From Entrepreneurial Alertness to Commitment to Digital Startup Activities: A Mediation Model of Perceived Desirability, Feasibility, and Intentions

Abstract

This study examines the transition from digital entrepreneurial alertness to digital startup intent in connection with perceived desirability, feasibility, and intentions. The theory of planned behavior (TPB) and the entrepreneurial event/potential model (EPM) form the foundation for a mediation model, which is examined by structural equation modeling (SEM) using AMOS on data gathered from 571 Saudi youth engaged in digital entrepreneurship. The results show that digital entrepreneurial alertness has a strong predictive power in relation to intent to start digital ventures, and that this is partly mediated by perceived desirability and feasibility. Intentions, however, fully mediate the relationship between alertness, desirability, feasibility, and actual digital entrepreneurial behavior. This study adds to digital entrepreneurship scholarship by de-mystifying the thought processes bridging opportunity recognition and action, particularly in emerging economies. This study validates the EPM framework and confirms its applicability to include digital entrepreneurial alertness (DEA) as a key antecedent of digital entrepreneurial intentions (DEI) and other factors. This study also highlights the theoretical relevance of the EPM by illustrating its utility in understanding youth decisions to pursue digital entrepreneurship, particularly in transitional countries such as Saudi Arabia. Policymakers and educators in Saudi Arabia should promote attention and amplify desirability/feasibility perceptions to stimulate youth engagement in digital ventures. This work highlights intentions as the determinative gateway between entrepreneurial cognition and concrete digital startup success.

1. Introduction

Leveraging digital technologies in new venture creation has attracted increasing attention from academics and practitioners alike (Camps et al., 2025; Dana et al., 2024; Kraus et al., 2019). This interest is driven by the influence of entrepreneurial activities on economic growth, fostering innovation and job opportunities (Nambisan, 2017). Increased interest in digital entrepreneurship has been especially evident among students (Aloulou et al., 2024; Mir et al., 2023; Sitaridis & Kitsios, 2024; Wibowo et al., 2023), and students’ desires to establish their own ventures have formed a particular focus of scholarly research into the opportunities brought about by digital technologies. Policymakers, universities, and academic institutions recognize the potential of digital entrepreneurship as a promising career path for youth (Biclesanu et al., 2023; Burlea-Schiopoiu & Popovici, 2024). As a result, there is growing interest in understanding the factors that influence digital entrepreneurial intentions (DEI hereafter) and actual digital entrepreneurial behaviors (ADEB hereafter) among students (Aloulou et al., 2024; Ganefri et al., 2025).

Although existing DEI literature has examined several potential factors, many others are still to be thoroughly investigated. Such research is especially needed in nations such as Saudi Arabia, in which rates of entrepreneurial intentions are among the lowest when compared to other Gulf countries such as Oman, Qatar, and the United Arab Emirates (GEM, 2025). The existing literature has also investigated various factors that potentially influence the connections between DEI and ADEB (e.g., education and entrepreneurial ecosystems) (Elia et al., 2020; Ganefri et al., 2025; Wibowo et al., 2023), but again, other potential factors have not been given sufficient attention. Limited research has been devoted to investigating the factors influencing Saudi undergraduates’ intentions to start entrepreneurial ventures using digital technologies, and the connections to actual entrepreneurial behavior (Al-Mamary et al., 2020; Aloulou et al., 2024).

Critical to entrepreneurial activities is the ability to identify and exploit opportunities, filling a gap in the market and/or solving a problem either by establishing a business or by providing a product to (Tang et al., 2012). Digital entrepreneurial alertness (DEA hereafter) refers to activities of active searching, environment scanning, and spotting and recognizing promising opportunities in the digital sphere. This concept has in recent years become an important element of digital entrepreneurship scholarship (Alzahrani & Bhunia, 2024; Wibowo et al., 2023).

Knowledge of how DEA and motivational factors including perceived desirability and feasibility influence DEI remains scant; similarly under-researched is the consequent influence of DEA on students’ digital entrepreneurship in Saudi Arabia. The field lacks empirical studies in particular. This study uses Kruger’s entrepreneurial potential model (EPM hereafter) as a lens to examine these relationships in the digital Saudi youth context. In a systematic review on entrepreneurial alertness, Chavoushi et al. (2021) highlighted the global absence of cognitive research into this concept and its outcomes from an “intrinsic perspective”; previous studies have approached the construct from a “phasic” viewpoint only. This lack of knowledge persists despite research confirming the positive influence of certain cognitive factors, such as entrepreneurial alertness and opportunity recognition, on youth entrepreneurial intention and venture creation (Biswas & Verma, 2021; Wibowo et al., 2023).

To fill these gaps in the literature, the aim of this paper is to investigate, using Ajzen’s theory of planned behavior (TPB hereafter) and EPM constructs including perceived desirability and feasibility, how DEA and DEI influence ADEB and commitment to digital startup activities among students in Saudi Arabia. The study adopts a quantitative methodology, drawing upon 571 surveys completed by Saudi students. Data were analyzed using structural equation modeling. Our analysis revealed that perceived desirability and feasibility partially mediate the relationship between DEA and DEI, whereas DEI fully mediates the relationship between perceived desirability and feasibility, DEA and commitment to startup activities.

This paper contributes to the literature by explaining how DEA and EPM antecedents including perceived desirability and feasibility mediate between DEA and DEI, as well as consequent impacts on students’ ADEB. This paper also contributes to the advancement of knowledge on the applicability of the EPM to predict entrepreneurial intentions in emerging countries and broadens the theoretical framework around digital entrepreneurship. The study shows how different EPM antecedents impact Saudi students’ entrepreneurial career choices in the digital realm. The remainder of this paper is structured as follows. In Section 2, the literature linking entrepreneurial alertness, perceived desirability, and feasibility to DEI and commitment to start-up activities is presented, and hypotheses are developed. In Section 3, the research design and methodology are outlined. Section 4 presents the findings which are discussed further in Section 5. The paper concludes by highlighting the contributions of the study and suggesting avenues for future research.

2. Theoretical Framework and Hypotheses Development

Digital entrepreneurship can be defined in several ways depending on perspective and context (Camps et al., 2025). It encompasses traditional concepts of entrepreneurship including ideation, identifying opportunities and funding, and other forms of resource gathering, as well as the utilization of technologies to improve innovation and creativity levels in an enterprise (Beliaeva et al., 2020; Nambisan et al., 2019). It can be understood as the process of founding new enterprises and/or transforming established ones by using the latest digital technologies (Zaheer et al., 2019). Another widely used definition, introduced by Sussan and Acs (2017, p. 66), includes the activities of “any agent that is engaged in any sort of venture be it commercial, social, government, or corporate that uses digital technologies.” Le Dinh et al. (2018) define digital entrepreneurship as entailing the merging of traditional entrepreneurial practices with new methods of establishing businesses within the digital landscape. This paper draws principally on the definition proposed by Sussan and Acs (2017), which provides a comprehensive view of digital entrepreneurship and is widely used in the literature.

2.1. Entrepreneurial Potential Model

Forming entrepreneurial intentions precedes entrepreneurial behaviors (Krueger et al., 2000). Several theories have proven the predictive ability of entrepreneurial intentions in relation to behaviors, for example, Ajzen’s TPB model and Shapero and Sokol’s (1982) entrepreneurial event model (also known as the entrepreneurial potential model—EPM) (Fayolle & Liñán, 2014). The EPM is a widely used, verified theory for investigating the intention–behavior relationship (Krueger et al., 2000; Krueger & Brazeal, 1994). It comprises three constructs: perceived desirability, feasibility, and propensity to act. The model was originally developed by expanding the TPB (Ajzen, 1991). Krueger and Brazeal (1994) proposed that both perceived desirability and feasibility are predictors of entrepreneurial intention and correspond to Azjen’s attitudes toward entrepreneurship and perceived control behavior. Several studies have confirmed that perceived desirability and feasibility are strong predictors of students’ intentions in entrepreneurship (Caputo et al., 2025; Guerrero et al., 2008).

2.2. Actual Behavior and Commitment to Startup Activities

Actual entrepreneurial behavior refers to individuals’ efforts to start a venture (Kautonen et al., 2013). A substantial body of research has investigated the relationship between intended and actual behavior in this context, drawing on the TPB (Al-Mamary & Alraja, 2022; Aloulou, 2017; Gieure et al., 2020; Kautonen et al., 2013; Lortie & Castogiovanni, 2015). For example, Aloulou (2017) confirmed that attitudes toward behavior (corresponding to perceived desirability) and perceived behavioral control (corresponding to perceived feasibility) strongly correlate with entrepreneurial intentions; and moreover, that intentions and perceived behavioral control are crucial in predicting actual behavior. These findings were consistent with those of previous studies conducted in Spain (Gieure et al., 2020) and Saudi Arabia (Al-Mamary & Alraja, 2022).

Despite these contributions, most research has focused solely on general entrepreneurial intentions or their antecedents, rarely looking beyond intentions in any single study. It is therefore important to gain a better understanding of how intentions lead to entrepreneurial endeavors in the digital sphere. Given the limited available literature on the factors shaping DEI and ADEB (Li et al., 2024), further studies are needed to examine this underexplored research area in more detail. However, from the cited materials, we can form the following hypothesis:

H1. 

DEI positively influences ADEB.

2.3. Perceived Desirability and Feasibility

Perceived desirability refers to individuals’ perceptions that starting a business is attractive and desirable (Krueger et al., 2000). Strong positive perceived desirability includes the eliciting of positive emotions, which in turn motivates individuals to form entrepreneurial intentions (Nguyen & Nguyen, 2024). Perceived feasibility refers to an individual’s belief that they feel prepared to start a business and become an entrepreneur, having the essential skills and knowledge to do so successfully (Krueger et al., 2000; Krueger & Brazeal, 1994). When venture creation is associated with higher levels of feasibility, the perceived risks associated with the venture creation task are reduced, boosting individuals’ confidence and motivating them to form entrepreneurial intentions and, later, behavior (Guerrero et al., 2008). Perceived desirability and feasibility are critical precursors of entrepreneurial motivation and intentions (Aloulou, 2022; Khoi et al., 2023). Several empirical studies have illustrated the role of perceived desirability and feasibility as EPM antecedents in forming entrepreneurial intentions (Alferaih, 2022) and subsequent entrepreneurial behavior (Guerrero et al., 2008; Tan et al., 2021).

A growing body of literature concludes that these relationships persist in relation to DEI (Nguyen & Nguyen, 2024; Wibowo et al., 2023). Yet, despite the extant evidence on the influence of these two constructs (Alferaih, 2022), little is known about the influence of perceived desirability and perceived feasibility on the ADEB of students in Saudi Arabia. Based on the literature presented, the following hypotheses are proposed:

H2a. 

Perceived desirability positively influences DEI.

H2b. 

Perceived desirability positively influences ADEB.

H3a. 

Perceived feasibility positively influences DEI.

H3b. 

Perceived feasibility positively influences ADEB.

2.4. Digital Entrepreneurial Alertness

Recognizing and developing business ideas is the essence of entrepreneurship: Spotting potential ideas early can be a key factor for success (Tang et al., 2012). Entrepreneurial alertness involves noticing potential ideas and opportunities for a start-up, which may be overlooked by others (Kirzner, 1979).

DEA refers to the ability to identify, act on, and exploit opportunities in digital spaces (Wibowo et al., 2023). According to the systematic review by Chavoushi et al. (2021), studies examining entrepreneurial alertness can be categorized into five main research streams. One group of studies has examined the antecedents of entrepreneurial alertness, such as prior knowledge, entrepreneurial passion, personality traits, and business training. These works investigate the role of cognitive factors such as entrepreneurial alertness and self-efficacy (Elnadi & Gheith, 2021; Tang et al., 2012; Urban, 2020), or innovative cognition (Mir et al., 2023), on individuals’ entrepreneurship intentions. Entrepreneurial alertness as a cognitive capability is believed to be an important antecedent of entrepreneurial intentions (Daniel et al., 2021).

A growing stream of literature is investigating how “spotting” or “creating” opportunities might influence individuals’ intent to start a business (Alzahrani & Bhunia, 2024; Wibowo et al., 2023). These studies have suggested that entrepreneurial intention is driven by entrepreneurial alertness, which enables individuals to spot and exploit opportunities and resources (Mcmullen & Shepherd, 2006). Individuals who are more alert to gaps in markets are better at recognizing promising opportunities and are more inclined to act on them (Daniel et al., 2021).

Entrepreneurial alertness has been found to influence not only entrepreneurial intentions but also perceived desirability and feasibility (Tan et al., 2021). Entrepreneurial alertness potentially aids individuals in identifying and evaluating business opportunities and recognizing them as appealing. It might enhance perceived desirability by promoting the ability to recognize promising opportunities and envision the positive outcome of pursuing such an opportunity. In fact, entrepreneurs can better assess the feasibility of ideas through greater capacities of evaluation and judgment, key components of entrepreneurial alertness (Tang et al., 2012). Studies show that individuals with high entrepreneurial alertness are more likely to feel confident acting on ideas when they believe that they have the necessary skills, resources, and knowledge to materialize these ideas in the form of a viable digital business (Li et al., 2024; Mcmullen & Shepherd, 2006). Given the importance of entrepreneurial alertness, Chavoushi et al. (2021) call for scholars to examine more closely how alertness drives both perceived desirability and feasibility toward entrepreneurship. Based on the literature, the following hypotheses are developed:

H4a. 

DEA positively influences perceived desirability.

H4b. 

DEA positively influences perceived feasibility.

H4c. 

DEA positively influences DEI.

H4d. 

DEA positively influences ADEB.

2.5. The Mediating Role of DEI

DEI, as defined by Mir et al. (2023, p. 6169), refers to “the intention of an individual to start a new business through means of digital technology including Internet, World Wide Web, Mobile technologies, Web 2.0 and related technologies.” As Krueger et al. (2000) argued, individuals’ actual behavior is strongly connected to the initial precursors of their intentions. Studies suggest that entrepreneurial behavior toward pursuing an entrepreneurial career path is preceded by entrepreneurial intentions influenced by several factors, including education, social norms, environmental factors, and personality traits (Shirokova et al., 2022).

A recent study conducted on Saudi students examined the mediation of DEA in the motivation–intention relationship and revealed a positive influence of education and DEA on DEI. DEA has elsewhere been found to be a critical factor in predicting DEI (Alzahrani & Bhunia, 2024). However, while this relationship has been widely illustrated, there is still a need for more research investigating the link between alertness and EPM antecedents and entrepreneurial intentions and actual behavior, especially in the rapidly changing digital entrepreneurship landscape (Wibowo et al., 2023). DEA enables individuals to identify opportunities in the digital space, which fosters DEI, which subsequently translates into actual behavior manifested in students’ commitment to starting a business) (Li et al., 2024; Shirokova et al., 2022). Building on previous studies, the following hypotheses are developed:

H5a. 

DEI mediates the relationship between DEA and ADEB.

H5b. 

DEI mediates the relationship between perceived desirability and ADEB.

H5c. 

DEI mediates the relationship between perceived feasibility and ADEB.

2.6. The Mediating Role of EPM Antecedents on the DEA and DEI Relationship

The literature agrees on the mediating roles of perceived desirability and feasibility in the intention–behavior relationship. In a study caried out in Vietnam by Tan et al. (2021), the two EPM antecedents were shown to mediate the relationship between personal traits, including risk-taking, innovativeness, and proactiveness, and Vietnamese students’ intentions and behaviors.

DEA has been shown to enable individuals to spot attractive and promising digital opportunities, so that they perceive digital entrepreneurial activities as more desirable (Wibowo et al., 2023). Furthermore, DEA supports individuals in their assessment of viable and attainable opportunities, enhancing perceived feasibility, which paves the way for the formation of intentions (Alzahrani & Bhunia, 2024; Daniel et al., 2021). Individuals’ actual behavior is indirectly influenced by the boosting of perceived desirability and feasibility, which in turn influences the formation of intentions and actual behaviors (Gieure et al., 2020). Based on these indications, the following hypotheses are proposed:

H6a. 

Perceived desirability mediates the relationship between DEA and DEI.

H6b. 

Perceived feasibility mediates the relationship between DEA and DEI.

2.7. Research Model

Figure 1 depicts the hypothesized relationships between entrepreneurial alertness, perceived desirability and feasibility, DEI, and ADEB. This research model will be tested using a structural equation modeling (SEM hereafter) analysis for direct and indirect linkages.

3. Materials and Methods

3.1. Sample and Data Collection Procedures

Student participants were chosen because they were in the career decision-making stage: According to the Global entrepreneurship monitor (GEM), there is increasing interest in entrepreneurship among young individuals (GEM, 2025).

The sample consisted of students from various universities in Saudi Arabia. Two sampling techniques were used. Convenience sampling, a commonly used technique in entrepreneurship research, was used to reach students at their universities (Abaddi, 2024; Aloulou et al., 2024; Mir et al., 2023). Snowball sampling was used to reach out to other participants.

A structured, closed-ended online questionnaire was developed and disseminated via email and social media platforms. The data collection process began in November 2024 and ended in January 2025. After issuing three reminders, a total of 612 observations were collected; 12 observations used in the pre-test of the survey were excluded, and 29 observations were removed from the sample because of the existence of outliers. The final sample includes 571 observations.

Table 1. Sample characteristics.

	Frequency	Percent
Gender
Male	145	25.4
Female	426	74.6
Age
23 years and less	441	77.2
Between 24 and 33 years	113	19.8
Between 34 and 43 years	12	2.1
44 years and more	5	0.9
Educational level
0 (before university level)	67	11.7
1st year	69	12.1
2nd year	132	23.1
3rd year	127	22.2
4th year	151	26.4
5th (Master student)	8	1.4
6th (Master graduate)	10	1.8
7th (PhD student)	5	0.9
8th (PhD holder)	2	0.4
Specialization
Business administration (all specializations)	143	25.0
Financial economics (all specializations)	93	16.3
Computer sciences (all specializations)	43	7.5
Engineering (all specializations)	19	3.3
Applied Sciences (all specializations)	29	5.1
Social Sciences (all specializations)	39	6.8
Humanities (all specializations)	40	7.0
Others (sharia, languages, sports…)	165	28.9
University
University in Riyadh city	470	80
University outside Riyadh city	114	20

provides an overview of the sample characteristics.

Approximately two-thirds of participants were female. Almost 78% of the participants were less than 23 years old. Regarding students’ education level, the majority (84%) were undergraduate students. Some 41% specialized in business and economics, and approximately 16% specialized in STEM and computer science.

3.2. Measurements

The study employed established self-report measures for the measurement of constructs, using a five-point Likert scale from “strongly disagree” to “strongly agree.”

DEA was measured using a ten-item scale originally developed by Tang et al. (2012) and validated by (Wibowo et al., 2023). Perceived desirability, perceived feasibility, and DEI were measured using a six-item scale drawing on Fellnhofer and Mueller (2018), Jaén and Liñán (2013), Liñán and Chen (2009), respectively.

ADEB was measured using a three-item scale adopted from Aloulou (2017) and Kautonen et al. (2013). The respondents were asked whether they had started, or had thought of starting, their own businesses in the last few months. They were then given a question with four options: (1) have not considered starting a business, (2) have considered it but have not taken action, (3) have not established a business but have begun preparations and aim to start the business soon, and (4) have started a business.

3.3. Strategy of Analysis

The constructs’ reliability and validity were assessed using factor loadings, Cronbach’s alpha, composite reliability (CR hereafter), correlation matrix, and several other indices. Then, a confirmatory factor analysis (CFA hereafter) was performed using AMOS version 21.0 to evaluate the relationships among the different factors and their indicators. Finally, SEM was conducted using AMOS to test the hypotheses.

3.3.1. Reliability and Validity Assessment

To assess reliability, Cronbach’s alpha coefficients, CR, and average variance extracted (AVE hereafter) were calculated (

Table 2. Reliability and validity indices.

Variable	Item Codings	Factor Loadings	% Variance	KMO	Cronbach’s Alpha	CR	AVE
Digital entrepreneurial alertness	DEA1	0.767	59.876	0.933	0.925	0.937	0.599
	DEA2	0.763
	DEA3	0.765
	DEA4	0.730
	DEA5	0.727
	DEA6	0.730
	DEA7	0.802
	DEA8	0.824
	DEA9	0.830
	DEA10	0.791
Perceived desirability	PerDesir1	0.827	59.948	0.869	0.865	0.899	0.600
	PerDesir2	0.774
	PerDesir3	0.838
	PerDesir4	0.684
	PerDesir5	0.748
	PerDesir6	0.765
Perceived feasibility	PerFeas1	0.770	63.732	0.895	0.885	0.913	0.637
	PerFeas2	0.818
	PerFeas3	0.770
	PerFeas4	0.818
	PerFeas5	0.780
	PerFeas6	0.831
Digital entrepreneurial intention	DEI1	0.813	70.155	0.828	0.893	0.922	0.702
	DEI2	0.848
	DEI3	-
	DEI4	0.843
	DEI5	0.835
	DEI6	0.849
Actual digital entrepreneurial behavior	-

). The factor loading exceeded 0.684 and the total explained variance exceeded 59.876%. The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy showed that the strength of the relationships between all the constructs was high, exceeding 0.869 (Tabachnick et al., 2019). Thus, the data were suitable for factor analysis. To assess internal reliability, Cronbach’s alpha coefficient, CR, and AVE were calculated. The alpha values were all greater than 0.70, ranging from 0.885 to 0.925 (see Table 2). The CR values ranged from 0.899 to 0.937, indicating good internal reliability. The AVE values exceeded the threshold of 0.5, ranging from 0.599 to 0.702. All constructs showed good internal consistency. To assess convergent validity, factor loadings and AVE criteria were used. The values of both exceeded 0.6, indicating good convergent validity.

3.3.2. Correlations Matrix and Discriminant Validity

The descriptive statistics and intercorrelations are presented in

Table 3. Mean, standard deviation, AVE, and correlation of constructs.

	Mean	S.D.	Gender	Age	DEA	PD	PF	DEI	ADEB	#Startup Activities
Gender	0.250	0.436	-
Age	1.270	0.539	−0.005	-
DEA	3.787	0.757	−0.087 *	0.124 **	0.770
PD	3.911	0.765	−0.073	0.130 **	0.534 **	0.775
PF	3.799	0.777	−0.089 *	0.090 *	0.690 **	0.566 **	0.798
DEI	3.755	0.871	−0.036	0.078	0.639 **	0.518 **	0.605 **	0.838
ADEB	2.140	0.796	0.076	0.032	0.215 **	0.111 **	0.148 **	0.211 **	-
#Startup Activities	1.600	1.910	0.021	0.015	0.170 **	0.020	0.121 **	0.145 **	0.471 **	-

Note: DEA = digital entrepreneurial alertness, PD = perceived desirability, PF = perceived feasibility, ADEB = actual digital entrepreneurial behavior. Diagonal elements represent the square root of AVE. * p < 0.05, and ** p < 0.01.

. Correlation analysis was conducted to assess the relationships between the constructs. As the table shows, there are several significant positive relationships between the predictor and criterion constructs. There are also significant negative (low-magnitude) relationships between the control variable (gender) and the predictor and criterion constructs.

Following Fornell and Larcker’s (1981) criteria, discriminant validity was established: The square root of the AVE for each construct was greater than any correlations with other constructs. As indicated in Table 3, all constructs showed discriminant validity.

3.3.3. Measurement Model and Model Fit

CFA was conducted to examine the measurement model, the validity of the scales, and the fit indices. All items loaded onto the predicted factors and produced significant t-values. The CFA results showed that the measurement model was a good fit for the study scales with CMIN/DF = 2.396; RMR = 0.038; GFI = 0.912; IFI = 0.958; TFI = 0.951; CFI = 0.958; RMSEA = 0.049.

3.3.4. Common Methods Bias

To detect common methods bias (CMB hereafter), the study employed two techniques. First, the Harman single-factor test was used as a post-hoc technique (Hair et al., 2021). Following the greater-than-one eigenvalue extraction criterion, five factors were extracted, accounting for 63.675% of the total variance. The first factor accounted for 44.742% of the variance (the result was less than 50% for all factors), while the remaining factors accounted for 18.933% of the variance. This study also employed a CFA, testing two different models. The first model was a one-factor, CMIN/DF = 7.620; RMR = 0.069; GFI = 0.699; IFI = 0.784; TFI = 0.765; CFI = 0.783; RMSEA = 0.108. The second model was a five-factor, CMIN/DF = 2.396; RMR = 0.038; GFI = 0.912; IFI = 0.958; TFI = 0.951; CFI = 0.958; RMSEA = 0.049. The poor fit of the one-factor model indicates that no single factor can account for most of the variance (Podsakoff et al., 2003). The fit improvement moving between the one- and five-factor models suggests that CMB is not a major issue in this study.

3.3.5. Structural Model

After conducting CFA, SEM was carried out for all constructs to test the study hypotheses and assess the goodness-of-fit indices. The model showed a good fit with CMIN/DF = 1.949; RMR = 0.032; GFI = 0.925; IFI = 0.967; TFI = 0.961; CFI = 0.967; and RMSEA = 0.041.

4. Results

4.1. Structural Modeling and Hypotheses Testing

The results in

Table 4. Path analysis results.

			Estimate	S_Estimate	S.E.	C.R.	p	Label
ADEB	<---	DEI	0.187	0.193	0.085	2.207	0.027	H1
DEI	<---	PD	0.217	0.183	0.062	3.503	***	H2a
ADEB	<---	PD	−0.138	−0.121	0.093	−1.494	0.135	H2b
DEI	<---	PF	0.413	0.374	0.074	5.593	***	H3a
ADEB	<---	PF	−0.025	−0.023	0.109	−0.226	0.821	H3b
PD	<---	DEA	0.726	0.709	0.056	13.039	***	H4a
PF	<---	DEA	0.874	0.794	0.059	14.701	***	H4b
DEI	<---	DEA	0.409	0.338	0.071	5.764	***	H4c
ADEB	<---	DEA	0.157	0.134	0.104	1.504	0.132	H4d

Note: DEA = digital entrepreneurial alertness, PD = perceived desirability, PF = perceived feasibility, ADEB = actual digital entrepreneurial behavior. *** p < 0.01.

show that DEI has a significant positive relationship with ADEB (β = 0.193, p = 0.027), providing support for H1. Moreover, the results indicate that perceived desirability has a significant positive relationship with DEI (β = 0.183, p < 0.001) but not with ADEB (β = −0.121, p = 0.135), lending support to H2a but rejecting H2b. Similarly, the results show that perceived feasibility has a significant positive relationship with DEI (β = 0.374, p < 0.001) but not with ADEB (β = −0.023, p = 0.821), lending support to H3a and rejecting H3b. DEA also showed significant positive relationships with perceived desirability (β = 0.709, p < 0.001), perceived feasibility (β = 0.794, p < 0.001), and DEI (β = 0.338, p < 0.001). Consequently, H4a, H4b, and H4c are supported. Nevertheless, DEA has no significant relationship with ADEB (β = 0.134, p = 0.132), rejecting H4d.

4.2. Mediation Analysis

The mediation results shown in

Table 5. Mediation analysis.

Hypothesis	From IV	Mediation	To DV	Direct Effect	Indirect Effect	Total Effect	Mediation Test
H5a	DEA	DEI	ADEB	0.134	0.044	0.177 *	Full mediation
H5b	PD	DEI	ADEB	−0.023	0.072	0.049	Full mediation
H5c	PF	DEI	ADEB	−0.121	0.035	−0.085	Full mediation
H6a	DEA	PD	DEI	0.338 ***	0.154	0.492 **	Partial mediation
H6b	DEA	PF	DEI	0.338 ***	0.327 **	0.666 **	Partial mediation

Note: DEA = digital entrepreneurial alertness; DEI = digital entrepreneurial intentions; PD = perceived desirability; PF = perceived feasibility; ADEB = actual digital entrepreneurial behavior. * p < 0.05, ** p < 0.01, and *** p < 0.01.

provide critical insights into the underlying mechanisms driving ADEB via DEA, DEI, perceived desirability, and perceived feasibility.

This table presents mediation model results regarding the relationships between DEA, perceived desirability and feasibility, DEI, and ADEB. All three hypotheses (H5a, H5b, H5c) exhibit complete mediation between the independent variables and the dependent variable ADEB by DEI.

H5a (DEA → DEI → ADEB): Shows a small but significant overall effect (0.177), with an insignificant direct effect (0.134) and significant indirect effect (0.044). Thus, the effect of DEA on ADEB is realized only through DEI. This is in line with models of entrepreneurial intention, such as the TPB, where intention acts as a strong mediator between cognition and behavior.

H5b (perceived desirability → DEI → ADEB): The direct effect is non-significant and negative (−0.023), but the indirect effect is larger and positive (0.072), verifying the mediating role of DEI.

H5c (perceived feasibility → DEI → ADEB): Again, the direct negative effect (−0.121) is compensated by a positive indirect effect (0.035), resulting in a small negative total effect (−0.085), which still confirms full mediation. These patterns support the hypothesis that desirability and feasibility influence behavior only through intention, in accordance with the EPM. Hence, real entrepreneurial action is dependent on intention; beliefs are insufficient unless they facilitate a firm intention.

These models explore more complex mediation chains with perceived desirability and feasibility, incorporating mediation of the relationship between DEA and DEI.

H6a (DEA → perceived desirability → DEI): Both the direct effect (0.338 *) and the indirect effect (0.154) are significant, and partial mediation is suggested, with perceived desirability explaining part of the effect of DEA on DEI.

H6b (DEA → perceived feasibility → DEI): This illustrates a stronger mediation pathway, with an extensive indirect effect (0.327) paired with a strong direct effect (0.338 ***), and therefore a robust total effect of (0.666 **), again confirming partial mediation.

This means entrepreneurial readiness has cognitive and motivational impact: It heightens prospect perceptions (DEA → perceived desirability/feasibility), but also directly impacts DEI independently of these judgments. DEA → perceived feasibility is the direct path with the strongest effect among all paths (β = 0.874), followed by DEA → perceived desirability (β = 0.726). This implies that individuals with greater alertness not only see opportunities but also develop more secure internal convictions concerning the desirability and achievability of taking action toward exploiting these opportunities.

5. Discussion and Implications

5.1. Discussion

The aim of this study was to examine the influence of DEA on Saudi youth ADEB through perceived desirability, feasibility, and DEI. Using the lenses of the TPB and the EPM, this study extends and validates the applicability of the EPM to digital entrepreneurship in the context of Saudi Arabia and introduces DEA as an important antecedent.

Consistent with previous studies (Al-Mamary & Alraja, 2022; Aloulou, 2017), we found a positive association between DEI and ADEB among Saudi youth. This finding confirms the importance of intention formation in motivating individuals to engage in entrepreneurial endeavors (Krueger et al., 2000). Previous literature has found that the TPB framework strongly predicts youth intentions toward entrepreneurship and subsequent entrepreneurial actions (Gieure et al., 2020), particularly in the context of Saudi Arabia (Alferaih, 2022; Al-Mamary & Alraja, 2022; Aloulou, 2017). Most studies examining Saudi university students’ entrepreneurial intentions show that, to a large extent, higher levels of entrepreneurial intention are associated with a higher likelihood of entrepreneurial engagement. This might suggest that the intention–behavior relationship is strengthened by the influence of factors such as government support for entrepreneurship, students’ positive attitudes toward entrepreneurship, and other factors, in forming students’ intentions toward entrepreneurial engagement. Thus, Saudi youths’ intentions to start digital ventures are increasingly likely to materialize into tangible digital businesses.

In contrast to previous studies (Aloulou, 2022; Khoi et al., 2023; Nguyen & Nguyen, 2024; Wibowo et al., 2023), the research found no positive correlation between perceived desirability, feasibility, DEI, and ADEB. One explanation is the intention–action gap in the Saudi context, as documented by Roomi et al. (2020), who highlight several factors exerting discreet influence on intention and actions. While the literature highlights several valid factors in predicting the intentions of youth in Saudi Arabia (Alferaih, 2022; Aloulou, 2022; Aloulou et al., 2024; Alzahrani & Bhunia, 2024; Elnadi & Gheith, 2021), the transformation of these intentions into action is often hindered by external factors such as limited access to funds or lack of training. It is also possible that the effect of such factors might be hindered because of the presence of contextual and cultural factors, such as prevailing social roles, expectations and norms, or institutional barriers, such as institutional voids that were not included in the study (McAdam et al., 2018).

The results also confirmed the influence of DEI antecedents, namely DEA, perceived desirability, and feasibility, on Saudi youth digital startup intentions. Both perceived feasibility and desirability are related to students’ DEI; these findings are confirming recent research conducted on Vietnamese students (Nguyen & Nguyen, 2024). Our results indicate that youth DEA is positively correlated with DEI. The findings are in line with recent studies highlighting the role of DEA as a cognitive capability in predicting DEI among Indonesian students (Wibowo et al., 2023) and Saudi students (Alzahrani & Bhunia, 2024). The findings confirm those of previous studies in suggesting the strong predictive power of DEA in relation to DEI and later success in launching digital ventures. This suggests in turn that Saudi youth have expanded their hands-on experience with digital skills and tools, thus enhancing their capabilities and readiness to spot and exploit promising ideas in the digital realm.

In addition, our findings reveal that DEA is positively correlated with perceived desirability and feasibility. Previous studies have reported similar findings (Tan et al., 2021; Tang et al., 2012). This might be explained by the Saudi government’s Vision 2030 program, which has invested in establishing entrepreneurship-related entities and building good digital infrastructure, allowing the country to keep pace with worldwide digital transformation and incubate newly emerged digital startups (MCIT, 2025). Furthermore, the Saudi government has focused closely on policies and initiatives linked to improving digital-related legislative and regulatory infrastructure, such as digital-connectivity infrastructure, setting up secure and affordable internet services and advancing digital awareness and skills (MCIT, 2025). All these efforts likely contribute significantly to enhancing both the perceived desirability and feasibility of digital entrepreneurship among Saudi youth.

Results from the mediating analysis highlight a multi-pathway mechanism. First, DEI is the central mediator translating upstream variables (DEA, desirability, and feasibility) into behavior. Second, DEA is a foundational cognitive capability that influences intention and behavior not only directly but also through motivational appraisals (desirability and feasibility). Third, the non-significant direct pathways between perceived desirability and feasibility and ADEB underline that perceptions alone are not enough—without intention, they do not necessarily materialize into action. Together, these findings reinforce the primacy of DEI as a mediator and support the use of a cognitive–motivational–behavioral chain in understanding digital entrepreneurship, with DEA playing a catalytic role across all stages.

5.2. Implications

5.2.1. Theoretical Contributions

The contribution of this study is twofold. First, it validates the EPM framework and extends its applicability and scope to include DEA as an important antecedent of DEI and other factors. This paper extends the theoretical basis of youth digital entrepreneurship by highlighting the ability of EPM to explain the influence of DEA, perceived desirability, and feasibility on DEI. The study enriches the theoretical relevance of the EPM by illustrating its usefulness in a new field (i.e., digital entrepreneurship) and confirms the framework’s utility in understanding youth decisions to pursue entrepreneurship across various contexts. Unlike previous studies that examined DEA as a mediator (Abaddi, 2024), and utilized either the lens of social cognitive theory (Alzahrani & Bhunia, 2024) or TPB (Wibowo et al., 2023), this study takes a different approach, examining DEA as a precursor and leaning on the EPM.

The existing framework also connects with established, relevant factors such as entrepreneurial cognition and digital/dynamic capabilities. We can highlight that DEA, for instance, can be considered as a cognitive capability (Chavoushi et al., 2021; Daniel et al., 2021) that exists alongside digital/dynamic capabilities, contributing to opportunity recognition and exploitation within various digital contexts.

Second, it contributes to the literature on digital entrepreneurship (Kraus et al., 2019; Zaheer et al., 2019), particularly in transitional countries such as Saudi Arabia (Aloulou et al., 2024). In particular, it responds to scholarly calls to examine how entrepreneurial alertness drives both perceived desirability and feasibility, facilitating entrepreneurship (Chavoushi et al., 2021).

5.2.2. Managerial Implications

This study provides valuable insights that could potentially increase levels of early-stage entrepreneurial activity in Saudi Arabia by helping youth overcome existing hurdles. Policymakers are encouraged to consider establishing suitable conditions for entrepreneurship, providing both financial and non-financial support to meet youth needs. The government should also continue to make available skill-building programs to provide entrepreneurs with much-needed tools and techniques to establish and grow sustainable digital enterprises. Customized programs could further incline youth intentions toward digital entrepreneurship by cultivating entrepreneurial mindsets. For example, DEA training could be integrated into entrepreneurship curricula, digital opportunity recognition workshops, and mentorships to equip students with skills and knowledge on how to scan, spot, and evaluate promising opportunities. Once youth DEA is established, it will consequently enhance attractiveness towards digital entrepreneurship (i.e., desirability) and beliefs about feeling prepared to start a digital business (i.e., feasibility). Targeted, well-crafted public discourse has proven to be fairly effective in changing public attitudes toward entrepreneurship as a promising career path for youth (Roomi et al., 2020). Continued engagement between governmental and semi-governmental organizations, society, digital entrepreneurs, and different stakeholders will help shed further light on the challenges entrepreneurs face and their requirements moving forward.

5.3. Conclusions

The study extends and validates the applicability of the EPM to digital entrepreneurship among undergraduate students in Saudi Arabia. The framework highlights the predictive explanatory power of DEA that acts as a precursor for perceived feasibility, perceived desirability, and DEI. The study findings also suggest that the government introduce customized programs aimed at cultivating entrepreneurial mindsets that lead to higher youth entrepreneurial intentions toward digital entrepreneurship. Programs could go hand in hand with targeted, well-crafted public discourse to positively impact youth attitudes toward entrepreneurship as a promising career path.

Despite its contributions, this study shows certain limitations. First, it focused on only one research context, Saudi Arabia, which restricts the generalizability of the findings. Comparative cross-country perspectives could broaden the study’s external validity and are required to move forward. A comparison between digital and non-digital entrepreneurs in Saudi Arabia could also help highlight differences or similarities in intentions and motivations around digital entrepreneurship and commitment to digital startups. Moreover, future empirical research could examine the interplay between cognitive factors, digital alertness, and digital and dynamic capabilities in establishing opportunity recognition and exploitation within various digital contexts. There is also a need to explore constructs at the institutional level, moving beyond individual-level constructs.

Second, this study used a non-probability sampling methods to approach the target population, which is a common approach in the literature investigating the entrepreneurial intentions of university students (Aloulou et al., 2024; Mir et al., 2023). However, this sampling method may lead to sample under-representation, which consequently affects the study’s generalizability and its external validity. Supported by previous studies, this target population is suitable for investigating entrepreneurial intention formation prospects, given that university students are in the career decision phase, and they are potential digital entrepreneurs (Abaddi, 2024; Aloulou et al., 2024; Mir et al., 2023).

Third, for future studies, it is recommended to replicate studies on DEI and ADEB across diverse regional and demographic samples. Although quantitative data on a large scale are important for generalizability, a longitudinal study would open new research avenues that would enrich the current findings, given that the cross-sectional approach limits the researcher’s ability to determine causal relationships among constructs.