduction The socioemotional wealth (SEW) framework originated from the seminal work of G´ omez-Mejía et al. (2007) is a pivotal advancement in family business research (Brigham &Payne, 2019, p. 326). It is widely used across various domains within family firm contexts, including firm strategic decisions (Dehlen, 2013; Feldman et al., 2014; Kotlar et al., 2018; Laffranchini et al., 2022), corporate governance (Bammens et al., 2011; Di Vito &Trottier, 2022; Labaki &D’Allura, 2021), social responsibility (Combs et al., 2023; Cruz et al., 2014; Dayan et al., 2019), and entrepreneurial outcomes (Block et al., 2013; Chrisman &Patel, 2012; Rodrigues et al., 2022). At its core, the SEW framework posits that family owners’decisions are motivated by their commitment to safeguarding the family SEW—the non-financial aspects of the firm—- which fulfills their affective needs (G´ omez-Mejía et al., 2007). Despite the extensive literature on SEW, a significant gap persists due to the lack of contextualization. Although studies often apply the SEW logic, they borrow arguments from others and ignore the specificities of each context (Krueger et al., 2021). This oversight is troubling, given that contextualized SEW research suggests that family owners’preferences are deeply influenced by their institutional and cultural surroundings (Berrone et al., 2012; Gomez-Mejia et al., 2020). The challenge of conducting “context-free”SEW research becomes apparent when considering the validation of SEW as a construct, which remains uncertain (Brigham &Payne, 2019). The widely used FIBER scale has significant variation across countries (Gerken et al., 2022; Hauck et al., * Corresponding author. E-mail addresses: [email protected] (V. Sanchez-Famoso), [email protected] (C. Cruz), [email protected] (M.M. Batterjee), [email protected] (J.-H. Mejia-Morelos), [email protected] (L. Cisneros), [email protected] (N. Tuyen LE). 1 ORCID: 0000–0003-3977–3337 2 ORCID: 0000–0001-7324–2367 3 ORCID: 0000–0001-5983–8973 4 ORCID: 0000–0003-3057–2563 Contents lists available at ScienceDirect Journal of Family Business Strategy journal homepage: www.elsevier.com/locate/jfbs https://doi.org/10.1016/j.jfbs.2024.100647 Journal of Family Business Strategy 16 (2025) 100647 Available online 30 December 2024 1877-8585/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC license ( http://creativecommons.org/licenses/bync/4.0/ ). 2016), highlighting the need for a deeper understanding of the reasons and implications of these differences. Unfortunately, there is no comprehensive explanation of how context matters for the conceptualization of SEW (G´ omez-Mejía &Herrero, 2022). This study addresses the challenges of integrating national context into SEW research, building on recent findings emphasizing the contextual specificities of family owners in different regions (Basco et al., 2019; Gomez-Mejia et al., 2020; Krueger et al., 2021). We highlight that failing to consider these contextual variations raises two critical issues: inadequate testing of the relevant assumptions of the SEW approach and threats to overall construct validity regarding the SEW measurement. Thus, we propose that both cultural and institutional aspects of the national context should be incorporated when conducting SEW research. These factors partly determine how family owners prioritize SEW dimensions and their willingness to trade off current SEW for future financial gains. Additionally, we emphasize the importance of enhancing construct validity in SEW research by testing the cross-country measurement of SEW measurement instruments, this is to say testing whether these instruments assess the SEW concept in a consistent manner across countries (Cheung &Rensvold, 2002; Harkness et al., 2003). If the SEW instrument shows cross country measurement invariance, we can confidently attribute the discrepancies identified in SEW across countries arise from varying cultural and institutional conditions, and not to limitations of the scale itself, thus enabling meaningful cross-country comparisons. To advance in this direction, we employ a comparing approach to SEW research and develop an exploratory study that measures SEW in five countries: Canada, Mexico, Saudi Arabia, Spain, and Vietnam. Surveying 1464 family owners across these countries, we initially investigate the cross-country measurement invariance of the FIBER scale. This scale is the most widely cited tool for SEW measurement (Swab et al., 2020). Our results confirm the cross country measurement invariance of the SEW scale. This means that, while there may be variations in the importance attributed to each FIBER dimension across national contexts, family owners’responses to the FIBER scale can be meaningfully compared across countries. Based on our findings, we conduct comparative SEW research to reveal the nuances of SEW dimensions across countries. This exploratory comparison reveals significant differences among the five countries regarding how family owners prioritize various SEW dimensions. We theorize that cultural and institutional aspects of the national context may help explain these variations. Overall, our results address recent calls for a more context-sensitive approach in family business research in general (Amato et al., 2022; Basco et al., 2019; Krueger et al., 2021) and SEW research in particular (Gomez-Mejia et al., 2024; G´ omez-Mejía &Herrero, 2022). Despite advances in the research on family businesses, the importance of context in shaping SEW antecedents and outcomes remains largely underexplored (Cruz et al., 2023; G´ omez-Mejía &Herrero, 2022). Moreover, by highlighting the nuances related to the specificities of family firms in different countries, we contribute to a better interpretation of context as a source of family firm heterogeneity (Basco et al., 2020) 2. Theoretical background Substantial academic evidence demonstrate that the conduct of firms is contingent upon their national context. Boundaries between countries represent a unique set of cultural norms that drive shared beliefs, values, customs, traditions, and behaviors (Sundin &Horowitz, 2002). Moreover, companies within a particular country are embedded in an institutional environment related to the formal and informal rules, laws, and regulatory systems that shapes internal and external corporate governance mechanisms to foster economic exchanges (Lien et al., 2016; Peng &Jiang, 2010). Cultural expectations and institutional frameworks vary widely across nations and have profound implications for the interrelationship among work, family, and business (Kossek &Ollier-Malaterre, 2013). Family firms and their owners are also prone to these influences. Family owners across various countries exhibit significant disparities in their goals and behavior (Gupta et al., 2010; Howorth et al., 2010). For instance, Howorth et al. (2010) revealed different tendencies among family firm owners in European countries, such as Spain, Greece, and Italy, in contrast to those in the United States. European family owners are often reluctant to part with their businesses, viewing them as integral extensions of their families. In contrast, family owners in the United States are more inclined to consider selling their businesses, provided the price is right. This cross-country variation in generational intentions among family owners is further corroborated by Corbetta and Montemerlo (1999), who highlighted a notable contrast between American and Italian family owners. Similarly, Sharma and Rao (2000) compared successor’s attributes in Indian and Canadian family firms, and found that, whereas it was extremely important for Indian family owners that the successor should be from their bloodline, this was not as important for Canadian owners. Recent years have seen growing recognition that the underlying logic of SEW reasoning may not operate in a standardized manner across national contexts (Cruz et al., 2023; Gomez-Mejia et al., 2024). The SEW approach, rooted in prior family-firm and behavioral studies, assumes that family firms are commonly motivated by and committed to preserving their SEW (Berrone et al., 2012), defined as the “non-economic utilities of family owners”(G´ omez-Mejía et al., 2007). It refers to family owners’intentional pursuit of noneconomic objectives, such as control, transgenerational succession, social capital, emotional connection to the firm, and reputation (Berrone et al., 2012). Hence, family owners represent the focal decision-making group regarding SEW (Swab et al., 2020). The operating strategies of families may differ across national boundaries based on cultural norms and institutional differences (Wright et al., 2014). Further, building on the behavioral agency model (Wiseman &Gomez-Mejia, 1998), the SEW approach argues that family firms base their decision on the framing of the problem and reference point when making strategic decisions (Kotlar et al., 2013). The reference point would be strongly shaped by situational aspects that occur at the family and/or business level (Basco, 2018; G´ omez-Mejía &Herrero, 2022). Understanding these SEW differences across national contexts where the family firm operates could substantially broaden our perception of the heterogeneity among family firms and generalizability of SEW research findings. Specifically, conducting “national context-free”SEW research—research that does not consider the specificities of the cultural and institutional setting where the family firms operate across national boundaries—poses two critical issues: inadequate testing of the relevant assumptions of the SEW approach and its influence on firm outcomes, and threats to overall construct validity regarding SEW measurement. 2.1. National context-free SEW research and inadequate theory testing The predominant depiction of SEW as an explanatory variable tends to treat it as a latent concept rather than operationalizing it directly (Gomez-Mejia et al., 2011). SEW is often equated with distal proxies like family ownership or management (e.g., Chirico et al., 2020;Cruz et al., 2014;Gomez-Mejia et al., 2018). This conceptualization portrays SEW as an asset as well as a liability for family firms, impacting strategic decision-making in both positive and negative ways (Swab et al., 2020). However, there is no consensus regarding the relationship between SEW and firm outcomes, particularly its impact on firm performance (G´ omez-Mejía &Herrero, 2022). Some suggest a positive correlation, hinting that SEW fosters commitment and longer investment horizons (Cruz et al., 2010; Zellweger et al., 2012). Others propose a potential “dark side,”positing that family firms might prioritize SEW over financial returns (Kellermanns et al., 2012). Cross-study comparisons are hindered by the diversity of contexts, thus affecting the establishment of prevalent positions. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 2 To address this inconclusiveness, Davila et al. (2023) conducted a meta-analysis and revealed a positive correlation between SEW and family firm performance. They highlighted two crucial nuances in the relationship: the non-uniform influence of SEW dimensions on performance and the negative relationship between SEW and specific strategic choices impacting firm performance. However, including culturally distant countries in the metanalyses raises concerns regarding the influence of the national culture in explaining the SEW-performance relationship. Examining these nuances through a national contextual lens could be significant in SEW research. Cultural settings may shape how family members prioritize SEW dimensions, influencing the relationship between SEW dimensions and performance (Gomez-Mejia et al., 2024; Swab et al., 2020). Davila et al. (2023) found that while most SEW dimensions correlate positively with firm performance, dynastic succession is negatively correlated. G´ omez-Mejía and Herrero (2022) found that in highly collectivistic societies, where family ties are paramount, family owners prioritized the renewal of family ties through succession (the “R”dimension of the FIBER scale). Consequently, including the national culture may explain SEW-performance variations across countries: highly collectivistic societies may prioritize dynastic succession, thus potentially impacting performance negatively. Examining the negative relationship between SEW and specific strategic choices impacting firm performance through a national contextual lens can also bring interesting findings to light. According to the SEW logic, family owners often do not pursue diverse strategic options to safeguard family control (Chrisman &Patel, 2012). This is particularly true for countries with high power distance, where cultural norms promote autocratic leadership and emphasize family authority, reinforcing family control over the business (Hofstede, 1983). Consequently, in such societies, the positive relationship between SEW and performance may weaken compared to those with lower scores in this cultural dimension. This may stem from an intensified focus on the family control dimension of SEW, potentially hindering family firms from engaging in value-enhancing strategic activities. Recent SEW studies suggest that family owners’behavior is influenced by the institutional framework within a specific country (Berrone et al., 2022). Gomez-Mejia et al. (2024) suggested that the volatile, uncertain, complex, and ambiguous (VUCA) nature of the Latin-American and Caribbean context (LAC) are important institutional factors that influence SEW in the region. Their review of the five FIBER dimensions considering the contextual peculiarities of the region concludes that because of cultural aspects, family businesses in this region are “SEW intensive.”This means that they give high priority to all SEW dimensions. However, the institutional environment shapes how family owners prioritize each dimension. For instance, while the concept of an extended family in LAC emphasizes the importance of the “I”(Family Identification) dimension of SEW, a VUCA environment moderates this relationship negatively, due to associated risks of kidnappings and extortions because of maintaining a strong public image in these countries. Further, a recent study conducted by Pinelli et al. (2023) examined SEW and family firm acquisitions and delved into the implications of considering how the institutional context affects the SEW approach. They suggest that SEW concerns make family firms more likely than nonfamily firms to undertake related acquisitions when operating in uncertain environments. This is done to avoid losses to the family’s current SEW. Nevertheless, family owners are more likely to undertake unrelated acquisitions when the target firm operates in a more developed institutional context. Taking a closer look at this from a contextual lens, we realize that institutional environments with robust frameworks supporting economic transactions, such as enforcement contracts, protection of property rights, and stable national governance, facilitate unrelated acquisitions. This is because these elements enhance predictability regarding prospective financial gains from such acquisitions, making them more appealing to family firms. To summarize, this discussion underscores the significant influence of cultural and institutional norms on SEW. Neglecting them may lead to inadequate testing of SEW assumptions. Therefore, incorporating cultural and institutional contexts in theoretical arguments is crucial for comprehensive SEW research. 2.2. National context-free SEW research and lack of construct validity In addition to the challenges related to inadequate theory testing, the issue of context-free SEW also raises concerns about construct validity. There is a lack of clarity regarding the validity of the SEW construct and its measurement (Brigham &Payne, 2019; Chua et al., 2015; Newbert & Craig, 2017). This criticism is evident in studies that test the SEW construct using the FIBER scale across culturally diverse countries such as Spain (G´ omez-Mejía &Herrero, 2022), Bangladesh (Razzak &Jassem, 2019), Finland (Filser et al., 2018), Mexico (Angulo et al., 2016), and the United Arab Emirates (Dayan et al., 2019). These studies yielded inconclusive results regarding the dimensions constituting the SEW construct. For example, while Razzak and Jassem (2019) concluded that all 27 items of the FIBER model load onto the scale, Dayan et al. (2019) suggested that the “F”dimension may not be significant in explaining SEW for United Arab Emirates families. To overcome these limitations, recent studies have attempted to validate the psychometric properties of the original FIBER scale using samples from different countries. The first attempt, Hauck et al. (2016), was built on 216 questionnaires—112 from German and 104 from Austrian firms. The authors acknowledge that the sample is entirely based on the German context and calls for future research to “revalidate the scale with a sample composed of more heterogeneous firms”(p. 143). Gerken et al. (2022) responded to this call by conducting a replication study using five samples from five different studies that used questionnaires in which the FIBER scale was included (Heider et al., 2021; Sch¨ afer, 2016; Schneider, 2018; Weimann, 2020). The total sample included 1206 responses from family owners in Europe, Asia, and the US. The authors validated the scale, but they did not acknowledge the influence of cultural context on SEW measurement. Last, G´ omez-Mejía and Herrero (2022) tested the FIBER scale in Spain and recognized the lack of contextualization, emphasizing that “the extent to which the overarching culture makes a difference in the content structure of SEW and its effects still remains an open question”(p. 8). Together, these studies highlight the significance of contextual variations among countries in shaping the family SEW. However, they also suggest that discrepancies among studies may have emerged from inconsistencies in the scale itself, including methodological disparities or divergent understandings of SEW across countries. If this is the case, the universality of SEW to capture family owners affective endowment would be questionable and this would invalidate conducting SEW cross country research. Hence, having a cross-cultural valid and reliable SEW instrument is an essential to contextualize SEW research. 2.3. Advancing SEW research: Context by comparing five national contexts The preceding discussion concludes that unconsidered contextual variation in SEW studies may lead to issues, including overall threats to validity and insufficient theory testing. To advance further, we incorporate the national context into SEW research, drawing inspiration from the recommendations of Bamberger (2008). This approach, recently applied to family firm studies by Amato et al. (2022),Gomez-Mejia et al. (2020), and Krueger et al. (2021), proposes a progressive introduction of (national) context research in family business studies in different stages. First, more SEW studies are conducted in diverse national settings (context by sampling). Next, more comparative SEW studies are developed across countries (context by comparing). Last, context is introduced into theoretical arguments of the SEW approach, adopting a context by theorizing research strategy. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 3 In existing SEW research that considers contextual factors, a prevalent approach involves incorporating context through a context by sampling strategy. This spectrum ranges from studies treating context as tangential (e.g., Berrone et al., 2010) to those highlighting its critical role in SEW antecedents and outcomes (e.g., Calabr` o et al., 2018). While context by sampling contributes to consistency in SEW research, it does not facilitate a nuanced understanding of the causal or moderating relationship between national context and SEW. Hence, we embrace the next step proposed by Bamberger (2008) and embark in “context by comparing”SEW research. We undertake a comparative study of SEW in five countries representing distinct national contexts. By gaining insight into the challenges associated with measuring SEW across national contexts and by shedding light on the heterogeneity of SEW across nations, we aim to enhance our comprehension of how various SEW dimensions are altered, modified, or constrained by national culture. Conducting context by comparing SEW research is a major challenge for SEW researchers, making it imperative to examine if available scales to measure SEW are valid across nations, as emphasized by crosscountry research (Lacko et al., 2022). Consequently, we initially focus on assessing whether the FIBER scale can be universally applied across countries, gauging its reliability in capturing SEW within the five-country context. The preferred method of conducting the statistical and psychometric assessment of the cross-cultural equivalence of an instrument is through testing its “measurement invariance”(Cheung & Rensvold, 2002; Harkness et al., 2003). Building on these established methodologies, we delve into testing the cross-country measurement invariance of the FIBER scale. Once we confirm the measurement invariance of the SEW scale, we focus on creating a comparative understanding of SEW across nations. We contend that developing this comparative knowledge is a crucial prerequisite before adopting a context theorizing SEW approach. 2.4. Testing the cross-country measurement invariance of the FIBER scale 2.4.1. Sample selection We chose five countries with distinct national contexts and where family firms have an important economic impact: Canada, Mexico, Saudi Arabia, Spain, and Vietnam. In Canada, family firms generate 49 percent of the private sector real gross domestic product (GDP) and account for 47 percent of private sector employment (Forbes &Basset, 2019). In Mexico, more than 70 percent of the businesses have a family structure and employ approximately 54 million people (Grant-Thornton, 2011). Family business in Saudi Arabia constitute up to 90 percent of all companies, employing 80 percent of the workforce and contributing to 60 percent of the region’s GDP (PwC, 2016). In Spain, family firms generate 70 percent of the total GDP, representing around 75 percent of total private employment (Spanish Family Enterprise Institute, 2021). Last, while official statistics are lacking, 95 percent of Vietnamese enterprises are family businesses (Nguyen Dang Tuan et al., 2019). Moreover, according to the PwC family Business Survey (2021), in Vietnam, the top 100 family businesses account for 25 percent of the country’s GDP. Furthermore, these five countries show important cultural differences, and thus, this is a stringent test to ensure the generalizability of our results. In Table 1a, we compare the cultural dimensions among countries using Hofstede’s dimensions and classify the countries’sample based on Hofstede scale scores to ensure that we have selected culturally distant countries (Debicki et al., 2016; Hofstede, 2001). As Table 1a shows, Canada is the most individualistic society, whereas Vietnam, Saudi Arabia, and Mexico are collectivist societies. Nevertheless, Canada has the lowest power-distance score in our sample, while Mexico and Saudi Arabia rank high in this dimension. Further, although there are two “Latin societies,”Mexico scores 97 in the “Indulgence”dimension, which is in sharp contrast with Spain’s score of 35. The selected countries also differ in their institutional context. As Table 1b shows, they rank very differently regarding the World Bank Governance Indicators (Kaufmann et al., 2010), which provide information on various dimensions of a country’s institutional context (i.e., voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption). For all six indicators, Canada’s ratings compare favorably with the rest of the countries, showing its strong institutional environment. On the contrary, Vietnam, Saudi Arabia, and Mexico rate below at least 60 percent of countries on most governance indicators, suggesting weak institutional frameworks. We followed the suggestions of recent cross-country studies dealing with scale validation (Kankaraˇ s&Moors, 2014; Lacko et al., 2022) and ensured the cross-cultural comparability of our sample. We used the same definition of family firms across the five countries. Following previous studies (Voordeckers et al., 2007; Westhead &Howorth, 2006), the selected firms met two conditions: a) at least 51 % of the ownership belongs to the family and b) at least one family member occupies a governance and/or management position. Nevertheless, as access to family firms’data varies a lot across countries, we built the database of family firms convergent with cross-cultural comparability (Lacko et al., 2022). In Canada, we compiled data provided by Business Families - HEC Montr´ eal and the National Bank Institute of Entrepreneurship - HEC Montr´ eal. Next, we complemented this data with those provided by the Business Families Foundation, which generously provided us access to its data on Qu´ ebec family businesses. For robustness check, we checked the Enterprise Register of Qu´ ebec to verify that the selected firms classify as family firms according to our two conditions. Thus, 1938 companies from all regions of the province of Qu´ ebec were contacted. Similarly, Saudi Arabian family firms were identified through databases provided by associations such as the Jeddah Chamber of Commerce, the Family Business Council in the Council of Saudi Chambers, the Family Business Forum, and the National Center for Family Business. Through these databases, we identified 1189 firms that were classified as family firms based on our definition. The selection of family firms in Spain was done using the SABI 1 database (Casillas et al 2024). To obtain a sample, based on our Table 1a Hofstede’s cultural dimensions scores in the five sampled countries. Hofstede Cultural Dimensions Canada Mexico Spain Vietnam Saudi Arabia Power Distance 39 81 57 70 95 Individualism 80 30 51 20 25 Masculinity 52 69 42 40 60 Uncertainty Avoidance 48 82 86 30 80 Long-Term Orientation 36 24 48 57 36 Indulgence 68 97 44 35 52 Table 1b Institutional differences among the five sampled countries. Canada Mexico Saudi Arabia Spain Vietnam Political Stability and Absence of Violence/ Terrorism 81.6 17.9 23.5 58 49 Governance Effectiveness 95.7 39 63.8 79.5 53.33 Regulatory Quality 96 55.7 49 52 39 Control of corruption 92.8 18 62.3 73 32.86 1 The SABI database contains information on 2600,000 companies from Spain and 800,000 companies form Portugal, and includes public and private Spanish and Portuguese companies. SABI is the Spanish/Portuguese (Iberic) version of the Orbis database, from Bureau Van Dick, which has worldwide coverage. Spain is one of the countries in which the coverage of this database is greater (Bajgar et al., 2020). V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 4 definition of family firms, we conducted an exhaustive review of the shareholding structures (percentage of common stock) and composition (name and surnames of shareholders) 2 of these firms. Thus, we selected companies from SABI that met the following criteria: i) non-listed Spanish companies with more than nine employees; ii) companies with financial information for (at least) the last five years; iii) companies not affected by special situations such as insolvency, wind-up, liquidation or zero activity; and iv) companies in which at least 51 percent of the firm owners were members of the family. From the original dataset of 27,355 Spanish enterprises that fell within the set of parameters, 1312 companies met the specified family criteria. In Mexico, there is not a single professional association with public databases for family business. Thus, we built our sample in three steps. First, we identified family firms listed in the National Institute of Statistics and Geography (INEGI, 2015). Second, we verified these data with scholars (experts) participating in research on family businesses in four universities in Mexico, as well as with consultants collaborating with family businesses. Then, we targeted family firms in the states of Mexico having the highest contribution to PIB: Campeche, Ciudad de Mexico, Estado de M´ exico, Jalisco, Nuevo L´ eon, and Veracruz (National Statistical Directory of Economic Units of INEGI, 2015). Last, we contacted 650 family firms that met our criteria to invite them to participate in the study. In Vietnam too, there is no professional association of family firms. Moreover, family owners in Vietnam receive high social consideration, so it is not easy to reach out to them. Therefore, we relied on our informal networks to set up contacts with directors and managers of family firms. The research team contacted each of these managers to explain to them the goal of the project as well as the content of the questionnaire. We contacted 295 firms. We followed a rigorous process to ensure the validity of the questionnaire across countries (Tan et al., 2020). We carefully translated the English version of the FIBER questionnaire into French, Arabic, Spanish, and Vietnamese. Then, all the translations were modified based on a review by three native social scientists with a doctoral degree. We carefully minimized the problems inherent in simple translation, such as linguistic or psychometric nonequivalence between the different language versions (Hulin &Mayer, 1986). Back translation (Brislin & Olmstead, 1973) was performed on the final versions by bilingual individuals. A comparison of the original and back-translated items indicated substantial similarity between the original and translated versions. We ran a pilot test to make sure that the meanings of the questions were adequate for each cultural context. For example, in Vietnam, the notion of kinship translated into Vietnamese may generate confusion, as the concept of kinship in Western cultures means a blood relationship. However, in Vietnam, it means the relationship between family members (in the sense of community). Thus, we took care of these cultural differences and adapted our final scale accordingly. The data were collected between July 2018 and June 2019. A company that specializes in market research administered the survey and collected data in all countries except Vietnam. In Vietnam, we selected a group of 20 students enrolled in Management Science courses in Vietnamese universities and trained them in data collection. With hard copies of the questionnaire, the students took on the role of interviewers and conducted face-to-face interviews with each respondent. The total final sample across the five countries comprised 1464 respondents: we received 495 completed questionnaires (25.5 percent response rate) from Canada, 114 (17.5 percent) from Mexico, 400 (33.6 percent) from Saudi Arabia, 160 (17.7 percent) from Spain, and in the case of Vietnam, we sent 500 questionnaires, out of which 295 interviews were completed successfully (59 percent). To test for non-response bias, we performed a Kolmogorov-Smirnov test (Siegel &Castellan, 1988) between responding and non-responding firms in Canada, Mexico, Saudi Arabia, and Spain. The analysis revealed no significant differences in the main demographic features between the two groups (i.e., number of employees, annual sales, firm size, and age). Table 2 shows some descriptive statistics. It reveals interesting differences in the family influence across the sample countries. For instance, family ownership is higher than 90 percent in all countries, except in Canada where the family ownership mean is 77 percent. Spain shows a higher percentage of family members involved in the TMT, while Saudi Arabia has the lowest family involvement in management responsibilities. Last, while most family firms in Canada and Mexico are in the first generation, in the rest of the countries, the majority of firms are controlled by the second generation. 3. Results Cross-country management studies pose a crucial question: can instruments validated in one country be readily applied in different contexts without assessing their cross-cultural applicability (Durvasula et al., 1993; Mavondo et al., 2003)? Responding research has predominantly aimed at mitigating the methodological and statistical challenges associated with the issue of measurement non-invariance. The most widely adopted approach to scrutinize the cross-cultural equivalence of an instrument is the examination of “measurement invariance” (Cheung &Rensvold, 2002; Harkness et al., 2003). In essence, this concept seeks to determine whether a measurement instrument assesses the same underlying concept consistently across various subgroups of respondents (Chen et al., 2008; Horn &Mcardle, 1992, p. 117). Before embarking on a comparative analysis of SEW across different countries, it is imperative to subject the FIBER scale to rigorous cross-country measurement invariance testing. To do so, we draw upon MultiGroup Confirmatory Factor Analysis (MGCFA) (J¨ oreskog, 1971) as it “represents the most powerful and versatile approach to testing for cross-national measurement invariance”(Steenkamp &Baumgartner, 1998, p. 78). It involves setting cross-group constraints and comparing more restricted models with less restricted ones (Byrne et al., 1989; Steenkamp &Baumgartner, 1998). Additionally, we follow the stepwise procedure proposed by these authors, testing from the least to the most demanding form of invariance. We started by testing whether the FIBER factor structure is adequate in each country by estimating a confirmatory factor analysis (CFA) model for each of the five data sets (countries). Table 3 shows the CFAs for all the datasets. All Cronbach’s alphas (CA) are above the recommended value of 0.70, and the composite reliability (CR) indices are above 0.70 (Fornell &Larcker, 1981). Moreover, the average variance extracted (AVE) of the constructs exceed the recommended threshold value of 0.50 (Fornell &Larcker, 1981). After testing that the scale is acceptable for each country, we tested whether it shows configural equivalence, that is, whether it exhibits the Table 2 Descriptive statistics of the sample. Canada Mexico Saudi Arabia Spain Vietnam Total Number of Respondents 495 114 400 160 295 1464 Male 395 68 338 80 142 1023 Female 100 46 62 80 153 441 Company Size Mean (No. of employees) 55 200 39 42 119 95 Ownership Mean (%) 77 94 88 92 89 88 Family in TMT (%) 54 58 48 66 54 56 Generation in Charge 1 1 2 2 2 2 2 In Spain, people have two surnames. The first is the first surname of the father, and the second is the first surname of the mother. Therefore, family relationships among shareholders are more evident than in other countries. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 5 Table 3 Validation of the measurement model –reliability and convergent validity. Canada Mexico Saudi Arab Spain Vietnam All Together Dimension Indicator Standardised Loading Standardised Loading Standardised Loading Standardised Loading Standardised Loading Standardised Loading FItem 1 0.707 CA =0.89 CR =0.89 AVE =0.59 0.610 CA =0.86 CR =0.86 AVE =0.51 0.681 CA =0.86 CR =0.86 AVE =0.51 0.806 CA =0.91 CR =0.91 AVE =0.62 0.710 CA =0.86 CR =0.87 AVE =0.54 0.721 CA =0.88 CR =0.88 AVE =0.55 Item 2 0.703 0.748 0.795 0.814 0.838 0.785 Item 3 0.790 0.797 0.683 0.765 0.739 0.737 Item 4 0.812 0.783 0.673 0.782 0.410 0.635 Item 5 0.799 0.777 0.705 0.742 0.820 0.780 Item 6 0.771 0.546 0.743 0.819 0.797 0.780 IItem 7 0.838 CA =0.85 CR =0.86 AVE =0.51 0.768 CA =0.87 CR =0.88 AVE =0.54 0.768 CA =0.88 CR =0.88 AVE =0.56 0.794 CA =0.88 CR =0.88 AVE =0.56 0.531 CA =0.86 CR =0.88 AVE =0.55 0.721 CA =0.88 CR =0.88 AVE =0.55 Item 8 0.655 0.784 0.797 0.753 0.724 0.749 Item 9 0.720 0.520 0.836 0.821 0.829 0.789 Item 10 0.622 0.840 0.729 0.686 0.850 0.735 Item 11 0.789 0.700 0.737 0.838 0.867 0.805 Item 12 0.647 0.772 0.580 0.578 0.572 0.644 BItem 13 0.831 CA =0.94 CR =0.94 AVE =0.75 0.703 CA =0.85 CR =0.86 AVE =0.55 0.656 CA =0.87 CR =0.87 AVE =0.58 0.664 CA =0.84 CR =0.85 AVE =0.54 0.772 CA =0.86 CR =0.86 AVE =0.55 0.725 CA =0.89 CR =0.89 AVE =0.61 Item 14 0.905 0.826 0.731 0.775 0.681 0.789 Item 15 0.916 0.773 0.763 0.850 0.759 0.827 Item 16 0.832 0.728 0.828 0.660 0.768 0.788 Item 17 0.842 0.663 0.807 0.705 0.713 0.772 EItem 18 0.681 CA =0.86 CR =0.86 AVE =0.52 0.539 CA =0.86 CR =0.86 AVE =0.50 0.730 CA =0.085 CR =0.85 AVE =0.50 0.763 CA =0.89 CR =0.89 AVE =0.58 0.666 CA =0.88 CR =0.88 AVE =0.56 0.680 CA =0.87 CR =0.87 AVE =0.53 Item 19 0.761 0.622 0.777 0.807 0.659 0.773 Item 20 0.851 0.808 0.814 0.821 0.850 0.821 Item 21 0.458 0.608 0.707 0.695 0.702 0.657 Item 22 0.841 0.809 0.479 0.723 0.755 0.707 Item 23 0.646 0.820 0.677 0.749 0.817 0.732 RItem 24 0.785 CA =0.80 CR =0.80 AVE =0.51 0.666 CA =0.79 CR =0.80 AVE =0.50 0.818 CA =0.84 CR =0.84 AVE =0.58 0.730 CA =0.86 CR =0.87 AVE =0.63 0.795 CA =0.87 CR =0.87 AVE =0.62 0.781 CA =0.84 CR =0.84 AVE =0.57 Item 25 0.526 0.680 0.674 0.702 0.747 0.657 Item 26 0.705 0.808 0.796 0.831 0.774 0.774 Item 27 0.805 0.675 0.745 0.887 0.831 0.796 Notes: CA: Cronbach Alpha; CR: Composite Reliabily; AVE: Average Variance Extracted. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 6 same configuration of factor loadings in different countries (Horn & Mcardle, 1992). We ran a MultiGroup Confirmatory Factor Analysis (MGCFA) for all the data together in one dataset. Table 4 shows that the Bentler-Bonett non-normed fit index (BBNNFI), comparative fit index (CFI), and incremental fit index (IFI) are all above 0.80, which indicates a good model fit (Srinivasan et al., 2002). 3 In addition, Table 4 shows the root mean square error of approximation (RMSEA) for the model. Its value lower than 0.095 indicates good fit (Hu &Bentler, 1995; Kline, 2005). Moreover, all factor loadings are greater than 0.40, which is the cut-off demonstrating adequate levels of fit (Ford et al., 1986; Nunnally &Bernstein, 1994), all ps <.001. Therefore, Tables 3 and 4 provide evidence of convergent validity and internal consistency of the FIBER scale for all the datasets. Thus, factors’structure does not differ in the five cultural groups (Lacko et al., 2022; Widaman &Reise, 1997). Evidence of discriminant validity is necessary to ensure configural invariance as well. Accordingly, we followed two criteria. First, we applied the Fornell-Larcker criterion that compares the square root of the AVE values with the latent variable correlations. To show discriminant validity, the square root of each dimension’s AVE of an instrument should be greater than its highest correlation with any other dimension. The heterotrait-monotrait (HTMT) criterion is the second and more conservative approach. The HTMT is computed as the geometric mean of the heterotrait-heteromethod correlations divided by the average of the monotrait-heteromethod correlations. If the HTMT value is below the (conservative) threshold value of 0.85, discriminant validity would be supported (Hair et al., 2017). Table 5 shows evidence of discriminant validity 4 (for all countries together) with both the Fornell-Larcker criterion and HTMT criterion. This implies that the FIBER scale has discriminant validity, that is, each dimension of the FIBER scale is unique and captures phenomena not represented by other dimensions in the scale. These results confirm the configural invariance of the FIBER scale. This means that family owners in all five countries conceptualize the preservation of their SEW in same manner. The second step involves testing whether the various dimensions and the items associated with each dimension are understood in the same way across countries, that is, testing the metric invariance of the FIBER scale. Metric cross-country invariance requires that the factor loadings between items and factors are invariant across nations. It is tested by restricting the factor loading of each item on its corresponding factor to be the same across groups. Once the factor loading of each item is restricted, we test if the difference between the configurational model (without restrictions) and metric invariance model (with restrictions) is significant. If it is, the metric invariance does not exist. In our case, the configural invariance model has a χ 2 (1570) =3806.87, while the metric invariance model has a χ 2 (1678) =3892.16. Thus, the difference between the models is (Δ χ 2 (108) =85.293, p >.10). Therefore, as the metric invariance model is not significantly worse than the fit of the configural invariance model, and as the RMSEA and other indicators of model fit have improved or at least remained equal to those in the configural model (see Table 6), metric invariance is supported. This means that the FIBER construct has the same metric and same meaning across groups. While metric invariance implies that the instrument used to measure the construct has been perceived similarly across national cultures (Wernsing, 2014), it does not imply that the scores of the instrument can be meaningfully compared across groups. Such comparisons are meaningful only if the items exhibit “scalar”invariance (Meredith, 1993; Steenkamp &Baumgartner, 1998). Scalar invariance show that item scores from different groups have the same scaling, origin, and interpretation across groups. To assess scalar invariance, both factor loadings and item intercepts are constrained to be equal across countries. We tested the scalar invariance of the FIBER scale by constraining all factor loadings and item intercepts. The model of metric invariance (with only factor loading of each item constrained) has a χ 2 (1678) =3892.16, while the model of scalar invariance (with both factor loadings and item intercepts constrained) has a χ 2 (1766) =187.842. Once we have the fit of the metric model and scalar model, we check if the difference between the fit of the two models is significant. If it is, scalar invariance would not be established. In our case, the difference between the two models is statistically significant (Δ χ 2 (88) =187.842, p <.010), such that we cannot ensure the scalar invariance of the SEW scale (see Table 6). Ideally, it would be preferable SEW measures that show scalar invariance, wherein the factor structure of the instrument is fully equivalent across countries. However, several researchers have emphasized that full measurement invariance does not hold, and thus represents too strict a requirement for group comparisons, especially in cross-country studies (Muth´ en &Christoffersson, 1981; Schmitt &Kuljanin, 2008; Steenkamp &Baumgartner, 1998). Therefore, Byrne et al. (1989) and Steenkamp and Baumgartner (1998) proposed that it is sufficient to have a minimum of two invariant loadings and intercepts to ensure partial scalar invariance, that is, it is not necessary to have the constraints of all the loadings and intercepts being equal and invariant across groups (Millsap &Kwok, 2004). Hence, as the last step, we tested the partial invariance of the FIBER scale. Partial scalar invariance is supported when the parameters of at least two indicators are equal across countries (Steenkamp &Baumgartner, 1998). This means we must ensure that at least two items per latent construct have equal loading and intercepts (Byrne et al., 1989). In our case, more than two indicators are constrained in terms of loadings and intercepts, and are equal across countries. Thus, partial scalar invariance was supported for the FIBER scale across all countries. It indicates that the interpretation of at least one question from each dimension differed across the groups. However, at least two questions from each dimension did not differ across the groups (that is, had the same meaning). Thus, partial scalar invariance of the FIBER scale ensures meaningful and valid group comparisons (Byrne et al., 1989; Steenkamp &Baumgartner, 1998). Scalar measurement invariance allows researchers to at least compare the group-specific (standardized) coefficients of the relationships in the structural model. Otherwise, when less than two items per latent variable (dimension) have equal loadings and/or intercepts, multigroup comparisons in SEM may be problematic (Henseler et al., 2016). Our test of the measurement invariance of the FIBER scale suggests that family owners in all five countries conceptualize the preservation of their SEW in same manner (i.e., the FIBER scale shows configural invariance). Further, the scale exhibits metric invariance, signifying that family owners across these countries respond to the scale’s dimensions consistently. Last, the confirmation of partial invariance underscores that the operational definition of SEW preservation using the FIBER Table 4 Model fit of the FIBER scale. ChiSquare Test Degrees of Freedom BBNNFI CFI IFI RMSEA Canada 749.872 314 0.932 0.940 0.940 0.053 Mexico 574.249 314 0.826 0.845 0.848 0.081 Saudi Arabia 762.277 314 0.911 0.921 0.921 0.060 Spain 753.218 314 0.827 0.845 0.847 0.094 Vietnam 727.875 314 0.893 0.904 0.905 0.067 All together 1422.08 314 0.938 0.944 0.944 0.049 3 The goodness of fit describes how well it fits a set of observations. In other words, how close the model-implied covariance matrix approximates the observed covariance matrix (Garson, 2016). Measures of goodness of fit typically summarize the discrepancy between observed values and the values expected under the model in question (Byrne, 2006). If a model has a good fit, it is supposed that the model quality is good and in our case that the FIBER scale is reliable. 4 The discriminant validity for all the countries independently is available on request. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 7 scale remains consistent across diverse cultural settings. To obtain further evidence of cross-cultural applicability of the FIBER scale, we ran measurement invariance tests by looking at company size and generation in control. 5 The scale shows partial scalar invariance across size and generations (see Table 7a and Table 7b). At least two indicators are constrained in terms of loadings and intercepts and are equal in the two multigroups in each dimension (one multigroup considers the size of the company, and the other considers the generation in control of the family firm). To further test the cross-country validity of the scale, we tested its nomological validity. Its importance is well documented in the literature (e.g., Cronbach &Meehl, 1955). Following Lee et al. (2019) and Ellinger et al. (2011), we ran the structural model and compare the chi-squares and the fit of the model with the measurement model. If the measurement model has a significantly better fit than the structural model (Ellinger et al., 2011; Lee et al., 2019), then the nomological validity is not tested (see Table 8a and Table 8b); or in our case, the models are not significantly different (1423.09 –1422,8 =1,01 not significant ). To further verify the nomological validity of the FIBER scale, we follow Lee et al. (2019). As can be seen, in Table 8a, all correlations among the different dimensions are significant and positive (p<0.01). These findings provide support the nomological validity (Lee et al., 2019) of the FIBER measure by indicating that all the dimensions of the FIBER scale are significantly positively correlated to one another. For researchers and practitioners, this verification implies that all the dimensions of the FIBER scale need to be understood with consideration of the variables’independency and high correlations. We conducted an additional test to explore the connections between SEW dimensions and other relevant and “popular”constructs used in family business literature as performance (Hern´ andez-Linares et al., 2020; Sanchez-Famoso et al., 2015). 6 In this sense, performance is measured using three questions using items adopted from Sorenson et al. (2009) with an alpha Cronbach equal to 0.82. As BBNNFI equals to 0.98; CFI equals to 0.98; IFI equals to 0.97, RMSEA equals to 0.059 and R 2 is 0.38, we can conclude that the structural model has an excellent fit (Kline, 2005). Thus, the accurate recomputation of covariances is possible (Bentler, 1990). All the dimensions have a significant effect on performance: i) “F”dimension (β=0.26; t-value =2.03); ii) “I” dimension (β=0.55; t-value =3.92); iii) “B”dimension (β=0.31; t-value =2.23); iv) “E”dimension (β=0.32; t-value =2.24); and v) “R” dimension (β=0.28; t-value =2.14). These findings show the nomological network of the FIBER scale. With these findings, we can confidently assert that the FIBER scale is robust and can be meaningfully compared across various national cultural contexts. This sets the stage for us to embark on making crosscountry comparisons of the results derived from the scale within these countries. 3.1. Conducting comparative SEW research: Examining the cross-country variations of the FIBER scale The analyses conducted in the preceding section provide support to the robustness of the FIBER scale, indicating its suitability for meaningful comparison across diverse national contexts. Importantly, we can confidently attribute the discrepancies identified in the dimensions of the scale in this section to differences arising from varying cultural and institutional conditions. Table 9 presents the mean scores for each SEW dimension across the five sampled countries. Although exploratory, the table provides preliminary evidence on how national context shapes the significance attributed to different SEW dimensions. As shown in Table 9, Mexico ranks highest in both, the Family Control and Renewal of Family Bonds, dimensions of SEW. Table 1a reveals that Mexico is characterized by a collectivistic and high powerdistance society. This observation aligns with Gomez-Mejia et al. (2024), suggesting that in Latin American and Caribbean (LAC) countries, which are characterized by high collectivism and power distance, family businesses prioritize extended family concepts. We speculate that these cultural aspects of the context foster family owners’unencumbered discretion to provide jobs for a greater number of relatives. They also facilitate them to obtain perquisites to maintain the extended family lifestyle. Moreover, a context where the family is conceptualized as extended also increases the importance of the “R”dimension, that is, the family’s concern to transfer the business to the next generation, as the Table 5 Discriminant validity (Fornell-Larcker criterion and HTMT criterion). FornellLarcker Criterion HTMT 85 Criterion F I B E R F I B E R F 0.789 F  I 0.487 0.790 I 0.550  B 0.350 0.342 0.829 B 0.394 0.386  E 0.115 0.203 0.186 0.781 E 0.134 0.230 0.208  R 0.377 0.391 0.391 0.196 0.818 R 0.438 0.454 0.458 0.222  Note: In the Fornell-Larcker criterion, the diagonal represents the square root of AVEs. Table 6 Test of measurement invariance of the FIBER scale. Chi-Square Test Degrees of Freedom Chi-Square ΔDegrees of Freedom Δp-Value RMSEA BBNNFI CFI IFI Single Group Solutions Canada 749.872 314    0.053 [0.048–0.058] 0.932 0.940 0.940 Mexico 547.249 314    0.081 [0.069–0.092] 0.826 0.845 0.848 Saudi Arabia 762.277 314    0.060 [0.054–0.065] 0.911 0.921 0.921 Spain 753.218 314    0.094 [0.085–0.102] 0.802 0.823 0.847 Vietnam 727.875 314    0.067 [0.060–0.073] 0.893 0.904 0.905 Measurement Invariance Equal Form 3806.87 1570    0.070 [0.067–0.073] 0.897 0.908 0.909 Equal Factor Loadings 3892.16 1678 85.293 108 0.948 0.067 [0.064–0.070] 0.905 0.909 0.909 Scalar Invariance Equal Intercepts 6617.92 1796 2725.763 118 0.000 0.098 [0.095–0.100] 0.788 0.806 0.807 5 We thank the anonymous reviewers for this insightful suggestion. 6 We thank one of our anonymous reviewers for this insightful suggestion. V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 8 extended family implies that more family members are available to lead the firm now or in the distant future. The relatively higher importance of these two dimensions in Mexico is also understood in the context of a high power-distance society, where there is a significant power gap between authority figures and subordinates. In a family business context, this may result in a more autocratic leadership style and a culture of respecting and complying with family authority, thereby amplifying the family’s control over the business both in short and long run. Interestingly, Mexico scores lowest in the “I”dimension, contrary to expectations in a collectivistic society where “social life revolves around the family and close relations”(Field, 1985, p. 87). In Saudi Arabia, however, this dimension is prioritized. While both countries share low individualism, differences in institutional stability may explain this variance. The greater instability and corruption that characterize the Mexican landscape compared to Saudi may explain why family owners may exercise caution in drawing excessive attention to the family’s identity. Unforeseen events or political changes could negatively impact both the family’s and business’s reputation. More importantly, as Table 7a Test of measurement invariance of the FIBER scale (taking into account company size). Chi-Square Test Degrees of Freedom Chi-Square Δ Degrees of Freedon Δ pValue RMSEA BBNNFI CFI IFI Single Group Solutions Companies (employees > 100) (n =754) 914.64 314 0.052 [0.048–0.055] 0.931 0.938 0.939 Companies (employees < 101) (n =710) 903.07 314 0.051 [0.047–0.055] 0.933 0.940 0.940 Measurement Invariance Equal Form 1844.71 628 0.051 [0.044–0.054] 0.932 0.939 0.939 Equal Factor Loadings 1881.42 655 36.71 27 0.101 0.051 [0.048–0.053] 0.934 0.938 0.938 Scalar Invariance Equal Intercepts 3.074,96 692 1193.54 37 0.000 0.070 [0.067–0.071] 0.871 0.882 0.882 Table 7b Test of measurement invariance of the FIBER scale (taking into account generation in control). Chi-Square Test Degrees of Freedom Chi-Square Δ Degrees of Freedon Δ pValue RMSEA BBNNFI CFI IFI Single Group Solutions Group 1 (Generation 1 and 2) (n =907) 1015.39 314 0.050 [0.047–0.054] 0.937 0.944 0.944 Group 2 (Generation >2) (n =557) 812.60 314 0.053 [0049–0.058] 0.925 0.933 0.933 Measurement Invariance Equal Form 1827.99 628 0.051 [0.048–0.054] 0.933 0.940 0.940 Equal Factor Loadings 1866.43 655 38.44 27 0.071 0.051 [0.048–0.054] 0.933 0.938 0.938 Scalar Invariance Equal Intercepts 3160.64 692 1294.21 37 0.000 0.070 [0.067–0.072] 0.872 0.883 0.883 Table 8a Nomological validity of the FIBER scale. Measurement Model Structural Model Chi-Square 1422.08 1423.09 Degrees of Freedom 314 313 BBNNFI 0.938 0.935 CFI 0.944 0.942 IFI 0.944 0.942 RMSEA 0.049 0.050 Table 8b Latent variable intercorrelations in the FIBER nomological network. F I B E I0.543 B0.398 0.386 E0.124 0.227 0.209 R0.430 0.441 0.437 0.221 Table 9 Cross-country comparison of the FIBER dimensions. SEW DIMENSIONS Canada Mexico Spain Vietnam Saudi Arabia SEW value Rank SEW value Rank SEW value Rank SEW value Rank SEW value Rank (F)amily Control and Influence 4.337 2 4.722 2 4.443 2 3.948 1 3.890 3 (I)dentification of Family Members with the Firm 4.375 1 3.319 5 4.426 3 3.768 4 4.085 1 (B)inding Social Ties 4.26 3 4.387 3 4.234 4 3.927 3 3.956 2 (E)motional Attachment of Family Members 3.952 5 4.058 4 3.292 5 3.436 5 3.826 5 (R)enewal of Family Bonds Through Dynastic Succession 3.998 4 4.883 1 4.478 1 3.94 2 3.874 4 V. Sanchez-Famoso et al. Journal of Family Business Strategy 16 (2025) 100647 9