Socio-Demographic Predictors of Entrepreneurial Intentions: The Mediating Role of Perceived Gender Discrimination Among Female Students

Abstract

Understanding entrepreneurial intention among female students has become increasingly important for addressing gender disparities in business creation and fostering economic development. Pursuing to promote inclusive entrepreneurship and reduce gender gaps in business creation aligns with Sustainable Development Goals, particularly SDG 5 (gender equality). This study examines how demographic and social variables influence entrepreneurial intentions, with perceived gender discrimination as a potential mediating factor. Data were collected through an online survey employing a structured questionnaire and analyzed using logistic regression models incorporating mediation analysis. The sample consisted of 360 female students from a university in the South–East region of Romania. Among the six socio-demographic variables examined, marital status and income satisfaction emerge as significant predictors. The results indicated that married students expressed higher entrepreneurial intentions, while the participants with higher income satisfaction reported lower entrepreneurial intentions. Perceived gender discrimination was not a significant mediator in the tested model, and all calculated indirect effects were statistically non-significant. The findings of the study offer valuable insights for the design and implementation of local entrepreneurship policies as well as for university strategies and curricula adjustments to better support young women in their entrepreneurship endeavors.

1. Introduction

Entrepreneurship has proven to be a critical driver of economic growth, innovation, and social development worldwide. It particularly addresses different contemporary challenges such as unemployment, environmental sustainability, and social inequality. Understanding entrepreneurial intentions has become increasingly important, as they are the primary predictor of entrepreneurial behavior and business creation [1,2]. In this context, the theory of planned behavior (TPB), developed by Ajzen, provides a widely accepted theoretical framework for understanding the antecedents of entrepreneurial intentions. It emphasizes the roles of personal attitudes, subjective norms, and perceived behavioral control [3]. While TPB highlights these elements as drivers of entrepreneurial intention, it does not explicitly consider structural barriers. Perceived gender discrimination can influence these elements by reducing women’s self-efficacy through its effect on perceived behavioral control, altering subjective norms, by reinforcing stereotypes and negatively affecting attitudes towards entrepreneurship. Thus, discrimination may shape how socio-demographic factors translate into intention. By incorporating gender discrimination as a potential mediator, the explanatory power of TPB can be increased and a recognized research gap would be addressed.

It must be noted that the entrepreneurial domain is not equally accessible to all individuals. Therefore, gender disparities in entrepreneurship remain a significant concern. Women entrepreneurs face unique challenges and barriers that influence their entrepreneurial intentions and subsequent business creation activities. Perceived gender discrimination is one such challenge that has emerged as particularly important factor [4,5,6], significantly impacting women’s entrepreneurial aspirations and decision-making processes [7]. Researchers have acknowledged the importance of gender-based disparities; yet, limited research has systematically examined how perceived gender discrimination mediates the relationship between individual characteristics and entrepreneurial intention. The perception of discrimination can influence how sociodemographic factors translate into entrepreneurial intentions, potentially explaining some of the observed gender differences in entrepreneurial behavior [5,8]. Only a limited number of studies have focused on female students, who represent the next generation of entrepreneurs capable of contributing to sustainable development of different countries and regions. Insights into how entrepreneurial intentions are formed among women, particularly among young female students, fall within the scope of SDG 5 (gender equality).

Numerous studies have tackled demographic and socioeconomic factors that influence entrepreneurial intention. Among these are age, living environment, educational level, marital status, monthly income satisfaction, and employment experience. These variables have been shown to exert different levels of influence on entrepreneurial intention, with effects that vary across gender lines. It is important to understand how these factors interact with perceived gender discrimination in shaping entrepreneurial intention. The results are valuable for developing effective policies and interventions to promote inclusive entrepreneurship [9,10,11,12].

Female students represent a critical demographic category for entrepreneurial development, as their future career trajectories can be significantly shaped by early entrepreneurial intentions. This category could also contribute to closing the gender gap in entrepreneurship. Universities can provide a unique setting for examining entrepreneurial intentions, as this period represents a formative stage in which career aspirations are shaped, and entrepreneurial skills are developed [13,14,15]. Gender-based discrimination and stereotypes may persist also within academic settings, influencing female students’ perceptions of entrepreneurship as a viable career option [16,17].

An important gap in the current literature concerns the mediating role of perceived gender discrimination in the relationship between individual characteristics and entrepreneurial intentions. While previous studies have examined the direct effects of demographic factors on entrepreneurial intention and the impact of discrimination on these intentions, very few have explored the mediation role of gender discrimination perceptions in these relationships [5,18].

Addressing gaps, the present study aims to examine the influence of several demographic and socioeconomic factors—including age, living environment, education, marital status, income satisfaction, and work experience—on entrepreneurial intention, with perceived gender discrimination as a mediating factor. The analysis focuses on female students, who represent both an underexplored population in entrepreneurship research and a demographic group with major potential for future entrepreneurial development. The study provides insights into how gender-based discrimination perceptions can influence the formation of entrepreneurial intention.

Romania provides a particularly relevant context for examining students’ entrepreneurial intentions. As an emerging economy within the European Union, Romania continues to face structural gender inequalities in the labor market and entrepreneurial activities, with women’s participation rates below the EU average [19]. This situation can be attributed to cultural stereotypes and limited access to financial resources, which act as major barriers. Meanwhile, universities serve as important incubators of entrepreneurial skills, offering the opportunity to study the interaction between gender related issues and socio-demographic factors in shaping students’ entrepreneurial intentions. The focus on Romania can contribute to a better understanding of how women—especially the young highly educated onesdecide to start a business, thereby supporting the sustainable development of the national economy and improving overall welfare. It may also provide policymakers with relevant insights into how SDG 5 (gender equality) can be approached within higher education system and the national entrepreneurship ecosystem.

Perceived gender discrimination is expected to mediate the relationship between socio-demographic factors and entrepreneurial intention because it operates as a mechanism through which social contexts are translated into career choices. Discrimination can weaken behavioral control, reinforce restrictive subjective norms, and reduce positive attitudes towards entrepreneurship, as suggested by the TPB. For example, married women may encounter stereotypes related to family responsibilities, students living in rural environments may face more traditional norms, and those with higher education levels may perceive less discrimination due to access to more inclusive environments. Thus, perceived gender discrimination can act as a channel linking socio-demographic variables to entrepreneurial intention.

Therefore, this paper addresses the following primary research question: Do socio-demographic factors (age group, residence, marital status, education, work experience, income satisfaction) predict entrepreneurial intention among female students and is this association is mediated by perceived gender discrimination?

To answer this question, the primary research objective of the study is to quantify the associations between female students’ socio-demographic factors (age group, residence, marital status, education, work experience, income satisfaction) and entrepreneurial intention, and to determine whether this relationship is mediated by perceived gender discrimination. The secondary research objectives derived from the primary objective are: (1) to identify the most significant associations between variables included, (2) to test the proposed mediation model and (3) to provide empirical evidence supporting the relationship between the socio-demographic factors considered and entrepreneurial intention among female students.

The paper is structured as follows. Section 2 reviews relevant literature on entrepreneurial intention, individual and social factors, perceived gender discrimination, and the application of mediation analysis in predicting entrepreneurial intention. Section 3 presents the research methodology, including data collection procedures, sample characteristics, variable definition, research hypothesis formulation, and the proposed mediation model design. Section 4 presents the results and discussions, highlighting the implications of the findings. Section 5 concludes with the findings, outlines the research limitations, and suggests directions for future research.

2. Literature Review

2.1. Theoretical Frameworks (TPB and Entrepreneurial Event Theory)

The Theory of Planned Behavior (TPB) furnishes the theoretical foundation for studying entrepreneurial intentions. It provides a comprehensive framework for understanding how individual attitudes, subjective norms and perceived behavioral control influence behavioral intentions [3]. The validity of TPB in explaining entrepreneurial intention across diverse contexts and populations has been constantly demonstrated by research. Several scholars have also emphasized that incorporating additional variables that reflect the complexity of entrepreneurial decision-making, can enhance the theory’s explanatory power [20,21]. In this context, extended models have been developed that include factors such as entrepreneurial self-efficacy, personal values, environmental factors, and demographic characteristics.

Shapero and Sokol proposed the entrepreneurial event theory. It provides an alternative theoretical perspective by emphasizing the role of precipitating events and environmental factors in the formation of entrepreneurial intentions [22]. According to this theory, entrepreneurial intentions are triggered by displacement events and shaped by the perceived desirability and feasibility of entrepreneurial action [23]. These two theories provide a better understanding of the complex nature of entrepreneurial intention formation [24].

More recently, sustainability perspectives have been incorporated into entrepreneurial intention research, further expanding the theoretical foundations [25,26]. Sustainable entrepreneurial intentions refer to the intention to establish ventures that address environmental and social challenges while generating economic value. This has emerged as a critical area of investigation, reflecting the growing recognition that entrepreneurship must contribute to sustainable development and address global challenges [27].

2.2. Gender Differences in Entrepreneurial Intention

Gender has been documented as one of the most significant factors influencing entrepreneurial intention [28]. Research examining the differences between male and female entrepreneurs consistently shows that men exhibit higher levels of entrepreneurial intention than women, across various cultural, economic and educational settings [29,30]. Data from the Global Entrepreneurship Monitor indicates that globally, women’s total entrepreneurial activity remains approximately three-quarters that of men’s [31].

Several perspectives have been proposed to explain these gender differences. Welsh et al. found that perceived management skills and gender discrimination significantly influence women entrepreneurs’ launch decisions, highlighting the complex connection between competence perceptions and experiences of discrimination [5]. Santos et al. revealed that personal attitudes toward entrepreneurship exert a less direct positive impact on entrepreneurial intentions among women compared to men, while social norms affect entrepreneurial intentions differently across genders [29].

Mahfud et al. identified significant difference between male and female students regarding the influence of risk-taking and innovativeness on entrepreneurial intention, with women demonstrating lower risk tolerance which may potentially affect their entrepreneurial intentions [32]. Méndez-Suárez et al. [6] found that institutional and social perceptions regarding effort contribute to gender differences in entrepreneurial activities. Societal beliefs about gender appropriate effort and achievement can create powerful obstacles for female entrepreneurship [6].

Recent research provides complementary insights into how gender differences shape university students’ entrepreneurial intentions. Cardella et al. [33] found that female students were generally less likely than males to express high entrepreneurial intentions and that the type of higher education program moderated these gender differences. These findings indicated that educational settings play an important role in shaping entrepreneurial intentions [33]. A study by Vu et al. addressed female students’ entrepreneurial intentions and emphasized the critical role of perceived gender inequality, which acts as a barrier and reduces entrepreneurial motivation [34]. Chin et al. studied subjective norms and entrepreneurial intentions among Malaysian students, finding that these norms have a stronger influence on female students, making their entrepreneurial intentions more sensitive to social approval and cultural expectations than those of their male peers [35]. These results reveal the complex relationship between gender and entrepreneurial intentions among female students, highlighting the need for further research on this topic.

2.3. Perceived Gender Discrimination

For underrepresented groups such as women, ethnic minorities, and other marginalized populations, perceived discrimination represents a significant barrier to entrepreneurial participation [18]. Discrimination in entrepreneurship manifests through limited access to financial resources, networks and market opportunities, as well as stereotypes and bias in business contexts [36].

Research on gender discrimination in entrepreneurship has shown that women entrepreneurs face numerous disadvantages. These include difficulties in accessing capital and business loans, challenges in building professional networks, and stereotypes related to competence and commitment to business ventures [37]. Such forms of discrimination can occur throughout different phases of the entrepreneurial process, from business planning to scaling and growth stages [38,39].

The influence of perceived discrimination on entrepreneurial intentions is complex. While perceived discrimination may deter individuals from pursuing entrepreneurial intentions, other studies demonstrate that discrimination can also act as a motivating factor, with people choosing self-employment as an alternative to discriminatory employment contexts [40,41]. This effect has been observed particularly among entrepreneurs from minority groups. They view business ownership as a path to economic autonomy and gain freedom from workplace discrimination [42].

The influence of perceived discrimination has been explored in relation to broader aspects of entrepreneurial behavior and outcomes. Thus, discrimination perceptions can influence decisions related to business strategy. People from the discriminated groups tend to orient their ventures toward niche markets or develop businesses that address the needs of their own communities [43]. Moreover, discrimination can restrict entrepreneurs’ access to different resources and support systems, ultimately limiting the growth potential of their businesses [38,44].

2.4. Socio-Demographic Predictors of Entrepreneurial Intention

Socio-demographic variables are frequently analyzed as predictors of entrepreneurial intentions. Age can be a significant predictor of entrepreneurial intention, with studies revealing a complex relationship between these two variables. Entrepreneurial intention tends to peak during young adulthood and gradually declines with age. Younger individuals often exhibit higher levels of entrepreneurial intention due to lower opportunity costs and greater risk tolerance [45]. Meanwhile, older individuals may possess greater human capital and financial resources [9]. Factors such as risk tolerance, innovation orientation, and fewer family responsibilities explain why younger people are more likely to demonstrate higher entrepreneurial intentions [10].

The living environment can exert both direct and indirect influences on entrepreneurial intention [46]. According to Fritsch & Storey, urban areas typically provide greater access to resources, better educational facilities, more networking opportunities, and a more developed entrepreneurial ecosystem [11], all of which can foster entrepreneurial mindsets. In this context, students are more likely to be exposed to entrepreneurial role models and startup culture. By contrast, rural environments are often characterized by limited access to capital, support infrastructure, and entrepreneurial education, which may reduce entrepreneurial aspirations, according to Pato & Teixeira [12]. Other studies have demonstrated that necessity-driven entrepreneurship is prevalent in rural areas, where—according to Laukkanen—limited job opportunities can create different motivational pathways that increase entrepreneurial intentions [47].

A complex relationship has been demonstrated between education level and entrepreneurial intention. Higher education generally provides valuable skills and knowledge relevant to entrepreneurship. Research has reported mixed results regarding the direct relationship between educational level and entrepreneurial intention [48,49,50]. For example, some studies suggest that higher education may decrease entrepreneurial intention, as individuals are offered alternative career opportunities, which emphasizes employment options, rather than self-employment [48].

For women, entrepreneurial decision-making is strongly influenced by marital status and family circumstances. Existing research shows that entrepreneurial activity can be facilitated or restrained by family responsibilities and support systems [39,51]. The presence of children may enhance risk aversion among women due to increased family responsibilities, creating an additional barrier for female entrepreneurs. At the same time, married individuals may have access to supplementary financial resources and emotional support which may stimulate entrepreneurial activities [52].

As far as employment experience is concerned, it provides valuable human capital for entrepreneurship. Studies have demonstrated that work experience generally enhances entrepreneurial intention. Exposure to diverse work environments enables the accumulation of industry-specific knowledge, management skills, and professional networks, which in turn can enhance entrepreneurial self-efficacy and opportunity recognition capabilities [53,54]. This increases people’s confidence in their ability to manage a business and face market challenges. However, the effect of employment experience may vary depending on the type and quality of the work performed. A positive experience can encourage the individuals’ commitment to their employee status, if they are satisfied with their current jobs, while negative employment experiences may motivate them to explore the entrepreneurial alternatives [55].

Income satisfaction and financial security are important factors in the entrepreneurial decision-making process. Since they are satisfied with their situation, individuals with high income, may not exhibit strong entrepreneurial intentions. At the same time, such individuals often have more resources available to invest in entrepreneurial ventures. People with lower income or dissatisfaction with their financial situation may be more motivated to engage in entrepreneurship, as a potential solution for higher earnings. The relationship between income satisfaction and entrepreneurial intention is complex, involving variables such as risk preferences and financial obligations, which can influence this relationship in both positive and negative ways [52,56].

2.5. Research Gaps

Despite extensive research on the topic, several important gaps persist in the process of understanding how socio-demographic factors and perceived gender discrimination interact to influence entrepreneurial intentions. The mediating role of perceived gender discrimination in the relationship between different socio-demographic variables and entrepreneurial intention remains an understudied subject. Further investigation is necessary into how gender interacts with other socio-demographic variables in shaping entrepreneurial intentions. Most studies have approached gender issues without considering how gender intersects with factors such as age, education, and socioeconomic status to influence entrepreneurial intentions. This combined approach is important for capturing the complexity of entrepreneurial intention formation among different groups.

Another notable gap is the limited number of longitudinal studies on the development of entrepreneurial intentions. Therefore, we must acknowledge the limited comprehension of how entrepreneurial intentions evolve over time and which factors most strongly influence this process. Such an approach would be particularly valuable for understanding female students’ entrepreneurial intentions, as these may change significantly during and after their educational experiences.

Future research directions should focus on sustainable entrepreneurship, with the aim of exploring how sustainability issues influence the formation of entrepreneurial intentions. In a context marked by growing environmental and social challenges, understanding how sustainability values interact with traditional entrepreneurial motivations represents an important research topic. This area of inquiry could be particularly relevant, for younger generations, who may be more sensitive to sustainability considerations when making career choices.

This study addresses these gaps by examining the influence of socio-demographic variables on entrepreneurial intentions among female students, while testing perceived gender discrimination as a mediating factor. In doing so, it advances contributing both theoretical understanding and practical implications for promoting inclusive entrepreneurship.

3. Research Methodology

3.1. Research Design and Approach

This study was based on quantitative cross-sectional research aimed to examine the influence of different demographic and socioeconomic variables on entrepreneurial intention among female students, with perceived gender discrimination included as a mediating factor. This design was selected because it enables the analysis of the relationships between variables at a specific point in time, within the target population [57,58].

The research framework was grounded in the Theory of Planned Behavior (TPB) developed by Ajzen [3]. It has been extensively validated in entrepreneurship research aimed at understanding the formation of entrepreneurial intentions [20]. Moreover, the theoretical framework also incorporates discrimination theory, which explains how perceived discrimination shapes entrepreneurial intentions [59].

3.2. Population and Sampling

The target population consisted of female university students enrolled in bachelor’s and master’s degree programs at Ovidius University of Constanta, the largest university in Constanta County, located in South-Est region of Romania. A convenience sampling method was employed to recruit participants, without randomization or stratification. According to Etikan et al. this is a common practice in entrepreneurship research, due to accessibility and practical considerations [60]. However, the lack of randomization and stratification introduces potential selection bias. Although the survey was open to all the female students at the university, the concentration of respondents in economic study programs should be acknowledged as a contextual feature of the sample, limiting representativeness and affecting the generalizability of the findings.

Data were collected using a structured online questionnaire, administered between February and March 2023. Participation in this study was entirely volunteer. The questionnaire link was distributed electronically through students’ courses groups, ensuring broad reach and participant anonymity, as recommended by Evans & Mathur, and Dillman et al. [61,62].

The number of valid responses was 360, representing the final sample. This sample exceeds the minimum requirements for structural equation modeling and mediation analysis, as recommended by Hair et al. [63] and Kline [64]. It provides adequate statistical power to detect medium effect sizes and ensures reliable parameter estimates in the proposed mediation model [65,66].

For the analysis presented in this article, eight variables were selected, described as follows.

The first variable delt with participants’ age, measured in five categories: (1) 18–25 years, (2) 26–30 years, (3) 31–35 years, (4) 36–40 years, and (5) over 40 years. It was coded as age.grp.

The second variable referred to the participants’ residence, with two categories: (1) rural and (2) urban. The variable was coded as residence.

The third variable tackled the respondents’ marital status and included two categories: not married and married. The variable was coded as marital.

The fourth variable addressed the participants’ education level, coded as education, with five categories: (1) first-year bachelor, (2) second-year bachelor, (3) third-year bachelor, (4) first-year master, and (5) second-year master [47,48].

The fifth variable investigated the participants’ previous work experience, coded as wrk.exp, and was measured via two categories (i.e., yes and no), distinguishing between participants with and without prior work experience. This variable was tackled in earlier studies [52,53].

The sixth variable measured satisfaction with monthly income, coded as inc.satisf. It was assessed using a 5-point Likert scale (1 = to a very small extent, 5 = to a very large extent). Income satisfaction was measured with a single global item, which is commonly applied in economic and entrepreneurship research, as it regards a subjective evaluation that respondents can easily assess. Previous studies in labor economics, psychology, and entrepreneurship, such as Reynolds et al. and GEM surveys, used single-item income satisfaction scales due to their conciseness and predictive validity [31,67].

The seventh variable addressed perceived gender discrimination, coded as perc.gend.disc. It was measured with a five 5-point Likert scale (5 = to a very large extent, 1 = to a very small extent), capturing participants’ perceptions of gender-based discrimination in entrepreneurial and business contexts.

The final variable investigated entrepreneurial intention, coded as entrep.int. It was measured using an adapted version of the validated formula proposed by Liñán and Chen [68]. The variable consisted of two categories: “I haven’t decided” and “I want to do it”. Dichotomous or single-item measures have been used in prior large-scale studies, such as GEM [31]. This question aimed to distinguish between participants who were undecided and those who expressed their entrepreneurial intention, with a single item considered reliable for capturing the general decision tendency.

Prior studies have demonstrated that variables such as age and education are associated with the perception of discrimination [5,17]. Based on these findings, the first hypothesis was formulated:

H1. 

Each predictor is significantly associated with the mediator (perceived gender discrimination).

Building on prior studies showing that discrimination reduces entrepreneurial intentions [4,18,40], the second hypothesis was proposed:

H2. 

Perceived gender discrimination is significantly associated with entrepreneurial intention.

Previous research has supported the use of mediation models in entrepreneurship studies [7,69], leading to the third hypothesis:

H3. 

Each predictor has a significant indirect effect on entrepreneurial intention through perceived gender discrimination.

Studies by Kautonen et al., and Nikolaev et al. have demonstrated a significant direct effect of socio-demographic variables on entrepreneurial intentions [9,56], which led to the fourth hypothesis:

H4. 

Each considered predictor has a total significant effect on entrepreneurial intention.

3.3. Mediation Model Design

To examine the relationship among variables, mediation analysis was performed using logistic regression models, following the parallel design approach recommended in mediation studies (Hayes 2022) [70]. The independent variables included age.grp, environment, marital status (marital), education, wrk.exp and inc.satisf. The dependent variable was entrep.int, with perc.gen.disc as the mediating variable. All variables were treated as blocks, without accounting for their individual response categories.

The variables age.grp, residence, marital status, education, wrk.exp and entrep.int were treated as nominal, while inc.satisf and perc.gen.disc were ordinal variables measured on a 5-point Likert scale. In the proposed model, age.grp, residence, marital, education, wrk.exp were included as factor, whereas inc.satisf and perc.gen.disc were treated as numeric. The variable entrep.int was coded “I haven’t decided” = 0, “I want to do it” = 1. Logistic regression and mediation model required a binary outcome variable to assess intention clearly.

A schematic of the proposed mediation model is presented in Figure 1, illustrating the relationships among the considered variables. It also includes the hypotheses that were tested along with the model.

The figure illustrated the mediation model that includes six socio-demographics variablesage group (age.grp), residence, marital status (marital), education level (education), work experience(wrk.exp) and income satisfaction (inc.satisf)—specified as predictors for entrepreneurial intention (entrep.int). The model aims to test whether part of the considered effects of these predictors may occur indirectly through the mediator, i.e., perceived gender discrimination (perc.gen.disc).

The main model involves a multivariable logistic regression for the dependent variable (Y), incorporating both the mediator and all predictors.

$$logit\left\{P\left(Y=1\right)\right\}={\beta }_0+{\beta }_{M}M+\sum _k{\beta }_k{X}_k$$

(1)

According to mediation analysis theory, three types of effect were calculated and reported: direct, indirect, and total [71]. The direct effect of the independent variables on the dependent variable is represented by the β_k coefficient in the previous formula. In the proposed model, each socio-demographic variable has a direct path to entrepreneurial intention (the dependent variable), capturing the influence of the predictors. For example, marital status may affect entrepreneurial intention through increased household resources pooling, social capital or family driven necessity [51]. Conversely, income satisfaction may negatively impact entrepreneurial intention, as individuals content with their current earnings may be less incentivized to engage in a new business venture.

The model includes indirect effects, which reflect the influence of socio-demographic factors on entrepreneurial intention, through their impact on perceived gender discrimination (the mediator). Previous research has shown that age may increase awareness of discrimination due to accumulated workplace exposure, while education levels and urban residence may reduce perceived gender discrimination, as these factors are associated with greater access to more inclusive environments and diversity-oriented networks.

The indirect effect is calculated as the product of the effect of the independent variables on the mediator (a_k) and the effect of the mediator on the dependent variable (b—the coefficient of M in the above logit model).

$$M={\alpha }_0+{\beta }_{M}M+{\sum }_k{a}_k{X}_k+\epsilon$$

(2)

The total effect of the independent variables on dependent variable, accounting for the mediator, was calculated as the sum of the direct and indirect effects.

$$total effect={\beta }_k+a_{k}b$$

(3)

To check if the mediation effect is statistically significant, the bootstrapping method proposed by Hayes (2022) was employed [70]. For all three types of effects coefficients, p-values, statistical significance and 95% CIs were reported. The mediation model was computed using R software, version 4.4.3.

The model treats all predictors in parallel, estimating the individual direct and indirect effect for each variable while controlling for the others. This approach is very important given the frequent intercorrelations and overlap among socio-demographic variables (i.e., age, marital status, work experience). Parallel designs are widely used in entrepreneurship research, as they allow for a more accurate decomposition of effects [72].

Multivariable regression was employed because the dependent variable, i.e., entrepreneurial intention, was measured dichotomously (0 = “I haven’t decided”, 1 = “I want to do it”). Logistic regression is the most appropriate method for modeling binary outcomes, allowing for the simultaneous inclusion of multiple socio-demographic predictors while controlling intercorrelations. This approach also integrates seamlessly with mediation models using bootstrapping as recommended by Hayes (2022) [70]. Alternative methods such as linear regression would have been inappropriate given the categorical nature of the outcome, whereas a probit model would have produced similar estimates but with less intuitive interpretation of effect sizes. Therefore, logistic regression was the most suitable solution for this study.

3.4. Descriptive Statistics

Given that the variables included in our analysis are nominal and categorical, we present the results of descriptive statistics (

Table 1. Descriptive statistics for the nominal variables.

Variable	Count	Categories	Frequency	Percentage
age.grp	360	18–25 years	294	81.67
		26–30 years	16	4.44
		31–35 years	10	2.78
		36–40 years	24	6.67
		Over 40 years	16	4.44
residence	360	rural	111	30.83
		urban	249	69.17
marital	360	not married	304	84.44
		married	56	15.56
education	360	1st year bachelor	62	17.22
		2nd year bachelor	95	26.39
		3rd year bachelor	142	39.44
		1st year master	28	7.78
		2nd year master	33	9.17
wrk.exp	360	no	85	23.61
		yes	275	76.69
entrep.int	360	I haven’t decided	113	31.39
		I want to do it	247	68.69

Source: Authors’ calculation.

).

The table reflects the demographic profile of the respondents. It reveals a predominantly young population, with 81.67% aged between 18 and 25 years, matching the typical age structure of university students, and aligning with prior studies on entrepreneurial intention among youth [68,73].

The majority of participants resides in urban areas (69.17%), where greater access to resources, networks, and role models has been shown to positively influencing entrepreneurial ambition [74]. Most respondents are unmarried (84.44%), suggesting fewer family-related constraints that might limit risk-taking, consistent with findings that marital status can shape entrepreneurial decision-making [39]. Regarding education, the largest segment consists of final-year bachelor’s students (39.44%), a group often at a career crossroads and therefore more inclined to consider entrepreneurship as a viable career path. Most participants reported prior work experience (76.69%), which can enhance self-efficacy and opportunity recognition [55].

Entrepreneurial intention levels are high, with 68.69% of the respondents expressing a desire to start a business. However, the existing undecided group creates the premise for targeted entrepreneurship education programs [75]. Overall, the data presented in the table depicts a young, urban and educated population with substantial entrepreneurial potential, reflecting global trends in student entrepreneurship research.

The table presents the descriptive statistics for income satisfaction and perceived gender discrimination (

Table 2. Descriptive statistics for ordinal variables.

Variable	Count	Min	Max	Median	IQR
inc.satisf	360	1	5	3	2
perc.gen.disc	360	1	5	3	2

Source: Authors’ calculation.

). For income satisfaction (Mdn = 3, IQR = 2), respondents reported a moderate level of contentment with personal income, with neither extreme dissatisfaction nor high satisfaction predominating. This situation may reflect a transitional financial stage typical of students and early-career individuals, for whom financial security is not yet fully established [76].

Perceived gender discrimination displays an identical central tendency (Mdn = 3, IQR = 2), suggesting that respondents experience a moderate prevalence of discriminatory events. Prior research has highlighted that even moderate levels of perceived discrimination can influence career attitudes, self-efficacy, and entrepreneurial decisions, particularly among women [5,69]. This may underscore the importance of considering perceived gender discrimination as a potential mediator in different models designed to explain entrepreneurial intentions [18].

4. Results and Discussions

The first step in our mediation analysis involved testing the associations among the considered variables, the predictors, the mediator and the outcome. Therefore, a correlation matrix was computed. Given the nominal and the ordinal nature of the variables considered polychoric correlations were tested using the hetcor function from the polycor R package, R software, version 4.4.3. The results are presented in Figure 2 below, which displays the coefficients and their significance levels.

Entrepreneurial intention (entrep.int) shows significant negative correlations with the age group (age.grp) (r = −0.18, p < 0.1); thus, younger respondents are more inclined toward entrepreneurship, which is consistent with recent findings that younger individuals demonstrate higher entrepreneurial risk tolerance and opportunity orientation [73]. A negative correlation was also observed with education (r = −0.17, p < 0.05), suggesting that higher educational attainment is slightly associated with lower entrepreneurial intention, possibly due to increased access to secure employment opportunities or reduced perceived necessity to start a business [77]. Similarly, marital status (marital) was negatively correlated (r = −0.20, p < 0.05). The same negative correlation occurs in the relationship with income satisfaction (inc.satisf) (r = −0.24, p < 0.01), indicating that the individuals more satisfied with their income are less likely to aspire to entrepreneurship, which aligns with the push–pull models where dissatisfaction serves as a motivator [78]. No significant positive correlations emerged for entrepreneurial intention, suggesting that, in this sample, certain socio-economic and demographic factors may function more as constraints than as enablers.

Perceived gender discrimination correlates negatively with age group (r = −0.25, p < 0.01), suggesting that older participants report lower perceptions of gender-based inequity. This pattern aligns with findings that increased workplace exposure over time can enhance awareness of gender discrimination and prepare individuals to manage it effectively [79]. It was also negatively associated with marital status (r = −0.16, p < 0.05) and education (r = −0.13, p < 0.05), indicating that married and more educated respondents report slightly lower perceptions of discrimination, possibly reflecting greater access to inclusive environments [80]. Perceived gender discrimination was negatively correlated with income satisfaction (r = −0.11, p < 0.05), suggesting that individuals more satisfied with their income perceive lower levels of discrimination. Income satisfaction was positively correlated with age group (r = 0.19, p < 0.01) and marital status (r = 0.20, p < 0.05), indicating that older and married individuals report higher levels of income satisfaction. These findings align with life-cycle income models and recent evidence linking family support to financial satisfaction [76].

The variable work experience (wrk.exp) was positively correlated with age group (r = 0.54, p < 0.01), residence (r = 0.17, p < 0.1), marital status (r = 0.55, p < 0.01), and education (r = 0.28, p < 0.01), reflecting the natural progression of career and family accumulation, as well as its higher prevalence among urban and more educated subpopulations. Marital status (marital) was positively correlated with age.grp (r = 0.88, p < 0.01), as expected, and with education (r = 0.29, p < 0.01). Residence was also positively correlated with age group (r = 0.26, p < 0.01), highlighting the urban concentration of higher education and formal employment opportunities [81].

The results suggest that, in this dataset, entrepreneurial intention tends to be higher among younger, less educated, and less income-satisfied individuals. Perceived gender discrimination is more pronounced among younger respondents and slightly less common among highly educated participants and those satisfied with their income. Our first research hypothesis (H1) was partially sustained, as four of the predictorsnamely age group, marital status, education and income satisfaction—were significantly associated with the mediator variable (i.e., perceived gender discrimination).

We tested the assumptions for the logistic regression involved in our mediation model. The linearity of the logit was assessed for the ordinal predictors (income satisfaction and perceived gender discrimination) using the Box-Tidwell test. The results supported the linearity assumption for both ordinal predictors: income satisfaction X ln (X) = 1.073, p = 0.095, and perceived gender discrimination X ln (X) = 0.866, p = 0.064, with p ≥ 0.05. No linearity of the logit assumption is required for nominal predictors. Model stability for the logistic outcome equation was evaluated by monitoring convergence, and screening for zero cells at each factor level to identify potential separation. The outcome model converged in 4 (four) iterations, reaching a stable maximum-likelihood solution quickly, indicating no convergence problems. No zero-cell levels were found across predictors, suggesting no separation due to empty cells, supporting stable estimation in the logistic outcome model. Overall, these checks indicate that the outcome model was numerically stable, with no convergence or separation issues detected.

We tested the multicollinearity assumption via GVIF (Generalized Variation Inflation Factor) diagnostics. All predictors exhibited adjusted GVIF values GVIF_adj ^1/(2Df) below 5, indicating no multicollinearity concerns, with the highest value being 1.53 for the marital status predictor. Detailed results of this analysis are presented in

Table 3. GVIF statistics.

Variable	Df	GVIF	GVIF_adj	VIF_like	Tolerance_like	Flag_VIFgt5	Flag_VIFgt10
age.grp	4	2.570454533	1.12525575	1.266200503	0.789764337	FALSE	FALSE
residence	1	1.044459585	1.021988055	1.044459585	0.957432929	FALSE	FALSE
marital	1	2.329887593	1.526396932	2.329887593	0.429205256	FALSE	FALSE
education	4	1.158048617	1.018511294	1.037365255	0.963980618	FALSE	FALSE
work.exp	1	1.139615487	1.067527745	1.139615487	0.877488952	FALSE	FALSE
inc.satisf	1	1.072311457	1.035524726	1.072311457	0.932564875	FALSE	FALSE
perc.gen.disc	1	1.082273448	1.040323723	1.082273448	0.923980905	FALSE	FALSE

Source: Authors’ calculation.

. The outcome variable (i.e., entrepreneurial intention) was binary with two categories, i.e., “I haven’t decided”/“I want to do it”.

Model fit was assessed using multiple indices. The overall model fit was LR_X²(13) = 24.37, p = 0.028, indicating that the model was significant but modest. This was reflected by the pseudo-R² values (Mc Fadden = 0.054, Nagelkerke = 0.092) suggesting a smaller influence of the predictors on the variance of the outcome. Model calibration was acceptable, with the Hosmer-Lemeshow test yielding _X²(8) = 11.3, p = 0.185. The model showed fair discrimination, with an AUC of 0.656 (95%CI 0.598–0.714). We expanded the model fit analysis by calculating the model classification metrics. At a 0.50 cutoff, the model yielded an accuracy of 0.689 (95%CI 0.638–0.736), a sensitivity of 0.943, a specificity of 0.133, a PPV of 0.704, an NPV of 0.517, a balanced accuracy of 0.538, and an F1 of 0.806. The no-information rate (baseline), which always predicted the majority class, was ≈0.686. At this cut-off, the accuracy was the same as the baseline. The results indicate that the model was aimed toward detecting positive cases, but generated many false positive ones. The model significantly improved fit over the intercept-only model (LR_X², p < 0.05). Pseudo-R² indicated modest effect sizes (McFadden = 0.054, Nagelkerke = 0.092), which is typical for behavioral outcomes and consistent with a strong baseline due to outcome prevalence. Similar values for Nagelkerke R² were reported by Djordjevic et al. (0.06) in one logistic model predicting entrepreneurial intention among youth in Serbia and Virasa et al. (0.132) in a study investigating the entrepreneurial intention among individuals in six ASEAN countries [82,83], showing that low pseudo-R² values are not uncommon when studying entrepreneurial intention. Discrimination, in case of our model, was fair (AUC ≈ 0.66), and calibration acceptable (Hosmer-Lemeshow test being nonsignificant). Thus, predictors add statistically reliable but modest information. The tested model was aimed to investigate the relationship between the socio-demographic factors and entrepreneurial intention among female students, considering perceived gender discrimination as a potential mediator in this association.

Next, we present and discuss the results of the proposed mediation model. First, the direct effects computed within the model are reported (

Table 4. Direct effects of independent variables.

Independent Variable (IV)	Direct Effect	CI_Lower	CI_Upper	p-Value	Significance
age.grp	−0.3692	−1.4413	0.8503	0.2299
residence	0.1294	−0.3540	0.6040	0.5966
marital	0.5943	0.0030	1.1760	0.0489	*
education	−0.5260	−1.3790	0.3155	0.0837	.
work.exp	−0.0489	−0.5871	0.4698	0.8554
inc.satisf	−0.4209	−0.6709	−0.1806	0.0005	***

*** p< 0.001, * p < 0.05, . p < 0.1. Source: Authors’ calculation.

). The table displays the direct effects of the six socio-demographic variables on entrepreneurial intention. It includes the coefficients that represent unstandardized effect sizes, the confidence intervals (CI_Lower/CI_Upper), p-values, and corresponding significance levels.

The direct effect of age group on entrepreneurial intention was negative but not significant (β = −0.3692, p = 0.2299), suggesting that although younger participants exhibited slightly higher entrepreneurial intention, this relationship was not statistically robust in the selected sample. Recent studies displayed mixed findings regarding this association: in some contexts, younger individuals manifest higher entrepreneurial intentions due to lower perceived risks and greater opportunity-seeking behavior [9], whereas in other contexts, older participants demonstrate higher intentions owing to the experience-related advantages [10].

For residence, a small and not significant positive effect was observed (β = 0.1294, p = 0.5966), indicating that living in urban areas does not strongly predict entrepreneurial intention when other factors are considered. In this sample, the results suggest that other predictors, such as income satisfaction or education level, exert a greater influence on the intention to start a business, despite prior research showing that urban areas provide better access to resources and networks [74].

Marital status shows a significant positive direct effect (β = 0.5943, p = 0.0489), indicating that married individuals were more likely to start a business. This finding aligns with previous studies suggesting that family responsibilities can increase motivation to generate additional income or leverage spousal support for entrepreneurial ventures [51]. At the same time, marriage may enhance access to social and financial capital, thus reducing perceived entrepreneurial risks.

Education level displayed a negative, marginally significant effect (β = −0.5260, p = 0.0837), suggesting the higher educational level may be associated with slightly lower entrepreneurial intentions. Prior research has shown that individuals with higher education studies are more likely to pursue stable employment paths, especially when entrepreneurship is not integrated in their curricula [48]. However, this situation may be reversed if entrepreneurship-focused education is implemented.

Work experience showed no significant association with entrepreneurial intention (β = 0.0489, p = 0.8554), contrasting with studies that linked prior professional exposure to increased opportunity recognition [55]. In this sample of students, however, work experience may be less relevant in strongly influencing entrepreneurial intentions.

Income satisfaction reflects a strong and significant negative direct effect (β = −0.4209, p = 0.0005), indicating that respondents who reported higher satisfaction with their income were less likely to manifest entrepreneurial intentions. This situation supports the push-pull theory of entrepreneurship, whereby dissatisfaction with current conditions serves as a pushes factor toward self-employment [56]. In this case, greater satisfaction appears to reduce the economic incentive to engage in entrepreneurial endeavors.

Two predictors showed significant influence: marital status (positive) and income satisfaction (negative). This underscores the role of personal circumstances in shaping entrepreneurial intentions. Marriage may provide both motivation and resources for developing a business, whereas higher levels of income satisfaction may reduce the perceived necessity for such actions. In addition, the marginal negative effect of education highlights the a potential need for a stronger alignment between academic training and entrepreneurial career opportunities.

The next table presents the indirect effects of the socio-demographic variables on entrepreneurial intention, with perceived gender discrimination acting as a mediator. For each independent variable, the indirect effect was computed as the product of the a-path (predictor → mediator) and b-path (mediator → outcome). Confidence intervals (CI_Lower/CI_Upper), p-values and significance levels were also included in order to assess the statistical significance of these mediated effects.

The results show that all indirect effects are small and statistically non-significant (p-values ranging from 0.562 to 0.986) (

Table 5. Indirect effects.

Independent Variable (IV)	a-Path	b-Path	Indirect Effect	CI_Lower	CI_Upper	p-Value	Significance
age.grp	−0.5112	0.0405	−0.0041	−0.0327	0.0175	0.680
residence	0.0111	0.0726	0.0002	−0.0069	0.0085	0.986
marital	0.3369	0.0542	0.0045	−0.0122	0.0229	0.562
education	−0.4045	0.0450	−0.0026	−0.0215	0.0104	0.690
work.exp	0.0680	0.0733	0.0011	−0.0064	0.0101	0.806
inc.satisf	−0.1319	0.0374	−0.0005	−0.0041	0.0025	0.672

Source: Authors’ calculation.

). The corresponding confidence intervals, for each effect, include zero, providing no evidence that perceived gender discrimination mediates the influence of any of the examined predictors on entrepreneurial intention. This lack of mediation suggests that, although discrimination may be present to varying degrees across subgroups, it does not significantly alter entrepreneurial intentions in this sample. Similar findings have been reported in studies involving younger, well-educated populations, where perceived gender discrimination did not play a mediating role. Instead, other psychological or contextual variables (such as entrepreneurial self-efficacity, opportunity recognition, access to resources) have been shown to play a mediating role [84,85].

For age group, the a_path is negative (−0.5112), indicating that older respondents reported lower perceived gender discrimination. The b_path (0.0405) is weak and non-significant and the indirect effect (−0.0041, p = 0.680) is negligible. For residence, the mediation effect was almost zero (0.0002, p = 0.986), indicating that urban versus rural living environments do not significantly influence entrepreneurial intention through perceived gender discrimination. A slightly higher perceived gender discrimination was reported among married individuals (a_path = 0.3369); however, this did not translate into higher entrepreneurial intention (b_path = 0.0542), creating a small and non-significant indirect effect (p = 0.562). Education showed a negative a_path (−0.4045) indicating that more educated individuals report lower levels of perceived gender discrimination. The corresponding indirect effect was also non-significant (p = 0.690), implying that this influence does not operate through the considered mediator. Work experience exhibited a minimal indirect effect (0.0011, p = 0.806) within the sample, suggesting that prior employment experience does not influence entrepreneurial intention via perceived gender discrimination. Similarly, income satisfaction showed a very weak and non-significant indirect effect (−0.0005, p = 0.672).

The findings do not support the second research hypothesis (H2), as the mediator was not significantly related to the outcome variable (entrepreneurial intention). The third hypothesis (H3) was not confirmed, as none of the predictors exhibited a significant indirect effect on entrepreneurial intention through perceived gender discrimination (all confidence intervals included zero).

These results indicate that in this sample—predominantly composed of young female students residing in urban areas—perceived gender discrimination does not serve as a key explanatory mechanism linking socio-demographic variables to entrepreneurial intention. The study indicates that, for the young, educated female students included in the sample, perceived gender discrimination may be less salient in shaping career decisions, in contrast to professional labor market contexts where discrimination is more pronounced. In Romania, EU membership and the policies promoting gender equality may have changed the perceived gender discrimination among younger population. Other studies on more stratified labor markets have indicated that perceived gender discrimination can directly motivate entrepreneurial escape strategies [42].

In practice, interventions aimed at stimulating entrepreneurial orientation among similar populations should prioritize competence-building, resources access, and opportunity exposure, rather than focusing on perceived gender discrimination. At the same time, maintaining broader equality initiatives remains essential as a matter of fairness and inclusion [86]. Our findings indicates that among female students, variables such as entrepreneurial self-efficacy, opportunity recognition, and institutional support can act as stronger mediators between socio-demographic factors and entrepreneurial intentions than perceived gender discrimination.

Finally, we present and discuss the total effects calculated within the proposed mediation model. The total effect combines the direct effect (independent variable → dependent variable) and indirect effect via perceived gender discrimination (mediator). These effects are listed in

Table 6. Total effects.

Independent Variable (IV)	Direct Effect	Indirect Effect	Total Effect	CI_Lower	CI_Upper	p-Value	Significance
age.grp	−0.3692	−0.0372	−0.4064	−1.5842	0.9188	0.2299
residence	0.1294	0.0008	0.1302	−0.3726	0.6242	0.5966
marital	0.5943	0.0245	0.6189	−0.0404	1.2685	0.0489	*
education	−0.5260	−0.0295	−0.5554	−1.4910	0.3686	0.0837	.
work.exp	−0.0489	0.0050	−0.0440	−0.6067	0.4993	0.8554
inc.satisf	−0.4209	−0.0096	−0.4305	−0.7070	−0.1637	0.0005	***

*** p < 0.001, * p < 0.05, . p < 0.1. Source: Authors’ calculation.

, which also includes confidence intervals (CI_Lower/CI_Upper), p-values and significance levels. The significance levels indicate whether there is a statistically meaningful relationship between each predictor and entrepreneurial intention when both pathways are considered.

The results indicate that age group, residence, and work experience did not exhibit statistically significant total effects. The lack of significance for age group may be attributed to the homogeneity of the sample, as most respondents were between 18 and 25 years old, generating limited variability. The non-significant effect of residence suggests that the academic environment may mitigate the differences between rural and urban backgrounds. Similarly, work experience did not significantly predict entrepreneurial intentions, which may be explained by the temporary and often limited nature of student jobs, which do not provide the entrepreneurial networks associated with the later-stage career experience.

For marital status, the total effect was indicated as significant, due to the direct effect, while the reported confidence interval included zero, reflecting a non-significant relationship. This total effect may be explained by the fact that married individuals could display higher levels of entrepreneurial intention due to family support or greater access to resources. Education showed only a weak effect possibly because higher educational attainment grants access to stable employment options, thereby reducing entrepreneurial intentions. Income satisfaction was the only variable with a total effect clearly significant, as its confidence interval excluded zero. In this case the strong negative total effect (β = −0.4209, p = 0.0005) reflects the pattern observed, indicating a higher entrepreneurial intention among individuals with lower satisfaction with their income. Those who are content with their income may be less motivated to exhibit the same levels of entrepreneurial intention, which is consistent with the push-pull theory of entrepreneurship. Overall, these findings partially support our fourth research hypothesis (H4).

The findings highlight that income satisfaction, which exhibits a negative influence, and marital status, with a positive association, are the most influential predictors tested in the model. The lack of significant indirect effects demonstrates that perceived gender discrimination does not mediate the relationships between the predictors and the outcome. This pattern is consistent with previous research on young, educated, urban populations where structural barriers are perceived but do not decisively shape entrepreneurial intentions [87].

A visual representation of the model results is provided in Figure 3.

In this figure, the coefficients for direct effects (path c), the a-path (predictor → mediator) and b-path (mediator → outcome) are presented. Gray arrows indicate non-significant relationships, while blue arrows represent statistically significant relationships.

The findings indicate that marital status and income satisfaction significantly predict entrepreneurial intention, whereas the other socio-demographic predictors do not show significant effects. Perceived gender discrimination did not mediate any of the tested relationships. The strong association of marital status may reflect the role of family support and pooled resources, while the significant negative effect of income satisfaction aligns with push-pull models of entrepreneurship, where dissatisfaction with current financial conditions motivates entrepreneurial aspirations.

This partial confirmation of our hypotheses highlights the contextual nature of entrepreneurial intention formation. Among young female students, entrepreneurial intention is primarily shaped by personal and financial circumstances, rather than structural barriers such as perceived gender discrimination. Moreover, the limited role of perceived gender discrimination may reflect generational changes and the protective nature of the university environment, where female students encounter fewer structural obstacles compared to women in the labor market. The findings suggest that other mechanisms, such as entrepreneurial self-efficacy, access to resources, or institutional support, may more effectively explain how socio-demographic characteristics translate into entrepreneurial intention.

5. Conclusions

This study examined the influence of socio-demographic variables on entrepreneurial intentions among female students, with a focus on the potential mediating role of perceived gender discrimination. The findings provide valuable insights into the complex dynamics shaping entrepreneurial aspirations within this selected demographic group. Among the six socio-demographic predictors tested, only two—namely marital status and income satisfaction—were significant. Married female students reported higher entrepreneurial intention, suggesting that family systems and shared resources may facilitate, rather than constrain, the decision to start a business. This finding challenges traditional assumptions of marriage as a barrier to female entrepreneurship and aligns with prior research highlighting the supportive role of family networks in entrepreneurial ventures [88].

The other significant predictor, income satisfaction, exhibited a strong negative relationship with entrepreneurial intention, revealing that female students who were more satisfied with their income reported lower entrepreneurial intentions. This finding supports the push-pull theory of entrepreneurship, which links dissatisfaction with current conditions to the motivation to pursue entrepreneurial endeavors. Perceived gender discrimination did not emerge as a significant mediator in the tested model. This suggests that, within this particular sample—predominantly composed of young, educated female students living in urban areas—perceived gender discrimination may be not the primary mechanism that could explain the influence of socio-demographic characteristics on entrepreneurial intentions. This pattern may reflect generational changes in gender-related attitudes, improved educational environments, or specific characteristics of the studied population [89].

The aim of the research was not to achieve broad generalization but rather focus on the contextual understanding of the socio-economic factors shaping entrepreneurial intentions within the specific local environment. Given the context marked by entrepreneurship oriented toward sustainability, policy efficiency is dependent on thorough understanding of local socio-economic contexts and realities. Encouraging entrepreneurial orientation among youngsters requires customized approaches, sensitive to local economic realities, institutional approaches, and cultural characteristics. Insights into how entrepreneurial intentions are formed among women, particularly among young female students, fall within the scope of SDG 5 (gender equality).

These findings contribute to the theoretical understanding of entrepreneurial intentions formation in several ways. First, they demonstrate that perceived gender discrimination did not mediate the relationship between socio-demographic variable and entrepreneurial outcomes (entrepreneurial intention), suggesting that its influence may vary across contexts and populations. Second, the findings highlight that family systems and shared resources can facilitate, rather than constrain, the intentions to start a business, as evidenced by married female students reporting higher levels of entrepreneurial intention. Third, by focusing on female students enrolled at a Romanian university, the study adds new evidence from an emerging economy to the broader international debates on gender and entrepreneurship.

The practical implications of this study cover several aspects. The findings suggest that interventions aimed at fostering entrepreneurship among female students should extend beyond addressing discrimination issues and also target other barriers and motivators. Universities can foster entrepreneurial intentions by integrating specific competence-building activities into their study programs, focusing on opportunity recognition, and providing tailored support within entrepreneurship education, especially for students who remain undecided. Policymakers should recognize that financial dissatisfaction may drive entrepreneurial intention more strongly than income stability when designing and implementing policies to support necessity-driven entrepreneurship. The findings of the study are valuable for the development of policy and programs to support local entrepreneurship initiatives and are relevant for interpretation in terms of contextual transferability. Although perceived gender discrimination was not significant in the sample studied, gender equality initiatives remain essential to ensure inclusivity in the broader labor market and entrepreneurial ecosystem.

Several limitations should be acknowledged when interpreting the results of our study. First, the cross-sectional design limits the understanding of how entrepreneurial intentions evolve over time. A longitudinal approach would provide deeper insights into the stability and development of these intentions throughout students’ academic and early career phases. For instance, Zulkifle et al. [90] tracked the evolution of these intentions among Malaysian university students before and after the pandemic and documented significant shifts.

Second, the measurement of perceived gender discrimination was based on established scales, it may not fully capture the complexity of discrimination experiences in contemporary educational and professional contexts. More nuanced measures—distinguishing between subtle and overt forms of discrimination or considering intersectional experiences—might reveal different results. For example, a study by Jones et al. [36] revealed that subtle discrimination in modern workplaces, although less visible, can affect individuals as much as, or even more than, overt forms of discrimination.

Third, the use of a convenience sample without randomization or stratification limits the representativeness of the sample and the generalizability of the findings. The participants were primarily drawn from economic study programs at a single university, which may not reflect the experiences of all female students in Romania or other contexts. Nevertheless, the results provide valuable insights into the mechanisms shaping entrepreneurial intention within this segment of student population.

Fourth, the model employed in this study was not designed to measure subtle degrees of intention but to differentiate between undecided participants and those with a clear entrepreneurial goal. This formula may reduce variability and mask some intention levels; therefore future research should consider using a multi-item scale that can more accurately capture varying intensities of entrepreneurial intention. This study also focused on entrepreneurial intention rather than on actual entrepreneurial behavior, which represents a common limitation in entrepreneurship research. Although intentions are recognized as strong predictors of behavior, the intention–action gap remains a significant concern in the entrepreneurship literature [91,92].

Considering future research directions, several options emerge from this study. Longitudinal studies, involving the same cohort of female students from university into the early stages of their careers could provide a deeper understanding of how entrepreneurial intentions translate into actual business creation and how different factors influence this evolution over time. Comparative studies examining differences across age groups may help identify generational effects in the relationship between perceived gender discrimination and entrepreneurial intention. Such research could determine whether the non-significant mediation effect observed in this study reflects broader intergenerational changes or is specific to younger generation [93]. Future research should also explore alternative mediating mechanisms that may better explain the relationship between individual characteristics and entrepreneurial intentions among female students. Variables such as entrepreneurial self-efficacy, social capital, opportunity recognition capabilities, and access to resources may serve as prove more relevant mediators in this population [94,95].

Cross-cultural studies examining these relationships across different national and institutional contexts would further enhance the understanding of how cultural factors influence the dynamics approached in this study. The influence of cultural values, institutional support systems, and gender norms on female students’ entrepreneurial intentions remains underexplored, as demonstrated by previous studies [30,96,97]. Research incorporating intersectional perspectives could provide deeper insights into how multiple identity dimensions interact to influence entrepreneurial intentions. Studies examining the interplay of factors such as ethnicity, socio-economic background and field of study with gender—in shaping entrepreneurial aspirations—could contribute to more nuanced theoretical understanding. Recent research has demonstrated the significance of these intersectional effects; for example, race and education jointly influence entrepreneurial intention [98], ethnic and gender identity together shape the barriers faced by women entrepreneurs [99], and innovation activity varies across intersections of gender, ethnicity and geographic context [100].

Intervention studies examining the effectiveness of different approaches to promoting entrepreneurial intentions among female students would provide valuable practical insights. Such research could evaluate the relative impact of skill-based training, mentorship programs, financial support and discrimination awareness initiatives on fostering entrepreneurial intentions. Existing evidence indicates that these components can play a critical role: entrepreneurial education, activities and commercialization support positively impact nascent entrepreneurial intention, mediated through self- efficacy and opportunity recognition [101]. Institutional interventions, including training, mentoring, finance access and networking are perceived by students as high priority factors in shaping their entrepreneurial intentions [102].

This study contributes to the understanding of entrepreneurial intention formation among female students, highlighting the complexity of these relationships. The findings suggest that contemporary approaches to promoting female entrepreneurship in university settings need to evolve beyond traditional gender-focus frameworks and address the specific needs and circumstances of today’s student population. As entrepreneurship increasingly gains recognition as a viable alternative career path, understanding these dynamics becomes essential. Such insights can inform the design of effective support systems that enable all students, and particularly the female students, to acknowledge their entrepreneurial potential.