Work and family conflict analysis of female entrepreneurs in Turkey and classification with rough set theory

The study examines how female entrepreneurs in Turkey navigate responsibilities in work and family domains, the resulting levels of work–family and family–work conflict, and whether these conflicts influence investment decisions. Entrepreneurship is central to economic development and involves decision-making under uncertainty; rough set theory is introduced as a tool to handle vagueness in such real-life classifications. The research focuses on female entrepreneurs’ dual roles and formulates three research questions: RQ1: Are the responsibilities of female entrepreneurs in both business and family decisive for conflict that they experience? RQ2: Are the responsibilities of female entrepreneurs in both business and family decisive for their investment decision? RQ3: What is the relationship between conflict level that they experience both in business and family and their investment decision? The study aims to quantify conflict levels and investment intentions and to model classifications under uncertainty using rough set theory.

The literature traces the evolution of women’s participation in the workforce and entrepreneurship, noting historical constraints and progressing rights in contexts such as Turkey after the founding of the Republic. Foundational questions in women’s entrepreneurship show that women are less likely to engage in entrepreneurship, tend to acquire financial resources differently and at lower levels, may enact somewhat different strategies, and their firms often underperform on traditional economic indicators compared to male-led firms. While motivations, values, and expectations can be similar across genders, differences appear in risk-taking, self-confidence, networking, and sectoral preferences. Work–family and family–work conflict are defined as bidirectional role conflicts (Greenhaus and Beutell), arising via time, strain, and behavior mechanisms, and associated with stress, health, and life satisfaction outcomes. Women typically experience more conflict due to greater domestic responsibilities; work–family conflict tends to be more prevalent than family–work conflict. Prior Turkish studies also find higher work-to-family effects. Investment decision literature emphasizes growth intentions as multidimensional and influenced by psychological, sociological, economic, and managerial factors; entrepreneurial intensity relates positively to growth intentions. Women’s investment and growth choices may be shaped by family responsibilities, with some preferring controllable firm sizes to balance life roles, though business ownership can reduce conflicts due to autonomy and flexibility. The study highlights a gap in the Turkish context and introduces rough set theory as a novel classification method for women entrepreneurs.

Design and sample: A face-to-face survey was conducted with 348 female entrepreneurs in Istanbul, Kocaeli, Bursa, and Yalova—regions with high economic contribution and relatively low gender inequality. The questionnaire covered demographics, business characteristics, entrepreneurial intensity, work–family and family–work conflict, and investment plans. Measurement: Items used five-point Likert scales; for rough set analysis, composite scores were categorized into three levels (low, medium, high). Entrepreneurial intensity, conflict levels, and investment decisions were discretized accordingly. Rough set framework: Rough set theory (Pawlak) was employed to handle uncertainty via lower and upper approximations and boundary regions. Objects (female entrepreneurs) were grouped into information granules based on identical condition attributes. Condition attributes: entrepreneurial time (duration), presence of an active partner, entrepreneurial intensity, marital status, age of youngest child, and educational status. Decision attributes: Stage 1 decision sets were work–family conflict (low, medium, high) and family–work conflict (low, medium, high). Stage 2 decision sets were four investment decision dimensions each at low, medium, high: functional (equipment/means of production), team-based (training/recruitment), product and market (product development/new markets), and combination (other investments). Granules and approximations: 226 granules were formed; 167 were singletons, indicating heterogeneity. For each decision set, lower approximations (certain members), upper approximations (possible members), and boundary sets (uncertain region) were computed, along with accuracy coefficients α = |lower|/|upper|. Membership functions were calculated for each granule as μX(x) = |X ∩ C(x)| / |C(x)| to derive set cardinalities by summing memberships. Correlation analysis: Relationships between conflict levels and investment decision categories were examined using Goodman–Kruskal’s gamma (for ordinal association). Kendall’s tau-b and Goodman–Kruskal’s lambda (nominal predictability) were also computed to check robustness. Reliability: Cronbach’s alpha was used to assess internal consistency of the scales (values not reported in detail in the text).

- Conflict distributions: Work–family conflict levels among 348 entrepreneurs were high 112, medium 88, low 148. Family–work conflict levels were high 26, medium 108, low 214, indicating family–work conflict is generally lower than work–family conflict. - Rough set Stage 1 approximations (example values from Table 3): For work–family conflict low (X1), lower=98, upper=209, boundary=111, α≈0.47; medium (X2) lower=57, upper=135, boundary=78, α≈0.42; high (X3) lower=68, upper=162, boundary=94, α≈0.42. For family–work conflict low (Y1) lower=134, upper=261, boundary=95, α≈0.51; medium (Y2) lower=62, upper=165, boundary=103, α≈0.38; high (Y3) lower=12, upper=67, boundary=55, α≈0.18. - Membership-based intensity (Table 5): Work–family conflict membership distribution across 226 granules: low 43%, medium 27%, high 30%. Family–work conflict: low 65%, medium 28%, high 7%. - Investment decision distributions (Stage 2 counts): Functional (F) low 76, medium 97, high 175; Team-based (TB) low 110, medium 104, high 134; Product & Market (PM) low 112, medium 105, high 131; Combination (C) low 116, medium 112, high 120. High-level functional investment plans are most frequent. - Rough set Stage 2 approximations (Table 4 examples): Functional high (F3) lower=106, upper=237, boundary=131, α≈0.45; Team-based high (TB3) lower=82, upper=207, boundary=124, α≈0.40; Product–Market high (PM3) lower=76, upper=200, boundary=124, α≈0.38; Combination high (C3) lower=71, upper=196, boundary=126, α≈0.36. - Membership-based investment intensities (Table 6): Functional high 49% of membership mass; Team-based high 40%; Product–Market high 38%; Combination high 36%. - Association between conflict and investment: Goodman–Kruskal gamma coefficients are near zero across all pairs (e.g., Functional vs Work–family γ≈−0.016; vs Family–work γ≈−0.121; Team-based −0.009/−0.067; Product–Market 0.032/−0.041; Combination 0.046/−0.040), indicating no meaningful ordinal association. Kendall’s tau-b and lambda analyses similarly show no significant relationships.

The findings address the research questions as follows. RQ1: Female entrepreneurs’ roles in business and family are associated with generally low conflict levels; family–work conflict is notably lower than work–family conflict, suggesting that work responsibilities more often spill over into family than vice versa. RQ2: When entrepreneurs are classified by shared role-related characteristics (granules), their investment decisions exhibit substantial high-level planning, especially for functional investments; however, the role-related condition attributes alone do not deterministically map to distinct investment categories, as indicated by sizable boundary regions and low approximation accuracies. RQ3: There is no significant relationship between the level of work–family or family–work conflict and the intensity/type of investment decisions; gamma coefficients are near zero, and alternative ordinal/nominal association measures corroborate this. The absence of detectable association may reflect that many participating entrepreneurs experience low conflict and prioritize work, maintaining investment planning regardless of family pressures. The discussion also notes potential sample selection dynamics: women with extremely high conflict may exit entrepreneurship or not participate, which could attenuate observed associations. The results contribute a novel rough set–based classification showing heterogeneity among women entrepreneurs and offering a nuanced view of uncertainty in decision classifications.

The study classified 348 female entrepreneurs in Turkey using rough set theory to analyze work–family and family–work conflicts alongside investment decisions. Most participants reported low conflict, with family–work conflict lower than work–family conflict. Investment intentions were generally high, particularly for functional investments, countering generalized views that women uniformly avoid growth. Across multiple association measures, no significant relationship emerged between conflict levels and investment decisions. The research fills a gap on Turkish female entrepreneurs by providing quantitative evidence of conflict intensities and investment plans and demonstrates the utility of rough set analysis for classifying decision situations under uncertainty. Policy implications include supporting childcare and financial mechanisms to reduce conflict-related barriers and potentially expand women’s entrepreneurship.

- Geographic scope: Data come from four provinces with the lowest gender inequality (Istanbul, Bursa, Yalova, Kocaeli), limiting generalizability to other regions of Turkey. - Sample composition: The study focuses solely on female entrepreneurs, not comparing with male counterparts; those experiencing very high conflict may be underrepresented or absent, affecting observed relationships. - Measurement and design: Ordinal categorization into low/medium/high and reliance on cross-sectional survey data constrain causal inference; many granules were singletons, reflecting heterogeneity and contributing to broad boundary regions. - Future research: Broaden geographic coverage; compare regions; examine unsuccessful female entrepreneurs; explore mompreneurship; investigate links between conflict levels and investment intensity using alternative designs and indirect questioning; enhance support mechanism analyses (e.g., childcare, microcredit).