Technology Admission in Education and Online Entrepreneurship Training at Micro, Small, and Medium Enterprises

The study explores the application of TAM in online entrepreneurship education at the Micro, Small, and Medium Enterprises (SMEs) Food Community in Bandung District, Indonesia. Quantitative methods with survey explanatory approaches are used to answer these research questions. The primary data for this research was collected through questionnaires. The target population consists of the SMEs Food Community, which is spread across 31 districts of Bandung district with 500 active members. A total of 141 valid responses were obtained for data analysis from the SMEs food Community who have followed entrepreneurial education on an online learning platform, online course, or educational portal that provides video-based courses, eBooks, or webinars. This training and education can be website development, digital marketing, content platforms, social media management, and online consulting services. The results of the study using Structural Equation Modeling-Partial Least Squares (SEM-PLS) found that the quality of online entrepreneurship education content has a positive impact on user-perceived usability, and the construction of an online enterprise education platform positively affects the perception of ease of use. Research results show that almost all hypotheses are accepted. Perceived usefulness (PU), perceived ease of use (PEOU), and Attitude Toward Use (ATT) positively influenced the Behavioral Intention to Use (BI) online entrepreneurship education platform.

Introduction

Micro, Small, and Medium Enterprises (SMEs) play a very significant role in the Indonesian economy, contributing 61% of the national GDP, as well as absorbing 117 million jobs, which is 97% of the total workforce. By 2023, about 66 million MSME were recorded, with West Java having the largest number, which is 1,494,723 MSME. Despite this, despite having a huge contribution, many MSMEs in Indonesia are still lagging behind in terms of the use of technology and innovation in today’s digital age. These constraints have become one of the main obstacles to their development and competitiveness (Anastasya, 2023).

By 2023, according to the Ministry of Commerce, as many as 22 million MSME have joined the digital economy. This figure only reached about 33.6% of the total MSME existing in Indonesia. This shows that there are still about two-thirds of MSME unconnected to the digital ecosystem. Many MSMEs have not yet used an online platform to market their products, conduct digital transactions, or use digital tools to manage their business more effectively. (INDEF, 2024)

To address this problem, it is necessary to improve MSME skills and access to technology, optimize the use of digital platforms, develop innovative products and services, and support from governments and related organizations through technology education and training programs. This step can be realized through online education that supports technology mastery and digital strategies.

The development of online education is often hampered by the inconsistencies between educational theory and technology today. As a new model of education, online education is influenced by a variety of factors, including educational concepts, individual understanding, external environments, and networking technologies. All of these factors cannot be fully understood in a single networking environment. Therefore, the Technology Admission Model (TAM) has been introduced in entrepreneurial education for MSME. Online entrepreneurship education can drive entrepreneur success by providing the resources, information, and knowledge needed by new entrepreneurs. This education helps them to recognize and exploit business opportunities, as well as overcome shortcomings in enterprise systems, thus forming a more integrated network of enterprises (Erwinet al., 2023).

Online education has grown rapidly in Indonesia since the Covid-19 pandemic. This growth is also in line with an increase in the number of users of online learning platforms in Indonesia. According to Fig. 1, online learning platform users in Indonesia reached 12.94 million by 2022 and is expected to continue to increase until 2027 (Sihaan, 2023).

Fig. 1. Number of online learning platform users in Indonesia (Sihaan, 2023).

On the other hand, the use of online learning platforms is experiencing various obstacles, especially in the training of MSME in Indonesia. Most MSMEs have low digital literacy, so they face difficulties during online training. (Tuluset al., 2021). Although the number of users of online learning platforms increases in Indonesia every year, there is still resistance to economic digitalization among MSME perpetrators. Many elderly MSME persons tend to be reluctant to follow technological developments and are unwilling to learn digital technology because it is considered complicated (Nisrina, 2023). MSME’s digital literacy is currently low, while digital literature is crucial to driving MSME’s business success (Anatan, 2023; Gunawan Wibowo, 2021; Haleemet al., 2022).

Based on these issues, the study aims to study online education for MSME perpetrators using the Technology Acceptance Model or Technology Acceptability Model (TAM). The TAM model has become an important framework for understanding how MSME persons receive online training platform technology. Most of the latest research on online entrepreneurship education is done in general, with little focus on the entrepreneurial networking aspects. Therefore, this research can be a reference for the development and enhancement of the innovation industry as well as online enterprise while supporting the establishment of a more integrated enterprise network. In this study, TAM is introduced as a new model in online entrepreneurship education, providing a new perspective in the study of online entrepreneurial education.

The research uses the conceptual framework of TAM that covers Platform Construction (PC), Content Quality (CQ), Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Attitude Toward Usage (ATT), and Behavioral Intention to Use (BI). Therefore, Perceived Usefulness and Perceived Ease of Use determine the intentions of users through their attitudes. (Mardianaet al., n.d.; Syarwani & Ermansyah, 2020).

Furthermore, behavioral intentions are influenced by user attitudes and the usefulness of technology. TAM considers user intent to be the most direct manifestation of user behavior. TAM also indicates that perception of usability is directly influenced by the perceptive ease of use; if technology is easier to use, the user will be more sensitive to its usefulness (Haleemet al., 2022; Maritsaet al., 2021; Pibriana, 2020; Ranugalihet al., 2020; Setyaningsih, 2023).

In the context of online education, factors such as the characteristics of the online teaching system, the methods of teaching, the quality of teachers, and the differences between students, especially MSME participants who attend online classes, influence their understanding of the usefulness and simplicity of the technology. It then affects their attitudes and intentions in using information technology, which ultimately determines the effectiveness of using online education (Mardhiyahet al., 2021). Based on the explanation, the study aims to explore the application of TAM in online entrepreneurship education in the UMKM Food Community in Bandung.

Literature Review

Technology Acceptance Model (TAM)

In the field of information systems and technology adoption, the Technology Acceptance Model (TAM) has been widely used to understand user acceptance and usage behavior. The centerpiece of this model is the construction of usability perception (PU) and the perception of ease of use (PEOU). These constructions are influenced by a variety of factors, including platform construction and content quality. This literature review explores the relationship between platform building and the quality of content, as well as how they affect PU and PEOU.

Platform design refers to the structural and functional design of an online platform. It covers aspects such as user interface design, navigation, system reliability, and performance. Effective platform construction can significantly improve the user experience, thus affecting PU and PEOU. Several studies have highlighted that well-built platforms improve PEOU. For example, Davis (1989) argues that ease of navigation and intuitive interface design contribute to user perception of ease. In addition, Venkatesh and Davis (2000) suggest that system reliability and rapid response times reduce user frustration and increase PEOU.

Although platform construction primarily affects the PEOU, it also indirectly affects PU. A smooth and efficient platform can lead to higher productivity and efficiency in the completion of tasks, thereby improving Davis (1989) perception of utility. Gefen and Straub (2000) found that users are more likely to consider a platform useful if it consistently works well and meets their needs efficiently.

Furthermore, content quality covers the accuracy, relevance, completeness, and timeliness of the information provided on the platform. High-quality content is crucial to ensuring that users find the platform valuable and easy to use. High-quality content directly affects PU by providing users with valuable information that helps them reach their goals. DeLone and McLean (1992) suggested that the success of an information system is partly determined by the quality of its content.

Furthermore, Zhang and Von Dran (2000) indicates that users find the platform more useful when its content is accurate, up-to-date, and relevant to their needs. While content quality is more closely related to PU, it also affects PEOU. Clear, well-organized, and easily accessible content can reduce the cognitive burden on users, thus making the platform easier to use. When users can quickly find the information they need, they tend to consider the platform user-friendly.

The combined impact of platform construction and content quality can create a synergistic impact on PU and PEOU. For example, a platform with an excellent user interface and high-quality content is likely to be considered useful and easy to use. As noted by Wixom and Todd (2005), the integration of good system design with valuable content results in higher levels of user satisfaction and acceptance. Empirical studies support the synergistic influence of platform construction and content quality. Similarly, that e-commerce platforms that are superior in content construction and quality experience higher user engagement and satisfaction.

The two main constructions in this model are the perceived usefulness (PU) and the perceptive ease of use (PEOU). Both constructions are believed to have a significant influence on attitudes towards use. (attitude toward use atau ATT). Some studies show that PU has a direct and significant influence on ATT. When users feel that the technology is beneficial and can improve their performance, their attitude towards the use of such technology tends to be positive (Venkatesh & Davis, 2000). For example, Teo (2011), who studied teachers, shows that PU is a powerful predictor of ATT’s use of educational technology.

Research by Igbariaet al. (1996) shows that PEOUs have a positive and significant influence on ATT, but this influence is stronger when mediated by PU. In other words, ease of use contributes to a positive attitude mainly through improved usability perception. Wixom and Todd (2005) proposed an integrated model that combines PU and PEOU in predicting ATT. They found that PU and TEOU together explain most of the variability in ATT, which suggests that both factors are important in determining user attitudes to technology.

The Technology Acceptance Model (TAM) indicates that ease of use and benefits of technology influence behavioral intentions, which in turn affect the end-user experience. Some researchers state that attitudes are the result of learning and experience and that ease of use and understanding of the benefits of technology can be enhanced through learning, thereby modifying user attitudes (Mooya & Phiri, 2021; Mulyonoet al., 2020). Therefore, psychological factors can transmit the influence of the external environment on the adoption of new technologies, and external parameters do not directly influence the intentions and behavior of use. This influence comes mainly from the internal understanding and attitude of the user (Liu & Chen, 2021).

Online Education

Online education is a learning model based on network technology, which is the integration of network technology with educational reform. Online education breaks through a rigid teaching model, expands online teaching channels, and further optimizes teaching resources. Online education has a variety of forms and platforms, which are tailored to people from all walks of life (Wu & Song, 2019).

In general, network technology-based educational models can greatly stimulate learning interests, improve the quality of personalized education, and help develop talents comprehensively (Wuet al., 2019). Most importantly, innovation and entrepreneurship education can train innovative talents and enterprising as well as raise their awareness of innovation, which is the key to innovation and enterprise programs (Dixit & Prakash, 2018). Meanwhile, innovation education and enterprises are responsible for nurturing talents with innovative abilities and businesses (Vermaet al., 2018).

Conceptual Framework

Based on the background that has been presented, this study aims to analyze the acceptance of technology in education and training online entrepreneurship at UMKM Food Community in Bandung district, the relationship between variables in this study can be seen in Fig. 2. The hypothesis in this study is as follows:

H₁: The Construction Platform (PC) has a significant positive influence on the Perception of Usability (PU). H₂: Content quality (CQ) has a significant positive influence on usage perception (PU). H₃: Construction Platform (PC) has a significant positive influence on the perception of ease of use (PEOU). H₄: Content quality (CQ) has a significant positive influence on the perception of ease of use (PEOU). H₅: Usage perception (PU) has a significant positive influence on usage attitudes. (ATT). H₆: The perception of ease of use (PEOU) has a significant positive influence on attitudes towards use. (ATT). H₇: Attitude toward Usage (ATT) has a significant positive influence on behavioral intentions to use (BI). H₈: Attitude toward Usage (ATT) partially mediates the influence of the perception of use (PU) on the intention of behavior to use (BI). H₉: Attitude toward Usage (ATT) partially mediates the influence of the Perception of Usability (PEOU) on the behavioral intention to use (BI).

Fig. 2. Conceptual framework.

Research Method

This research uses descriptive quantitative methods with an explanatory survey approach. Quantitative methods are based on the philosophy of positivism, used to research a particular population or sample, data collection using research instruments; data analysis is quantitative/statistical with the aim of describing and testing the hypothesis that has been established (Sugiyono, 2017). The variables in the study are Construction (PC), Content Quality (CQ), Perceived Usefulness (PU), Perceptured Ease of Use (PEOU), Attitude Toward of Usage (ATT), and Behavioral Intention to Use. (BI). These questions are based on items that have been validated from previous research. The variables in this model are derived from a literature review in research journals (Su & Li, 2021; Gaffaret al., 2022). The analysis process involves two main stages, namely, the measurement model and the structural model.

The population in this study consists of the UMKM Food Community, which is spread out in 31 districts of the Bandung district and has 500 active members. The number of samples in this study is determined using Ferdinand’s method, which is the number of indicators × 5–10. Of the 350 questionnaires distributed online (April–May 2024), 141 were completed and valid for data analysis. The sample of this research uses purposive sampling with the main criteria of having followed entrepreneurial education on an online learning platform, online course, or educational portal that provides video-based courses, eBooks, or webinars. This training and education can be web development, digital marketing, content platforms, social media management, and online consulting services.

The data analysis is done using Structural Equation Modeling-Partial Least Squares (SEM-PLS) with the help of the smart application PLS 4. According to Abdillah and Hartono (2015) and Handayani (2016), Partial Leost Square (PLS) is a structure equation modeling equation analysis that uses variable bases simultaneously that can perform measurement model testing and structural model testing. The application or software used in this study to calculate the results of data analysis is smartPLS. SmartPLS is a stand-alone application used for calculating Structural Equation Modelling (SEM). Table I explains the Operationalization of Research Variables.

Table I. Operationalization of Research Variables

No	Variable	Indicator	Item	Questionnaire statement
1	Attitude Toward Use (ATT)	Self-confidence in online learning	ATT1	I feel confident to follow online learning in the learning process
	(Singhet al., 2020)	Self-learning facilities through online training	ATT2	I feel easier to learn on my own through online training
2	Behavioral Intention to Use (BI)	Intent to use E-Learning platform routine	BI1	I plan to use this e-learning platform routinely to learn online entrepreneurship in the next few months
	(Su & Li, 2021)	The desire to integrate the e-learning platform into the learning process	BI2	I intend to integrate the use of this e-learning platform in all aspects of my entrepreneurial learning
3	Content Quality (CQ)	Richness and quality of content	CQ1	The resources obtained are rich and the quality of the content is good
	(Su & Li, 2021)	Values and attractions of entrepreneurship activities design	CQ2	The design of business-related activities is very valuable and interesting
		Platform basic service benefits and quality	CQ3	The basic service provided by the platform is very nice and helpful
4	Platform Construction (PC)	Influence and number of platform users	PC1	Online entrepreneurship education platform is very influential and has many users
	(Su & Li, 2021)	Platform operation and maintenance capabilities	PC2	The online entrepreneurship education platform has strong operational and maintenance capabilities and can operate steadily
5	Perceived Ease of Use (PEOU)	Accessibility of resources and services	PEOU1	Easy to get the resources and services you need
	(Gaffaret al., 2022)	Platform participation facilities	PEOU2	It’s easy to participate in this platform without much effort
6	Perceived Usefulness (PU)	The usefulness of knowledge and resources	PU1	This platform can provide a lot of useful knowledge and resources
	(Gaffaret al., 2022)	Usefulness for personal growth and development	PU2	This platform is useful for personal growth and development

Result and Discussion

Research Results

Respondent Demographics

The study tested a total of 141 respondents whose answers were valid and reliable for further analysis. Table II presents a summary of the demographics of respondents, including gender, length of work, last education, and age. Out of 141 respondents, based on the gender classification, female respondents completed the survey more than men, reaching 65.96% compared to men 34,04%. Then, in terms of age, most respondents were 20–30 years old (48.23%), and while the least were respondents who were over 50 years old, only 14 people (9.93%). The second-largest age to fill in the questionnaire was 41–50 years (21.28%), then 31–40 years (20.57%). Next, according to the last education, the majority of respondents had a Bachelor (S1) education background with 34.4%, followed by High School and Undergraduate (D3/D4), then Master (S2) and Doctoral. (S3). Lastly, from the long term of UMKM Food Community, respondents who completed the questionnaire had a 0–2 yearlong business of 63 people (44.68%), then a 3–5 years long business (32.62%), a business of more than 8 years there were 17 people and a 6–8 yearlong job of at least 15 people (10.64%).

Table II. Respondent Demographics

Item	Amount	Precentage
Gender
Female	93	65.96%
Male	48	34.04%
Total	141	100.00%
Age
20–30 years	68	48.23%
31–40 years	29	20.57%
41–50 years	30	21.28%
Over 50 years	14	9.93%
Total	141	100.00%
Last education
High school	35	24.82%
Diploma	35	24.82%
Bachelor	48	34.04%
Magister	16	11.35%
Doctor	7	4.96%
Total	141	100.00%
Length of business
0–2 years	63	44.68%
3–5 years	46	32.62%
6–8 years	15	10.64%
Over 8 years	17	12.06%
Total	141	100.00%

Measurement Model Analysis

In this research method, all the variable indicators used have been described, but based on the Model Measurement test, there are several Outer Loading, Construct Reliability and Validity, and Discriminant Validities. The retest results are shown in Table III.

Table III. Construct Reliability and Validity

Outer model		Convergent validity		Reliability
		Outer loading	AVE	Cronbach’s alpha	Composite reliability
		>0.70	>0.50	>0.70	>0.70
ATT	ATT1	0.879	0.795	0.743	0.886
	ATT2	0.904
BI	BI1	0.936	0.878	0.861	0.935
	BI2	0.938
CQ	CQ1	0.804	0.701	0.788	0.876
	CQ2	0.837
	CQ3	0.870
PC	PC1	0.895	0.810	0.765	0.895
	PC2	0.904
PEOU	PEOU1	0.909	0.803	0.756	0.891
	PEOU2	0.884
PU	PU1	0.902	0.818	0.777	0.900
	PU2	0.907

Referring to Table III, objective information obtained that all indicators are significant in measuring their latent variables and have factor load coefficients above the required minimum value of 0.70. From AVE statistics, the test results showed that six measurement models gave an AVE value greater than 0.50. This means that all indicators used to measure the latent variables being studied have sufficient convergence validity. Furthermore, from Cronbach’s Alpha and Composite Reliability, the statistical test results showed that Cron Bach’s alpha and composite reliability values for six measurement models yielded a value above the required minimum value of 0.70. This indicated that the six measuring models had adequate internal reliability. The next step in testing the validity of discrimination is to use the cross-loading test. Cross-loading itself is the Outer Loading value test in which a variable structure must have a larger value for its own variable compared to other variables.

Table IV provides information that, seen from cross-loading analysis results, all indicators used to measure six measurement models give factor load coefficient values that are greater than cross-load values. This shows that the six measuring models have sufficient discriminatory validity. Further, the validity of the study was followed by a validity test of discrimination through the Fornell-Larcker criteria.

Table IV. Cross Loading

Constructs	ATT	BI	CQ	PC	PEOU	PU
ATT1	0.879	0.490	0.566	0.379	0.501	0.584
ATT2	0.904	0.618	0.566	0.496	0.539	0.572
BI1	0.580	0.936	0.563	0.499	0.689	0.635
BI2	0.591	0.938	0.511	0.504	0.610	0.628
CQ1	0.515	0.436	0.804	0.497	0.450	0.497
CQ2	0.512	0.473	0.837	0.582	0.446	0.543
CQ3	0.564	0.523	0.870	0.503	0.513	0.655
PC1	0.409	0.432	0.577	0.895	0.408	0.493
PC2	0.478	0.530	0.551	0.904	0.494	0.448
PEOU1	0.588	0.626	0.516	0.491	0.909	0.632
PEOU2	0.452	0.616	0.493	0.405	0.884	0.641
PU1	0.588	0.596	0.657	0.483	0.581	0.902
PU2	0.584	0.622	0.575	0.461	0.700	0.907

Based on Table V, it is seen that according to the Fornell-Larcker criteria, six measurement models have a larger square AVE root value than the correlation coefficient between the measured structures and other structures. This means that the six measuring models have sufficient discriminatory validity.

Table V. Fornell-Larcker Criterion

	ATT	BI	CQ	PC	PEOU	PU
ATT	0.891
BI	0.625	0.937
CQ	0.634	0.573	0.838
PC	0.494	0.535	0.627	0.900
PEOU	0.584	0.692	0.563	0.502	0.896
PU	0.648	0.673	0.681	0.522	0.709	0.904

Analysis Structural Model

The proposed hypothesis is tested for data analysis and to verify the relationship in the proposed model. The regression value shows the magnitude of the direct influence coefficient between variables, as shown in Table VI.

Table VI. Direct Effect (Path Coefficient, P-Value, T-Value, and R²)

Model	Track coefficient	P-value	T-value	R2	Test results
Model PU
H1: PC influences PU	0.157	0.172	1.365	0.478	H0 accepted
PC -> PU					H1 rejected
H2: CQ influences PU	0.582	0.000	5.536		H0 rejected
CQ -> PU					H2 accepted
Model PEOU
H3: PC influences PEOU	0.156	0.079	1.968	0.529	H0 rejected
PC -> PEOU					H3 accepted
H4: CQ influences PEOU	0.070	0.115	0.610		H0 accepted
CQ -> PEOU					H4 rejected
Model ATT
H5: PU influences ATT	0.471	0.000	4.901	0.451	H0 rejected
PU -> ATT					H5 accepted
H6: PEOU influences ATT	0.250	0.015	2.430		H0 rejected
PEOU -> ATT					H6 accepted
Model BI
H7: ATT influences BI	0.625	0.000	9.124	0.390	H0 rejected
ATT -> BI					H6 accepted

Referring to Table VI, the following information is obtained:

1. For the PU model, the test results showed that the path coefficient from PC to PU is insignificant (p1 = 0.157, p = 0.172 > 0.05), and H1 is not supported data. This means that the PC does not have a direct effect on the PU. Meanwhile, for the path factor from CQ to PU of p2 = 0.582 with p = 0.000 < 0.05, it means that CQ has a significant effect on PU (H2 data supported). Thus, it can be concluded that only CQ affects PU in this study. The determination coefficient for the PU model is 0.478 (moderate). This means that 47.8% of the variation that occurs in the PU can be explained by this model. The remaining 52.2% are described by other variables that the model does not describe. This means that the model has sufficient predictive validity or relevance in predicting variations on latent PU variables.
2. For the PEOU model, the test results showed that the path coefficient from PC to PEOU is significant (p3 = 0,156, p = 0,049 < 0,055) H3 supported data. This means that the PC has a direct influence on the PEOU. Meanwhile, for the CQ to PU path coefficients of p4 = 0.070 with p = 0.542 > 0.05, this means that CQ is not influenced by the PEOU (H4 is not supported by the data. Thus, it can be concluded that only the PC affects the PEOU in this study. The determination coefficient for the PEOU model is 0.529 (moderate). This means that 52.9% of the variation that occurs in the PEOU can be explained by this model. The remaining 47.1% is described by other variables that are not described in the model. This means that the model has sufficient predictive validity or relevance in predicting variations on the PEOU latent variable.
3. For the ATT model, the results of the hypothesis test indicated that PU and PEOU had a significant influence on ATU in this study. The PU to ATT path coefficient is p5 = 0.471, p = 0,000 < 0.05, which means H5 supported data where PU has a direct influence on ATT. Then, for the PEOU to ATT path coefficient of p6 = 0.250, p = 0,015 < 0,05, that means H6 supports data where PEOU has a direct influence over ATU. The determination coefficient for the ATT model is 0.451 (moderate). This means that 45.1% of the variation that occurs in the ATT can be explained by this model. The remaining 54.9% are described by other variables that are not described in the model. This means that the model has sufficient predictive validity or relevance in predicting variations on the latent variable ATT.
4. For the BI Model, the result shows a path coefficient from ATT to BI (p7 = 0.625, p = 0.000 < 0.05) (H7 supported data, very significant. Thus, it can be concluded that the ATU directly affects the BI. The determination coefficient for the BI model is 0.390 (low). This means that only 39% of the ATU variables can explain the variations that occur in BI. The remaining 61% are described by other variables that are not described in the model. This means that the model has sufficient predictive validity or relevance in predicting variations on the BI latent variable.

The above explanation explains the results of the direct effect and path coefficient tests for the four models tested.

To test the H8 and H9 hypotheses, use the output specific indirect effect. Table VII shows the specific indirect effect output. Here are the results obtained:

1. H₈ data supported. That is, ATT significantly mediates the influence of PU on BI (p8 = 0.294, p = 0.000; 95% CIBC ≠ 0).
2. H₉ data supported. That is, ATT significantly mediates the influence of PEOU on BI (p9 = 0.156, p = < 0.05; 95% CIBC ≠ 0).

Table VII. Specific Indirect Effect

Model	Specific indirect effect	P-value	95% CIBC*		Test results
			2.5%	97.5%
H8 = ATT mediates the influence of PU on BI	0.294	0.000	0.173	0.440	H0 rejected
PU -> ATT -> BI					H8 accepted
H9 = ATT mediates the influence of PEOU on BI	0.156	0.032	0.019	0.307	H0 rejected
PEOU -> ATT -> BI					H9 accepted

Referring to the findings of the study, given the size of the estimated specific indirect effect coefficient, the ATT variable tends to have a relatively stronger role in mediating the influence of PU on BI.

Discussion

Based on the results of the research described in Table V, the PC has a positive but non-significant influence on the perceived usefulness (PU). This means that the platform’s operational and maintenance capabilities, as well as the influence and number of users of the platform, cannot be a detrimental measure that the user feels the benefits of the online training platform. The results of this study are reversed compared to the research Su & Li (2021). This is due to differences in research locations, and the digital literacy capabilities of entrepreneurs who undertake online training on previous research with current research. Locations that have better access to technology can affect respondents’ ability to use digital technology, and locations with more traditional cultures will have lower levels of digital literacy than more modern locations (Zaahidahet al., 2023). However, the results shown by CQ against PU compared to the best of PC, CQ has a positive and significant influence on PC. This situation occurs due to the wealth and the quality of content, value, and power of the attractive design of online training activities, as well as the benefits and qualities of the basic service of the platform, which can be felt the benefits by MSME participants who follow online training. The results are in line with the research conducted by Su & Li (2021).

The findings from Hypothesis 3 prove that the PC has a positive and significant influence on the PEOU. This means that the higher the construction of the platform, the easier it is for the online training platform to be used by MSME. This implies that the construction platform provided by the training platform is easy to use by respondents at the time of online training. However, the results demonstrated by Hypothesis 4 prove that CQ is positive but not significantly influenced by the platform. This study compares with a study conducted by Sarinet al. (2023), which says that when a consumer perceives the content of a complete and adequate platform and provides a variety of information that can meet their needs, then they will feel that the platform will be easily used.

The findings from Hypotheses 5 and 6 prove that PU and PEOU have a positive and significant influence on ATT. This means that the higher the perception of the usefulness and ease of use of technology, the higher the positive attitude of the user towards the technology and the greater their desire to use it. When individuals feel that a technology is useful and easy to use, they tend to have a more positive attitude towards the technology and are more interested in using it. The results of this study are consistent with the research carried out by Almulla (2021). PU and PEOU are often connected. Technologies that are considered useful (PU) but difficult to use (low PEOU) may still not be well accepted.

On the contrary, technologies that are easy to use (high PEOU) but not useful (low PU) may also not be attractive. A combination of the highest of the two usually gives the most significant impact on the positive attitude of the user. The TAM model states that PU and PEOU jointly influence attitudes towards use (ATT), which then influence behavioral intentions to use technology (BI) and eventually actual use (AU).

The findings from hypothesis 7 prove that ATT influences BI. TAM assumes that individuals make decisions rationally based on their evaluation of the benefits and ease of use of technology. ATT reflects a combination of cognitive components (confidence in benefits and ease of use) and affective components (Feelings of technology). These cognitive and affective components together form attitudes that then influence the intention to use technology. Based on the theory of attitudes and behaviors, a positive attitude to a behavior increases the likelihood of an individual having the intention to do so. In the context of TAM, positive attitudes to technology increase the probability that an individual intends to use it.

It has been explained earlier that, PU and PEOU influence the ATT. If one feels that the technology is useful (PU) and easy to use (PEOU), they will tend to have a positive attitude toward the technology. The findings of hypotheses 8 and 9 prove that ATT can mediate the influence of PU and BI as well as the influences of PEOU and BI. This means that when individuals feel that a technology is beneficial and easy to use, they will have more positive attitudes to the technology and be more interested in using it. This positive attitude will then increase the chances of users to use the technology and benefit from it (Al-Adwanet al., 2023; Almulla, 2021; Sitiet al., 2021).

When the user feels that the technology is useful (PU), a positive attitude towards the technology (ATT) will be formed. This negative attitude will then affect the user’s intention to use the technology. This means that the PU not only directly affects the BI but also indirectly through the ATT. Similarly, when the user thinks that it is easy to use (PEOU), positive attitudes toward the technology will form.

Conclusion

Overall, good content quality and platform construction improve the ease of use and perception of the usefulness of technology, which ultimately shapes the positive attitude of users towards the technology and encourages their intention to use it. PU and PEOU are cognitive attitude-forming factors (ATT). Users evaluate technology based on its ability to improve performance (PU) and its ease of use (PEOU). These evaluations create positive or negative attitudes toward technology. A positive attitude reflects the belief that the technology will deliver a good experience and the desired results, which ultimately encourages the intention to use the technology.

Online entrepreneurship education can drive MSME’s success by providing the resources of information and knowledge required by MSMEs. It helps them to recognize and harness business opportunities, as well as overcome shortcomings in the entrepreneurial system, forming a more integrated network of enterprise. Technological innovation in education is very useful in achieving this goal by introducing more effective and efficient learning methods.

Technology developers must ensure that the product or service they offer is truly beneficial to the user. This can be done by adding features that increase productivity, efficiency, and user satisfaction. Besides being useful, technology should be easy to use. An intuitive and simple interface design, clear user guidance, as well as good technical support can improve ease of use.