Abstract
Interpersonal conflict research has traditionally emphasized cultural determinants of conflict styles while overlooking their context-sensitive nature. Addressing this gap, our study examines how generational differences, cultural values, and socio-demographic factors relate to romantic conflict management among Chinese women. The evolving conflict styles reflect how macro-social changes influence interpersonal relationships, capturing the dynamic interplay between tradition and modernity. Drawing from Rahim and Ting-Toomey’s conflict frameworks, we examined seven conflict styles through a multi-generational online survey (N = 469): dominating, avoiding, obliging, productive integrating, third-party help, emotional expression, and passive-aggressive. Results revealed distinct generational patterns, with Gen X (Chinese women born in 1960–1979) reporting higher use of power-oriented approaches (e.g., dominating and avoiding) compared to younger generations, who indicated greater inclinations toward problem-solving strategies (e.g., third-party help and obliging). Cultural values emerged as significant correlates, with individualism positively associated with both confrontational (dominating) and indirect strategies (passive-aggressive), whereas collectivism was positively related to cooperative approaches (productive integrating and obliging) but negatively related to avoidance. Socio-demographic characteristics, including argument frequency, education, and relationship status, were also significantly related to conflict styles, highlighting the influence of contextual variability in romantic conflict management. Our findings offer both theoretical insights and practical guidance for romantic relationship counseling and conflict resolution interventions.
Keywords
Conflict management within romantic relationships plays a crucial role in relationship satisfaction and emotional well-being (Johnson et al., 2015; Meyer & Sledge, 2022; Szachter et al., 2025). In contemporary China, conflict resolution reflects broader social changes, with women driving these shifts in romantic relationships (Song & Ji, 2020). While traditional Chinese society was deeply patriarchal, rooted in Confucian values that formed one of the most restrictive systems for women in history (Greenhalgh, 1985), dramatic social transformations—including the 1949 Communist Revolution and the 1978 Opening-up Policy—have fundamentally reshaped women’s roles and conflict management approaches in relationships (Song & Ji, 2020).
The focus on Chinese women is particularly crucial as they face unique challenges and opportunities. They are at the forefront of embracing egalitarian values while still navigating traditional expectations. This coexistence of modern and traditional norms in the Chinese context, which Ji (2015) terms “modern-tradition mosaic” (p. 1036), creates potential relationship tensions that their male counterparts, who generally show more resistance to modern gender roles, may not experience (Qian & Li, 2020; Song & Lai, 2020). Ji (2017) further conceptualizes this phenomenon as a form of “mosaic temporality” (p. 3), highlighting the institutional-cultural mismatch in which rapid institutional changes (e.g., legal reforms, educational expansion, labor market participation) outpace slower transformations in cultural norms, values, and practices. This mismatch places Chinese women at the intersection of competing expectations: while institutions increasingly encourage gender equality, cultural traditions continue to reinforce more conventional gender roles. Despite their pivotal role in driving social change and their distinct experiences in negotiating between traditional and modern values, Chinese women’s conflict management strategies remain understudied.
While previous research has demonstrated how conflict styles vary across cultures (Cai & Fink, 2002; Ting-Toomey et al., 2000), less attention has been paid to populations undergoing rapid cultural transitions. In the Chinese context, recent studies have begun to explore such changes. For example, Li et al. (2017) documented shifting marital conflict patterns over time, and Li et al. (2020) showed how interparental conflict shaped young adults’ romantic relationship quality through emotional security and parent-child attachment. However, these studies have not directly examined how Chinese women navigate romantic conflict in the context of rapid social change.
This study addresses this gap by examining how multiple factors are associated with Chinese women’s conflict management styles. First, generational cohorts reflect distinct socialization experiences (Todd, 2005), with older generations emphasizing Confucian harmony (Zhang et al., 2005) while younger generations, influenced by globalization, pursue greater autonomy and self-expression (Chen, 2015). Second, cultural values play a crucial role as China transitions from collectivism to individualism, driven by the opening-up and one-child policies (Chen, 2015). Third, socio-demographic characteristics are increasingly important given women’s rising socioeconomic status and education levels—a transformation that has particularly impacted women’s relationship dynamics and their approach to conflict resolution (Qian & Li, 2020; Song & Lai, 2020).
This research contributes both theoretically and practically. Theoretically, it offers a multi-dimensional framework for understanding conflict management styles in non-Western societies undergoing rapid social transformation, integrating generational, cultural, and socio-demographic perspectives within romantic relationship contexts. Practically, understanding Chinese women’s conflict management approaches can inform policies and interventions promoting gender equality and relationship well-being in contemporary Chinese society, particularly given their unique position at the intersection of traditional expectations and modern aspirations (Qian & Li, 2020). By focusing on Chinese women specifically, this study addresses critical gaps in our understanding of how individuals navigate relationship conflicts while simultaneously managing broader societal transitions in gender roles and cultural values.
Refined theoretical framework of conflict styles
Research on conflict management styles has been extensively developed in organizational communication, with Rahim’s (1983, 1985) five-style model remaining foundational to the field. This model identifies five primary approaches to managing interpersonal conflicts: dominating, avoiding, integrating, compromising, and obliging. These styles represent different strategies individuals employ when facing conflict (Cai & Fink, 2002), including highly confrontational approaches that prioritize self-interest and often lead to win-lose outcomes (dominating), withdrawal behaviors aimed at preventing conflict escalation (avoiding), and strategies prioritizing relational harmony by yielding to others’ needs, often at the expense of personal interests (obliging).
However, recent scholarship has suggested the need for refinements to this traditional framework, particularly regarding the conceptual overlap between integrating and compromising styles. While integrating involves actively seeking mutually beneficial solutions through negotiation, compromising focuses on reaching middle-ground agreements through shared concessions (Thomas, 1976). Cai and Fink (2002) noted that the compromising style lacks conceptual distinctiveness, as it represents a moderate concern for both self and others without the proactive problem-solving characteristic of integration. Supporting this view, Liu et al.’s (2024) interviews with intercultural couples revealed that partners struggled to differentiate between integrating and compromising approaches during actual conflict episodes. To address these theoretical and practical challenges, this study combines these styles into a unified, productive integrating style, which emphasizes active collaboration over passive concessions. This refinement not only enhances the framework’s conceptual precision and applicability to romantic relationships but also aligns with Li et al.’s (2017) and Liu et al.’s (2024) observations that individuals strategically alternate between multiple conflict styles rather than adopting them exclusively.
While Rahim’s (1983) model was originally developed in organizational contexts, and conflict styles such as dominating, avoiding, and obliging are applicable across various interpersonal settings (e.g., family, friends, coworkers), romantic relationships constitute a distinctive context for examining how these strategies operate. Romantic conflicts often involve deeper emotional interdependence, identity negotiation, and relational boundary management than disputes in other relationships (Diener et al., 2018). Unlike familial relationships structured by hierarchies and obligatory bonds, or professional relationships governed by formal roles, romantic partnerships are characterized by voluntary commitment, mutual dependency, and emotional intimacy—dynamics that fundamentally shape how conflict strategies are selected, enacted, and interpreted (Klein & Milardo, 2000). Consequently, romantic conflicts are typically more emotionally charged and private, underscoring the importance of examining conflict strategies as they function within romantic contexts.
To better capture these relational dynamics in the Chinese cultural context, this study incorporates Ting-Toomey et al.’s (2000) Conflict Style Dimension (CSD) framework, which adds three styles particularly relevant in collectivistic cultures and romantic contexts: emotional expression, third-party help, and passive aggression. These additions capture important nuances in conflict management, especially in collectivistic cultures and romantic relationships where indirect or emotion-centered approaches often predominate (Ting-Toomey & Oetzel, 2001). Taken together, this study builds on a refined conflict styles framework comprising seven strategies: dominating, avoiding, obliging, productive integrating, emotional expression, third-party help, and passive aggression. Rather than claiming these strategies as exclusive to romantic relationships, our framework examines how the distinctive relational dynamics of romantic partnerships shape their function and consequences.
Generation and conflict management in the Chinese context
Generations have become increasingly central to research in the Chinese context due to their role in driving social changes (Hu & Scott, 2016). Scholars have proposed various generational categorizations: some defined those shaped by the Cultural Revolution (1966–1976) as a “Lost” or “Delayed” generation (Chen, 1999; Roland & Yang, 2017), while Sun and Wang (2010) classified cohorts based on major social movements following the establishment of the People’s Republic of China in 1949. Erickson (2009) offered chronological categories, including Traditionalists (1928–1945), Boomers (1946–1960/1964), Generation X (Gen X; 1961/1965–1979), and Generation Y (Gen Y; 1980–1995). More recently, scholars have identified Generation Z (Gen Z) as those born between 1990 and the early 2000s (Chen & Yang, 2023).
Drawing from these varied definitions, this study examines three generational groups in the Chinese context: Gen X (1960-1979), Gen Y (1980-1995), and Gen Z (1996-2005). These groups were derived by synthesizing prior Chinese (e.g., Erickson, 2009) and international scholarship (e.g., Dimock, 2019) and aligning the cohort boundaries with major sociopolitical transitions in China. Specifically, Gen X experienced their formative years (young adulthood) during the Reform and Opening Up era of the late 1970s and 1980s, a period marked by de-collectivization, decentralization, and legalization, which distinguished their cultural values from those of Traditionalists and Boomers (Sun & Wang, 2010). We adopt Erickson’s (2009) categorization of Gen Y and extend Gen Z to include those born through2005 1 , incorporating the newest cohort of emerging adults. While we use the labels Gen X/Y/Z for clarity and comparability with broader generational research, the categories themselves are grounded in historical and cultural milestones specific to the Chinese context.
Socialization and life course theories suggest that each cohort’s experiences within its historical environment foster distinct collective memories and habitus (Elder, 1994). The dramatic economic, political, and cultural transformations in Chinese society have prompted researchers to examine cross-generational value changes. Gao et al. (2024) observed that older generations tend to be more materialistic and stability-oriented, suggesting they may adopt more conservative and harmony-oriented conflict management styles, such as obliging and avoiding. Conversely, younger generations, having experienced greater economic prosperity, tend toward postmaterialist values emphasizing autonomy and self-expression (Sun & Wang, 2010), potentially predisposing them to more assertive conflict management styles like dominating and expressing emotions. Additionally, Gao et al. (2024) found that younger generations hold more positive attitudes toward the openness of private spheres. As such, with improved access to public education, younger generations, particularly Gen Z, may be more likely to seek third-party help from others, such as professional therapists and online communities.
Zhang et al. (2005) provided empirical evidence for generational differences in conflict management styles, demonstrating that older adults tend to favor accommodation while younger adults prefer more proactive problem-solving approaches. However, their study focused on organizational contexts and did not address gender specificity. To address the gap in understanding generational differences among women in romantic relationships, this research poses the following research question:
Culture orientations and conflict management in romantic relationships
Cultural orientations significantly influence communication behaviors in individualistic (IND) and collectivistic (COL) cultures. Research shows that IND cultures tend to favor low-context communication (Gudykunst et al., 1996), characterized by explicit, direct messages in which meaning is conveyed primarily through the content itself (Hall, 1976). In contrast, COL cultures often employ high-context communication (Gudykunst et al., 1996), where messages are more implicit and indirect, with meaning embedded in the speaker, the relationship, or the broader sociocultural context (Hall, 1976). This distinction is particularly evident in intercultural relationships, where Chinese partners typically adopt more indirect, subtle communication styles compared to their Western counterparts (Halford et al., 2018).
Conflict in romantic relationships involves perceived discrepancies between partners in expectations, goals, face concerns, and interaction styles (Ting-Toomey & Oetzel, 2001). The two key frameworks reviewed above help understand how cultural values shape conflict behaviors. First, Rahim’s (1983) dual-concern model differentiates conflict styles based on concern for self and concern for others, with IND cultures typically aligning with assertive styles while COL cultures favor cooperative approaches. Second, Ting-Toomey’s (1988) face-negotiation theory demonstrates how face needs influence conflict strategies across cultures, complementing Thomas’s (1976, 1992) classification of conflict styles by assertiveness and cooperativeness.
The dominating conflict style is characterized by high self-concern and self-face maintenance, and research shows it is more prevalent in low-context cultures (Croucher et al., 2012). Ting-Toomey et al. (1991) found that the U.S. group reported significantly higher use of dominating compared to the Japanese and Korean groups. However, they also noted differences within East Asian countries, with Chinese participants reporting a higher use of dominating than the Japanese and Korean groups. Based on the existing evidence, we hypothesize that:
Chinese women with higher IND are more inclined to adopt dominating conflict styles.
Avoiding and obliging styles represent passive cooperative approaches, both reflecting high cooperativeness and low assertiveness (Antonioni, 1998). Obliging involves self-sacrifice for others’ needs, resulting in a lose-win outcome, while avoiding leads to mutual withdrawal, creating a lose-lose situation as both parties’ needs remain unmet (Antonioni, 1998; Ting-Toomey et al., 1991). Research consistently shows that collectivists tend to favor obliging, as demonstrated by studies showing higher obliging usage among people from China, Taiwan, and Japan compared to those from Korea and the U.S. (Ting-Toomey et al., 1991). While avoiding is common across cultures, Chinese and Taiwanese individuals show higher adoption rates compared to Japanese, Korean, and American counterparts (Ting-Toomey et al., 1991). Importantly, Asian cultures view these styles positively, unlike Western contexts where they may be perceived negatively (Ting-Toomey et al., 2000). For instance, Gomez and Taylor (2018) found that while Americans generally perceive avoiding as ineffective, Chinese participants view it as demonstrating concern for both oneself and others. In summary, there is consistent evidence of a significant positive association between COL and the preference for obliging, while the relationship between COL and the use of avoiding remains less clear. As such, the following hypothesis and research question are proposed:
Chinese women with stronger COL are more inclined to adopt obliging conflict styles in their romantic relationships.
Productive integrating styles combine integrating and compromising strategies, which are particularly prevalent in collectivistic societies due to their focus on harmony and avoiding confrontation to preserve social cohesion (Cai & Fink, 2002). These styles emphasize harmony and mutual satisfaction, aligning with collectivistic values, as evidenced in Kirkbride et al. (1991) that Hong Kong Chinese showed a preference for compromising compared to British expatriates. In traditional Chinese families, collectivistic conflict styles such as compromising are pervasive (Wang & Liu, 2010), often requiring individuals to sacrifice personal needs for the collective good (Cai & Fink, 2002). Interestingly, the integrating style shows some variation across cultures, with both U.S. and Chinese groups reporting significantly higher usage compared to Japanese groups (Ting-Toomey et al., 1991). Based on these findings, we predict that:
Chinese women with stronger COL are more inclined to adopt productive integrating styles to manage conflict.
Lastly, three additional conflict styles are predominantly found in collectivistic cultures: third-party help, passive aggression, and emotional expression. Research by Zhang et al. (2012) demonstrates that in romantic relationships, Chinese participants prefer third-party help and passive aggression, while Americans favor overt expressions of anger or exiting the conflict when dealing with emotional issues. This preference for indirect conflict resolution is further evidenced in the literature, where studies (e.g., Kozan & Ergin, 1998; Wall & Blum, 1991) have consistently shown greater reliance on third-party involvement in conflicts within collectivistic societies. Gomez and Taylor (2018) provide more support, showing that collectivists often rely on their social networks to resolve conflicts in a non-confrontational manner. Ting-Toomey (1988) similarly argued that collectivists might prefer having others intervene rather than confronting issues directly, primarily to ensure face-saving during the process. Regarding emotional expression, Neuliep and Johnson (2016) found that Americans preferred more confrontational styles like dominating and emotional expression, aligning with their low-context, individualistic culture, while collectivistic cultures favored more accommodating approaches. Based on the literature, the following hypotheses are proposed:
Chinese women with higher COL are more inclined to adopt third-party help (H4a) and passive-aggression (H4b) to manage conflict in romantic relationships.
Chinese women with higher IND are more inclined to adopt emotional expression to manage conflict in romantic relationships.
Socio-demographics and conflict management in romantic relationships
Women’s social status in China has experienced significant changes recently, creating new challenges in family settings as cultural expectations evolve alongside traditional norms (Qian & Li, 2020). These shifts make it crucial to understand how socio-demographic factors influence conflict management preferences. Research has shown that education, gender, and work tenure significantly impact conflict management styles in workplace settings (İslamoğlu et al., 2008). Jang and Kim’s (2018) qualitative study further demonstrated that women’s educational level, work experiences, and financial status are crucial in shaping their conflict management approaches. Early family experiences, particularly sibling relationships, also play a vital role. Shalash et al. (2013) found that conflict styles developed among siblings often transfer to adult relationships.
Furthermore, living arrangements and parental status significantly influence conflict management approaches. Jamieson (2020) found that living alone can promote more honest dialogue and gender equality, potentially leading to more assertive conflict styles. Delatorre and Wagner (2018) observed that cohabiting women tend to use more confrontational strategies than married women, while parents of both genders showed higher avoiding scores than non-parents.
Meanwhile, education and financial status consistently emerge as significant factors. Studies show that individuals with higher education and income levels tend to employ more collaborative conflict styles (Li et al., 2017; İslamoğlu et al., 2008). Jang and Kim’s (2018) research confirmed that women with higher education, work experience, and income often adopt compromising conflict styles, using their financial independence and negotiation skills to maintain relationship harmony while preserving family dynamics.
Additional contextual factors include religiosity and relationship length. Prior research indicates that religiosity is associated with greater use of constructive conflict resolution strategies, while relationship length corresponds with systematic shifts in conflict management patterns over time, including increased conflict avoidance and reduced use of destructive strategies, particularly among women (Delatorre & Wagner, 2018). Recent work further highlights that engagement with relationship conflict varies across relational contexts and stages, underscoring the importance of considering temporal and contextual factors when examining conflict management preferences (Jöhnk et al., 2025).
Notably, research on the relationship between conflict management styles and demographic variables has shown inconsistent results, especially regarding gender (Cetine & Hacifazlioglu, 2004; Rahim, 1983). These inconsistencies, combined with China’s evolving cultural landscape and changing gender dynamics, necessitate further research into how socio-demographic factors influence conflict management preferences among Chinese women. This raises two key research questions:
Method
Participants
Before collecting the data, a priori power analysis using G*Power 3.1.9.7 indicated that a sample size of 107 participants would be sufficient to detect medium-sized effects (f 2 = 0.15) with 95% power (α = .05) in a hierarchical multiple regression with up to 13 predictors, including main effects and interactions.
In May 2024, we recruited a sample of 469 participants via Wenjuanxing (Changsha Ranxing Information Technology Co., Ltd, Hunan, China), the largest online survey platform in China. Wenjuanxing, comparable to Qualtrics and SurveyMonkey, provides comprehensive online survey services for enterprises, research institutions, and individuals. The platform’s extensive user base facilitated targeted recruitment based on our inclusion criteria. Given the scope and focus of this study on Chinese women from three generations, and to ensure participants had recent and relevant experiences, eligibility criteria were: (1) Chinese women 2 , (2) aged 18–65 (born between 1960 and 2005), and (3) currently in a romantic relationship or in one that had ended within the 12 months prior to the survey. Individuals were excluded if they failed to meet these criteria. Using quota sampling, we obtained balanced representation across generational cohorts: Gen X (born 1960–1979, n = 156), Gen Y (born 1980–1995, n = 156), and Gen Z (born 1996–2005, n = 157).
Participants ranged in age from 18 to 62 years (M = 35.92, Mdn = 32, SD = 10.53) and resided across 27 provinces in mainland China (see Appendix A for a map of the data collection locations). The majority were employed full-time (77.2%, n = 362) or self-employed (3.4%, n = 16), representing over 60 different occupations, including education, healthcare, administration, finance, and service industries. Additionally, 3.6% (n = 17) of participants were homemakers, 7.9% (n = 37) were students, 5.3% (n = 25) were unemployed, and 2.6% (n = 12) were retired. Educational levels ranged from elementary school to doctoral degrees, with most participants holding a bachelor’s degree or higher (84.2%, n = 395). Around half of the sample (52.9%, n = 248) reported having siblings rather than being an only child. Of these, 62.5% (n = 155) had one sibling, 22.2% (n = 55) had two siblings, and 15.3% (n = 38) had three or more. Regarding residency arrangement, 8.3% (n = 39) lived alone, 37.3% (n = 175) lived with a partner only, 41.6% (n = 195) lived with a partner and others (e.g., family members or roommates), and 12.8% (n = 60) lived with others who were not their romantic partner.
Most participants reported having male partners (98.7%, n = 463), while a small proportion had female partners (1.3%, n = 6)3. The majority of their partners were of Han ethnicity (96.6%, n = 453), and Appendix A provides the detailed breakdown of all reported ethnic groups. Most partners were reported as employed full-time (79.1%, n = 371) or self-employed (6%, n = 28), with 1.7% (n = 8) being homemakers, 6.8% (n = 32) students, 3.8% (n = 18) unemployed, and 2.6% (n = 12) retired. Relationship lengths ranged from 1 to 610 months (50.83 years), with an average duration of 108.55 months (SD = 108.29) and a median of 54 months. Gen Z participants reported the shortest average relationship length (M = 33.41 months, SD = 19.94), followed by Gen Y (M = 86.01 months, SD = 70.39), and Gen X (M = 206.76 months, SD = 118.84).
Marital status was assessed to capture respondents’ current legal marital and cohabitation status, consistent with common categorizations in the Chinese context. Of the participants, 60.8% (n = 285) were married, 25.2% (n = 158) were single (i.e., never legally married, regardless of current romantic relationship status; this includes those currently dating but not cohabiting as well as those not currently in a relationship), 8.5% (n = 40) were cohabiting (living with a romantic partner but never legally married), 5.3% (n = 25) were separated or divorced, and one participant was widowed. The sample was nearly evenly split between those with children (53.9%, n = 253) and those without children (46.1%, n = 216). Among participants with children, most had one child (77.1%, n = 195), followed by two children (21.7%, n = 55), and a very small number had three (1.2%, n = 3). Financially, 27.1% (n = 127) of participants reported that both partners were independent and managed money separately; 42% (n = 197) indicated that they contributed equally and managed finances jointly; 28.4% (n = 133) reported being largely or completely financially dependent on their partners; and a small proportion (2.6%, n = 12) indicated that neither partner had income and that they were financially dependent on a third party (usually their parents). Detailed demographic characteristics by generational group are presented in Appendix A.
Procedures
This study employed a cross-sectional survey design. The online questionnaire comprised three sections: (1) an adapted version of the Individualism and Collectivism Scale (Triandis & Gelfand, 1998) to assess IND-COL cultural value orientation; (2) the adapted Rahim Organizational Conflict Inventory-II (ROCI-II; Rahim, 1983) and the Conflict Style Dimension (CSD) scale (Ting-Toomey et al., 2000) to measure conflict style preferences; and (3) a demographic questionnaire collecting information about age, educational level, marital status, occupation, and other relevant characteristics.
Participants were instructed to recall a specific disagreement in a current or past romantic relationship that had occurred within the previous 12 months and to answer the questions assessing their conflict management styles for that episode. Because such episodes are often emotionally salient and memorable, they can typically be recalled retrospectively with reasonable accuracy, and prior work has supported the validity of retrospective self-reports of conflict interactions (Backer-Fulghum & Sanford, 2015). This approach helped ensure that the reported conflict management behaviors were both meaningful and representative, as they were grounded in a concrete relational context and allowed us to capture the possibility of multiple strategies being employed simultaneously within a specific conflict episode.
An Institutional Review Board at the university where the authors are affiliated approved this study, and all participants provided informed consent before participation. Survey completion took approximately 20–30 minutes, and participants received a small incentive for their participation directly from the research company.
Measurements
Major variables, including cultural values and conflict styles, were measured on 5-point scales, with higher scores indicating greater levels of the measured construct. All scales were adapted from established instruments based on the theoretical frameworks and conceptual definitions of the constructs (Rahim, 1983; Ting-Toomey et al., 2000; Triandis & Gelfand, 1998), with refinements to accommodate the Chinese cultural context, following the Model for Culturally Contextualized Communication Measurement (MCM 3 ; Liu et al., 2022).
Confirmatory factor analyses (CFA) were conducted in JASP to assess the construct validity and internal consistency of both cultural values and conflict styles, which are theoretically conceptualized as latent variables. For conflict styles, the refined seven-factor model exhibited substantially better fit (CFI = .94, TLI = .92, RMSEA = .04, SRMR = .04) than other tested solutions, including the original eight-factor model (CFI = .65, TLI = .60, RMSEA = .09, SRMR = .15). For cultural orientations, the two-factor model representing individualism and collectivism also demonstrated good fit (CFI = .95, TLI = .93, RMSEA = .05, SRMR = .04), following established criteria (Hu & Bentler, 1999). These results confirm that our refined, theory-driven measurement models fit the empirical data well.
To ensure that the constructs were measured equivalently across generational groups, multigroup CFA was conducted to test measurement invariance for cultural orientations and conflict styles. For individualism and collectivism, configural invariance was first established with an acceptable model fit (CFI = .94, RMSEA = .054). Constraining factor loadings to be equal across generations did not significantly worsen model fit (Δχ2(14) = 14.90, p = .39; ΔCFI = .001; ΔRMSEA = .004), supporting metric invariance. Further constraining item intercepts likewise resulted in no significant loss of fit (Δχ2 (14) = 16.54, p = .28; ΔCFI = .005; ΔRMSEA = .002), indicating scalar invariance.
For the seven conflict styles, configural invariance was supported with adequate baseline fit (CFI = .93, RMSEA = .049). Imposing equality constraints on factor loadings did not significantly worsen model fit ( Δχ2(34)= 35.91, p = .38; ΔCFI = .000; ΔRMSEA = .001), supporting metric invariance. Scalar invariance was also established, as constraining item intercepts produced no significant decrement in fit (Δ χ2(34) = 38.16, p = .29; ΔCFI = .002; ΔRMSEA = .001). Together, these results indicate that both cultural orientations and conflict styles were measured equivalently across generational groups, permitting meaningful comparisons across cohorts.
Zero-order correlations, means, and standard deviations of major variables.
Note. *p < .05, **p < .01.
Cultural values
Cultural values were measured using an adapted version of the Individualism and Collectivism Scale (ICS; Triandis & Gelfand, 1998). The original ICS is a 16-item instrument using 9-point Likert scales (1 = never or definitely no; 9 = always or definitely yes) assessing four dimensions: vertical individualism/collectivism and horizontal individualism/collectivism. For this study, we modified the scale to use 5-point Likert responses for measurement consistency. Higher scores on both vertical and horizontal measures indicated greater individualism (IND) or collectivism (COL), respectively. Following CFA, we retained nine items across two subscales: IND (3 items, e.g., “I enjoy being unique and different from others in many ways”; Cronbach’s α = .56) and COL (6 items, e.g., “It is important for me to maintain harmony within my group”; α = .68).
Conflict styles
Conflict styles were assessed using an adapted version of the Rahim Organizational Conflict Inventory–II (ROCI-II; Rahim, 1983) and the Conflict Style Dimension (CSD) scale by Ting-Toomey et al. (2000), with culturally contextualized modifications for Chinese romantic relationships (Liu et al., 2022). The original ROCI-II comprises 35 items rated on 5-point scales, with higher scores indicating greater use of a particular style. Our adapted version measured seven distinct conflict management styles: dominating (3 items, e.g., “I used my power to win a competitive edge”; α = .65), avoiding (3 items, e.g., “I sucked it up and held my resentment in silence”; α = .63), obliging (4 items, e.g., “I gave in to the wishes of my partner”; α = .73), productive integrating 4 (5 items, e.g., “I tried to get us to work together to settle our differences”; α = .81), emotional expression (2 items, e.g., “I used my feelings to guide my conflict behaviors”; α = .53), third-party help (4 items, e.g., “I typically go through a close friend to settle our conflict”; α = .86), and passive-aggression (3 items, e.g., “I said nasty things about my partner to other people”; α = .64).
Socio-demographic variables
We collected comprehensive demographic information to describe the sample and control for potential confounding variables. These variables included age, gender, relationship length, marital status, argument frequency, children (number, sex, and age), residence location, living arrangement (alone, with partner, or other), employment status, financial status, siblings (number and sex), education level, and partner characteristics (nationality, ethnicity, gender, and employment status).
Plan of analysis
We adopted a multi-step analytical approach to examine the measurement structure, address our research questions, and test hypotheses. Prior to the main analyses, we assessed the clarity and cultural appropriateness of the survey items and validated the measures through confirmatory factor analyses (CFA) in JASP. Details of the CFA and measurement validation are reported in the Measurement section above.
After establishing the measurement model, we proceeded with the main analyses in SPSS. To address RQ1, we conducted a series of one-way ANOVAs to examine generational differences (Gen X, Y, and Z) in Chinese women’s use of conflict management styles in romantic relationships, with generational group as the independent variable and conflict management styles as the dependent variables. Post hoc Tukey’s HSD tests were performed where significant differences were detected.
To examine how cultural values and socio-demographic variables were associated with conflict styles (H1–H5 and RQ2), we conducted hierarchical multiple regression analyses. Predictors were entered in two blocks: socio-demographic variables (Block 1) and cultural orientations (Block 2), with each of the seven conflict styles examined as a dependent variable in separate models. No missing data were observed across any study variables; therefore, all cases were retained for analysis.
Specifically, Block 1 incorporates variables capturing key relational and structural contexts in which romantic conflict occurs, including: (1) participant’s life-course positioning and natal family background (age, number of siblings, and number of children), reflecting their life stage and early family socialization experiences relevant to conflict learning; (2) partner characteristics (partner’s age and employment status); (3) relationship structure and dynamics (relationship length, marital status, co-residence patterns, and argument frequency); and (4) economic positioning and resource arrangements within the relationship (participant’s education level, employment status, financial independence or dependence on the partner). Together, these variables provide important contextual grounding for understanding how women in our sample manage romantic conflicts.
Results
Generational differences in conflict management styles (RQ1)
RQ1 examines how conflict management styles vary among Chinese women from Gen X, Y, and Z in their romantic relationships. Results from one-way ANOVAs (see Figure 1), with generational groups as the independent factor and conflict management styles as the dependent variable, revealed significant generational differences in four conflict management styles: dominating [F(2,466) = 4.86, p = .008], avoiding [F(2,466) = 4.58, p = .011], obliging [F(2,466) = 5.32, p = .005], and emotional expression [F(2,466) = 7.14, p = .001]. Third-party help showed marginal significance [F(2,466) = 2.93, p = .055], whereas passive-aggressive and cooperative styles showed no significant generational differences. ANOVA results on conflict management styles across the three generational groups.
Post-hoc Tukey’s HSD tests showed that for dominating, Gen X (M = 2.90, SD = 0.85) scored significantly higher than both Gen Y (M = 2.68, SD = 0.74) and Gen Z (M = 2.64, SD = 0.77). For avoiding, Gen X (M = 2.40, SD = 0.79) reported significantly higher use than Gen Z (M = 2.13, SD = 0.84), whereas Gen Y (M = 2.22, SD = 0.76) did not differ significantly from either cohort. For emotional expression, Gen X (M = 3.38, SD = 0.84) reported significantly higher use than Gen Y (M = 3.00, SD = 0.93). Gen Z (M = 3.22, SD = 0.91) reported marginally higher use than Gen Y (p = .077) but did not differ significantly from Gen X. Lastly, for obliging, Gen Y (M = 3.10, SD = 0.78) reported significantly higher use than both Gen X (M = 2.89, SD = 0.78) and Gen Z (M = 2.84, SD = 0.70).
Across all generations, productive integrating was the most frequently used strategy (Total M = 3.97, SD = 0.72), while avoiding was the least used (Total M = 2.25, SD = 0.80). Third-party help showed a generational trend with Gen Z (M = 3.06, SD = 1.03) using it more frequently than Gen X (M = 2.80, SD = 0.98), though this difference was marginally significant (p = .053).
The roles of cultural values and socio-demographic characteristics
Hierarchical linear regression results for socio-demographics and culture orientations on conflict styles.
B, β, and t values are the statistics displayed in the given block in the regression model. *p < .05, **p < .01, ***p < .001.
aNote. Dummy-coded marital status with “Married” as the reference group.
bDummy-coded living arrangement with “Live with non-romantic others (e.g., roommates, family)” as the reference group.
cDummy-coded financial status with “Share financial management and responsibility equally” as the reference group.
Individualism and collectivism (H1-H5 and RQ2)
We predicted that Chinese women with higher individualistic (IND) orientations tend to adopt more dominating (H1) and emotional expression (H5) to manage conflict in romantic relationships. These two hypotheses were both supported, with IND significantly associated with dominating (β = 0.35, p < .001) and emotional expression (β = 0.18, p = .007).
Meanwhile, we proposed that Chinese women with higher collectivistic (COL) orientations would use more obliging (H2), productive integration (H3), third-party help (H4a), and passive-aggression (H4b) to manage conflict in romantic relationships. H2 (β = 0.28, p < .001), H3 (β = 0.43, p < .001), and H4a (β = 0.36, p < .001) were all supported. However, contrary to H4b′s prediction, COL did not significantly associate with the usage of passive aggression. Instead, participants with stronger IND tended to report a greater use of passive aggression (β = 0.20, p < .001). In addition, IND was also shown as significantly associated with obliging and third-party help styles (β = 0.23, p < .001).
RQ2 examined the use of avoiding among Chinese women with varying degrees of IND and COL. Results showed that IND was significantly positively related to avoiding (β = 0.32, p < .001), whereas COL was significantly negatively related to avoiding (β = −0.20, p = .001).
Socio-demographic characteristics (RQ3 and RQ4)
RQ3 explored how socio-demographics are associated with conflict management styles among Chinese women. The findings revealed that among the variables examined, the strongest predictor was argument frequency, which exhibited a significant positive association with passive aggression (β = 0.38, p < .001), dominating (β = 0.23, p < .001), avoiding (β = 0.13, p = .042), and emotional expression (β = 0.23, p < .001). In contrast, a negative association was found for productive integrating (β = −0.12, p = .049). Relationship length was negatively associated with third-party help (β = −0.28, p = .006), and a marginally negative association was found with passive aggression (β = −0.17, p = .075).
Family relationships also play a role. Significant positive associations were found between the number of children and third-party help (β = 0.19, p = .002) as well as passive aggression (β = 0.15, p = .016). The number of siblings showed positive associations with dominating (β = 0.15, p = .029), obliging (β = 0.16, p = .018), and third-party help (β = 0.15, p = .018). A marginal association was also observed for avoiding (β = 0.12, p = .077).
Education positively predicted productive integrating (β = 0.16, p = .016) and third-party help (β = 0.15, p = .019). A marginal negative association was found with passive aggression (β = −0.12, p = .056).
Being single or cohabiting (i.e., never married) was negatively associated with avoiding (β = −0.14, p = .036), indicating that never-married Chinese women reported substantially lower use of avoiding compared to those who were married. Besides, a marginally positive association was observed with productive integrating (β = 0.12, p = .084).
Financial independence was positively associated with avoiding (β = 0.16, p = .018) and negatively associated with productive integrating (β = −0.17, p = .018). Finally, for those with no income, marginal positive associations were observed with avoiding (β = 0.11, p = .078) and obliging (β = 0.11, p = .073). A summary of the findings for all hypotheses and RQs is presented in Appendix B.
Lastly, given China’s rapid social transformation and the possibility that cultural values and generational position may be intertwined, we conducted exploratory post hoc analyses to examine whether the associations between cultural values and conflict styles varied by generational cohort. Interaction terms between generation (Gen Y, Gen Z; Gen X as reference) and individualism or collectivism were added to the regression models. The overall pattern of results remained consistent with the main effects reported above. Interactions accounted for minimal increases in explained variance (ΔR2 = .00–.03), and most interaction effects were not statistically significant. Only the Generation × Individualism interaction for avoiding reached statistical significance. Specifically, the positive relationship between individualism and avoiding was significantly stronger for Gen Y (B = .31, p = .012) and Gen Z (B = .45, p < .05) compared to Gen X. This pattern suggests that individualistic values are more strongly associated with avoiding style among younger cohorts. Full results are presented in Appendix C.
Discussion
This study examined generational differences, cultural values, and socio-demographic factors in Chinese women’s use of conflict styles in their romantic relationships. Guided by the conflict styles frameworks from Rahim (1983, 1985) and Ting-Toomey et al. (2000), our findings reveal a complex interplay of the factors in shaping Chinese women’s conflict management experiences.
The evolution of conflict management across generations
Our findings demonstrate distinct generational patterns in conflict management, supporting Elder’s (1994) claim that generational experiences foster collective approaches to interpersonal dynamics. Three key trends emerged, reflecting research highlighting the evolving interplay of traditional and modern values in romantic relationships (Zeng, 2020): a decline in power-oriented approaches, an increase in problem-solving strategies, and a sustained emphasis on relational harmony.
Shifting away from power-oriented approaches
A notable finding was Gen X’s significantly higher use of dominating and avoiding styles compared to younger generations, particularly Gen Z. This pattern aligns with life course theory’s assertion that older generations often become more authority-oriented (Gao et al., 2024). Gen X women came of age during China’s socialist revolution and the Cultural Revolution, which emphasized gender equity in the public sphere (Roberts, 2010; Stockman, 1994). This likely fostered their use of dominating styles to assert authority in relationships. Their higher use of avoiding strategies can be attributed to the era’s limited tolerance for marital discord (Hu & Scott, 2016; Wang & Zhou, 2010).
In contrast, younger generations’ significantly lower use of dominating and avoiding styles reflects China’s cultural shift toward egalitarianism and self-autonomy (Chen, 2015; Gao et al., 2024). This extends Zhang et al.’s (2005) findings from professional to romantic settings, suggesting a broader societal move away from a power-oriented social hierarchy.
Emergence of problem-solving approaches
Furthermore, our findings of Gen Z’s greater reliance on third-party help and emotional expression reveal an important generational shift toward problem-solving strategies. This aligns with the general expectations that young women are equipped with more developed interpersonal skills and display more positive behaviors in conflicts with partners (Creasey, 2002). By contrast, although Gen X also reported higher emotional expression than Gen Y, their concurrent preference for non-problem-solving approaches (e.g., domination and avoidance) suggests that their emotional expression may function more as emotional release than a tool for resolution. This finding complicates earlier claims that older couples display fewer negative emotions in conflict (Carstensen et al., 1995), and underscores the need to distinguish between emotional expression as catharsis versus as a deliberate problem-solving strategy.
This distinction is especially critical for understanding the role of emotional expression in romantic relationships. Although it occurs in other interpersonal and familial conflicts, romantic partnerships are more emotionally charged, involving deeper emotional interdependence and intimacy (Cramer, 2000; Diener et al., 2018). As a result, emotional disclosure carries different stakes in romantic relationships than in contexts such as parent–child or sibling relationships governed by hierarchical norms in China. The fact that emotional expression can function to either repair or escalate conflict among romantic couples (Kleef & Côté, 2018) makes it particularly important to understand when and how it is employed. Gen Z women’s stronger tendency (compared to older generations) toward emotional expression reflects this generation’s emphasis on authenticity and openness in intimate partnerships (Melnikova & Tychinina, 2022), echoing Liu et al.’s (2024) argument that the value of emotional expression should be recognized and examined through a more nuanced lens. This constructive use of emotional expression may be explained by a broader relational goal toward mutual growth and understanding, as prior research shows that relational self-expansion and shared personal growth are associated with lower levels of conflict and more constructive engagement in romantic relationships (West et al., 2024). Collectively, these findings point to a generational shift in the function of emotional expression specific to romantic contexts, with younger women increasingly using it as a constructive tool for managing conflict and solving problems, rather than merely a means of venting and releasing—a pattern reflecting evolving conflict strategies particularly relevant to romantic relationships given the enduring hierarchical structures in Chinese family relationships (Song & Ji, 2020).
In addition, Gen Y’s higher use of obliging reflects a more conservative problem-solving approach, whereas both Gen Y and Gen Z’s greater use of third-party help suggests increasing openness to external assistance (Melnikova & Tychinina, 2022). This pattern is particularly important to understand in the context of romantic relationships. Although third-party involvement occurs in managing other interpersonal conflicts, it functions differently in romantic partnerships. For example, in familial conflicts, third-party involvement by parents or elders typically follows predetermined hierarchies and is normative and expected (Klein & Milardo, 2000). In romantic relationships, by contrast, third-party help is discretionary and self-initiated, with partners proactively selecting external parties from their social networks rather than defaulting to family authorities. This growing willingness to seek outside support also mirrors broader societal transformations in contemporary China, including expanded private spaces and population mobility since the 1980s (Huang, 2020), the “sexual revolution” and modernization, which have fostered autonomy and openness (Gao et al., 2024, p. 9), as well as the rise of social media, which normalizes sharing personal and relational struggles publicly (Cole et al., 2018). Together, these shifts may have encouraged younger generations to turn to external perspectives and feedback when managing romantic conflicts—a practice that represents a significant departure from traditional Chinese values emphasizing privacy in intimate relationships (Gao et al., 2024).
Persistent focus on relational harmony
Another critical finding is the consistent emphasis on relational harmony across all generations, evidenced by the highest use of productive integrating and the lowest use of avoiding across all groups. This persistence of harmony-oriented approaches, despite generational shifts in other conflict management strategies, supports previous research on the enduring influence of collectivistic cultural norms (Wang & Liu, 2010) and relationship-oriented values (Chi et al., 2020) in Chinese society, highlighting the role of cultural orientations discussed below.
It is important to note that these generational patterns should be interpreted in light of the self-enrolled online nature of the sample. Self-enrolled surveys tend to attract individuals who are more reflective about their relationships and more willing to articulate conflict experiences online, which may shape how conflict is interpreted and reported (Khazaal et al., 2014). As such, the observed generational differences may have reflected conflict orientations among women comfortable with online disclosure of intimate relationship dynamics, rather than population-level estimates of conflict behavior.
Beyond the binary: Rethinking individualism-collectivism in conflict styles
The study provides robust evidence for the influential role of cultural orientations, particularly individualism-collectivism, in predicting conflict management styles, aligning with an extensive body of research (Gomez & Taylor, 2018; Gudykunst et al., 1996; Ting-Toomey et al., 1991). Meanwhile, our findings reveal complex relationships between cultural orientations and conflict styles, challenging traditional binary conceptualizations of individualism-collectivism dynamics. Exploratory post hoc analyses further suggest that the associations between cultural orientations and conflict styles remain largely consistent across generational cohorts.
A particularly unexpected finding was that individualism, rather than collectivism, predicted passive aggression, challenging the conventional assumption that indirect conflict styles are uniquely tied to collectivistic orientations (Ting-Toomey et al., 1991). This pattern may reflect the tension between individualistic values and limited freedom of emotional expression within a predominantly collectivistic society. For Chinese women with stronger individualistic tendencies, the issue may not be whether they want to express emotion (i.e., “overt expression”), but whether they feel permitted to do so (i.e., “freedom of expression”; see Mongrain & Vettese, 2003, p. 545). In this context, passive aggression may function as a strategic outlet, enabling self-assertion while avoiding direct confrontation that could violate social norms. As such, the availability of expressive space, not merely conflict style preference, emerges as a critical factor in how individualism is enacted under cultural constraint.
This interpretation also aligns with broader cultural dynamics in China, where modern individualistic values coexist with enduring traditional expectations. Ji’s (2015) concept of modern-tradition mosaic offers a useful framework here: rather than replacing tradition, modern norms are layered onto older cultural logics, prompting individuals, especially women, to adopt hybrid strategies in navigating relational conflicts. These tensions may be particularly salient in romantic relationships, where emotional intimacy intersects with strong social expectations. In such contexts, passive aggression may represent one expression of this mosaic dynamic, offering a means to signal discontent or assert one’s needs indirectly when direct confrontation carries high relational costs and is often expected to be avoided. This may be especially consequential given that relationship satisfaction is more closely tied to overall life satisfaction than in other types of relationships (Cramer, 2000).
This finding, along with other results showing individualism’s positive associations with both direct (dominating, emotional expression) and indirect strategies (avoiding, passive aggression), suggests that individualistic tendencies manifest in more complex ways than previously theorized. The positive relationship between individualism and dominating style aligns with Ting-Toomey et al.’s (1991) findings regarding individualistic values and confrontational approaches, but the relationship with passive aggression challenges simple categorizations of direct styles as individualistic and indirect styles as collectivistic. Post hoc analyses also revealed that the positive association between individualism and avoiding was stronger among younger generational cohorts. Although exploratory, this pattern may reflect how younger individualistic Chinese women navigate relational autonomy differently than their Gen X counterparts (Gao et al., 2024)—potentially using avoidance to protect independence in relationships with lower commitment costs, or as a response to tension between individualistic values and persistent cultural expectations for relational harmony (Ji, 2017). This preliminary finding warrants further investigation to clarify the underlying mechanisms driving this generational variation.
On the other hand, collectivism negatively predicted avoiding while positively predicted obliging, productive integrating, and third-party help. This pattern suggests that collectivistic values in Chinese romantic relationships do not necessarily lead to indirect or avoidant approaches, as the literature suggested (Ting-Toomey, 1988). Instead, women with stronger collectivistic orientations appear to actively engage in conflict resolution while choosing approaches that preserve relational harmony.
Another important finding is that both individualism and collectivism positively predict third-party help and obliging strategies, suggesting different cultural orientations can lead to similar behavioral outcomes through distinct motivations. For women with stronger collectivistic values, seeking third-party help and engaging in obliging behaviors may reflect traditional approaches to maintaining harmony and face through respected mediation (Ting-Toomey, 1988). In contrast, those with stronger individualistic tendencies might view third-party help as tools for problem resolution and obliging as a strategic choice in relationship maintenance. This dual cultural basis for these conflict styles extends Ting-Toomey’s (1988) theoretical framework by suggesting that similar conflict styles might be motivated by different cultural values. Future research can explore these cultural dynamics further, particularly the underlying motivations and contexts that shape how individualistic and collectivistic values manifest in specific conflict management behaviors.
The influences of socio-demographics
Socio-demographic factors were significantly associated with conflict management styles, reflecting the impact of both situational dynamics and behavioral tendencies. Frequent arguments exhibit a dual influence, correlating with both confrontational (e.g., dominating, emotional expression) and indirect styles (e.g., avoiding, passive aggression). This finding aligns with research indicating that frequent disputes heighten emotional intensity, prompting more confrontational behaviors (Weingart et al., 2015), while also producing conflict fatigue that fosters avoidance (Liu & Roloff, 2015). Furthermore, the fact that argument frequency emerged as the most robust predictor across conflict styles resonates with Cao et al.’s (2015) observation that Chinese couples’ interactions display both stability and heterogeneity. On the one hand, frequent disputes may signal enduring conflict orientations, consistent with Cao et al.’s (2015) finding that many couples maintain stable communication profiles across contexts. On the other hand, our finding that argument frequency were associated with both confrontational and avoidant styles reflects the situational variability noted by Cao et al. (2015), where couples alternated between warmly supportive, emotionally quarrelling, and traditionally undemonstrative modes depending on interactional contexts.
Longer relationship duration correlated with reduced reliance on third-party help, indicating that couples gradually develop effective or private conflict strategies. This pattern is consistent with Chinese cultural values of face maintenance and privacy (Gao et al., 2024; Ting-Toomey, 1988), commonly expressed in the saying “jiā chǒu bù kě wài yáng” (家丑不可外扬), which discourages exposing domestic problems to outsiders. In addition, as relationships become longer, conflicts are more likely to be embedded in shared histories and ongoing relational patterns, making them more central to the relationship. Recent research suggests that as conflicts become more central to the relationship and less likely to be resolved quickly, individuals tend to shift away from externally focused, solution-seeking strategies and toward more internal or dyadic ways of managing conflict (Jöhnk et al., 2025). Together, these cultural norms and relational dynamics may help explain why couples in longer relationships are less likely to seek third-party help.
Family dynamics also played a role in shaping conflict management. For instance, women with more children reported greater use of third-party help and passive aggression, likely reflecting efforts to reduce direct confrontation in family settings. This finding aligns with prior research showing that interparental conflict undermines children’s well-being (Hess, 2022) and can foster insecure attachment and avoidant tendencies in later romantic relationships (Li et al., 2020). As such, mothers may deliberately rely on less overt strategies to manage disputes as a means of protecting children from direct exposure to conflict. Similarly, sibling relationships were associated with conflict strategies: dominating, obliging, and third-party help were more common among individuals with more siblings, possibly reflecting behaviors learned through sibling interactions (Shalash et al., 2013). Although these associations were moderate, the findings align with and extend previous research on conflict management in family contexts (e.g., Song & Zhang, 2012) and warrant further investigation across diverse samples.
Moreover, education was a strong predictor of constructive conflict strategies, including productive integration and third-party help, and was negatively associated with passive aggression. This finding aligns with prior research (e.g., Rainer & Smith, 2012) showing that education enhances relational and sexual satisfaction by fostering open and effective communication, suggesting that higher levels of education may equip individuals with more advanced interpersonal skills for managing romantic conflict. Beyond interpersonal communication skills, one possible explanation is that higher levels of education may also be associated with greater capacity for self-reflection and clearer self-concepts, which recent research has linked to higher relationship satisfaction in the face of conflict (Szachter et al., 2025).
Finally, single or cohabiting (i.e., never-married) Chinese women reported lower reliance on conflict avoidance and a greater tendency toward productive integration than married women. This pattern may reflect their higher levels of education and career aspirations (commonly associated with delayed marriage), which challenge traditional norms that prioritize relational harmony over personal needs (You et al., 2021; Zeng, 2020). These findings also suggest a growing emphasis on egalitarian partnerships and higher standards for relational satisfaction, such as mutual understanding, rather than compliance with marriage-related pressures (e.g., the “leftover women” stigma), echoing Gui’s (2020, 2023) observations from interviews with never-married Chinese professional women. Further research is needed to examine these evolving dynamics, which may reflect broader transformations in gender norms and relationship expectations.
Implications
This research advances conflict management theories by illustrating the interplay of generational, cultural, and socio-demographic factors in shaping conflict styles. It refines existing frameworks by integrating Rahim’s (1983, 1985) five-style model with Ting-Toomey et al.’s (2000) Conflict Style Dimension, supporting a more dynamic, situation-dependent model of conflict management. The results challenge assumptions about the consistent adoption of single conflict strategies based solely on cultural values.
The findings also have important practical implications for relationship counselling and conflict resolution interventions in contemporary Chinese society. Practitioners such as therapists should recognize that conflict management strategies emerge from a complex interaction of generational experiences, cultural values, and socio-demographic factors. This understanding suggests moving away from one-size-fits-all approaches based solely on cultural assumptions. Instead, interventions should consider how generational shifts, particularly the move from power-oriented to problem-solving approaches among younger generations, interact with persistent cultural values emphasizing relational harmony.
Notably, younger cohorts’ greater openness to seeking third-party help points to the potential of integrating online support groups or mediated counseling platforms, which can preserve privacy while offering accessible avenues for assistance. Additionally, practitioners should acknowledge how individual circumstances, including education level, relationship status, and family dynamics, may influence clients’ preferred conflict styles. This nuanced approach can enhance the effectiveness of relationship counselling and conflict resolution programs in addressing the diverse needs of Chinese women in romantic relationships.
Limitations and future directions
Despite its contributions, this study has limitations. One concerns measurement reliability. The IND and COL scales demonstrated relatively low internal consistency, a common challenge in assessing cultural orientations across contexts due to inherent conceptual and measurement variability (Oyserman et al., 2002). Some conflict style subscales also showed less-than-ideal reliability. We recognize that lower reliability may weaken the observed relationships and reduce the precision of our estimates; therefore, results involving these measures should be interpreted with caution. Part of this limitation may reflect our theoretical refinement—merging compromising and integrating into a new construct, productive integrating, to align with cultural face values. Although conceptually grounded and supported by our factor analyses, this new construct warrants further empirical validation. Future research should refine measurement scales to improve reliability, test measurement invariance across generations, and potentially employ data-driven strategies (e.g., latent class analysis, exploratory factor analysis) to further assess the distinctiveness of conflict styles across cultural contexts.
Second, the cross-sectional design of this study provides only a snapshot of conflict management and precludes causal inferences about how individuals may shift between or combine conflict strategies across time or situations. Future research using longitudinal or experimental designs is needed to clarify the temporal ordering and potential causal mechanisms underlying the observed associations. Beyond these design considerations, the study’s reliance on self-report data from one partner in the romantic relationship introduces additional methodological challenges. Although asking participants to recall a specific recent conflict enhanced ecological validity, self-report measures remain susceptible to social desirability bias, and the process of recalling and reporting an emotionally intense conflict episode may also have introduced respondent fatigue.
Moreover, since the study is not based on dyadic data, it does not model interactional processes between partners. While respondents’ conflict styles are situated within relational and structural contexts through measured variables such as partners’ age, employment status, and financial arrangements, these indicators function as structural proxies rather than direct measures of couple-level dynamics, such as assortative mating patterns (e.g., educational or age homogamy), power asymmetries arising from relative resource positioning between partners, or mutual adaptation in conflict behavior. Additionally, micro-contextual factors, such as conflict topic, initiator, or shifting interactional dynamics over the course of an exchange, could meaningfully shape conflict styles but were not examined in the present study.
Future studies could strengthen validity and deepen theoretical insight by employing longitudinal, experimental, qualitative, or mixed-method designs, such as daily diaries, dyadic analyses, or observational methods. Person-centered dyadic and observational approaches, such as those demonstrated by Cao et al. (2015) and Park and Harris (2026), would be particularly valuable for examining both the stability and variability of romantic conflict processes, elucidating how conflict strategies emerge and evolve within dyadic interactions.
Third, our analytical approach treated generational, cultural, and socio-demographic factors as independent predictors. While this allowed us to examine the unique contribution of each set of variables, it does not fully capture potential interactive or synergistic relationships among these factors. Exploratory post hoc analyses suggested that the association between individualism and avoiding style was stronger among younger cohorts of Chinese women; however, these analyses were not guided by a priori hypotheses and should be interpreted cautiously as preliminary findings.
To address these limitations, future research would benefit from developing and testing explicit theoretical predictions regarding how cultural values may operate differently across generational contexts and from employing more integrative analytical approaches, such as structural equation modeling and multilevel modeling, to better disentangle interactive effects and examine potential mediators (e.g., emotional regulation skills, stress, or anxiety). Longitudinal and qualitative studies could further illuminate the mechanisms underlying generational differences in conflict management (e.g., avoiding), while cross-cultural comparisons with other rapidly transitioning societies could clarify whether these patterns reflect universal processes or context-specific dynamics unique to contemporary China.
Furthermore, since participants were recruited via self-enrolled online surveys, the sample may be skewed toward individuals from urban areas, who are more digitally literate, educated, and willing to reflect on and disclose intimate relationship dynamics. This self-selection may interact with generational characteristics; for instance, younger participants may be particularly comfortable with online disclosure. While measurement invariance tests showed our key constructs functioned comparably across generations, future research should employ more diverse recruitment strategies (e.g., probability sampling, offline recruitment) to examine whether the observed patterns generalize to women less inclined toward online research participation.
We also did not collect information on participants’ gender identity or disability status. As such, the findings reflect the experiences of cisgender women only and do not account for how gender diversity or disability may shape relational experiences and conflict management. This focus on cisgender Chinese women may further limit the generalizability of findings to other cultural or gender groups. Future research should incorporate these demographic dimensions to provide a more inclusive and comprehensive understanding of conflict dynamics across diverse populations.
Finally, while we included partner and relationship-level variables to contextualize participants’ conflict styles, limitations remain in capturing the full complexity of couple and family dynamics. Regarding partner characteristics, we measured partner employment status but not detailed occupational information (e.g., occupational category, work hours, job demands), which limited our ability to assess how work-related stress, time availability, income levels, and occupational prestige may shape conflict interactions. Regarding natal family background, our measures were restricted to sibling information and did not include parental education, socioeconomic status, or parental conflict patterns—factors that prior research has shown to influence conflict management through intergenerational transmission and early socialization (Li et al., 2020; Shalash et al., 2013). Additionally, we did not collect information on separation duration or whether separated/divorced participants remained in contact with former partners, factors that may have shaped how conflict episodes were recalled and reported. Future research should incorporate more comprehensive measures of partner characteristics, natal family socioeconomic background and parental conflict patterns, as well as post-relationship dynamics, to provide a fuller understanding of how relational and family contexts shape women’s conflict management across the life course and relationship trajectory.
Conclusion
This study examines the multifaceted nature of conflict management styles among Chinese women, in relation to generational, cultural, and socio-demographic factors. The findings reveal a generational shift from indirect and power-oriented to problem-solving approaches, alongside a persistent emphasis on relational harmony across all cohorts. Cultural values remain persistent and prominent within the broader framework of individualism-collectivism, yet they display nuanced variations in the Chinese context, with individualism associated with both direct (dominating) and indirect (passive-aggressive) strategies. Our findings also show that socio-demographic factors, such as argument frequency, are associated with both confrontational and indirect conflict strategies, highlighting the situational dynamics of conflict management. These results deepen our understanding of how individualism and collectivism operate within a predominantly collectivistic cultural context and underscore the need for context-sensitive approaches to conflict resolution. Insights from this study can guide practitioners and counselors in developing tailored interventions that account for generational, cultural, and socio-demographic influences, thereby promoting more effective conflict management in romantic relationships.
Supplemental material
Supplemental Material - Tangled threads of romantic conflict: The intersection of generation, culture, and socio-demographics in shaping Chinese Women’s romantic conflict styles
Supplemental Material for Tangled threads of romantic conflict: The intersection of generation, culture, and socio-demographics in shaping Chinese Women’s romantic conflict styles by Jiaqi Zeng and Rain Wuyu Liu in Journal of Social and Personal Relationships.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this research was funded by the “Special Research Funding Award” from the Department of Communication, University of Arizona.
Ethical approval
All participants in this study provided written informed consent prior to their participation. The study was approved by the Human Subjects Protection Program (HSPP) of the University of Arizona (Approval Date: May 6, 2024).
Data Availability Statement
The data used in the research cannot be publicly shared but are available upon request. The data can be obtained by emailing the first author, Jiaqi Zeng:
Supplemental material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
