Exploring the determinants of electronic commerce adoption in a municipality with limited internet access

Abstract

In many municipalities in Colombia, the diffusion of electronic commerce (e-commerce) and the expansion of internet coverage have led to economic development. However, some rural regions and municipalities still have limited access to this technology, and its benefits have not been fully realized there yet. This study aims to identify the characteristics of this digital gap in Colombia by analyzing e-commerce acceptance in a mostly rural municipality: Andes (Antioquia). The objective is to establish the determinants of consumer adoption behavior there and provide the basis for new public policies and business strategies. The Technology Acceptance Model (TAM) and hypotheses in a structural equation model were used to represent e-commerce acceptance behavior in Andes. The reliability and validity of the measurement model were established by Confirmatory Factor Analysis (CFA) using a sample of 400 questionnaire responses by internet users in Andes. The results indicate that Ease of Use, Perceived Usefulness, and Attitude toward e-commerce significantly influence their Behavioral Intention to use the latter. The findings suggest that, in the Colombian countryside, the Perceived Usefulness of this technology should be improved, and the strategies implemented to drive the adoption of this technology should be tailored for that context. Although acknowledging limitations in its exploration of cultural factors, this study contributes to the understanding e-commerce adoption in Andes. Local businesses and authorities should work together to promote regional economic growth and digital transformation.

Keywords:

• Electronic commerce
• e-commerce
• technology acceptance
• TAM
• online behavior

REVIEWING EDITOR:

• Balan Sundarakani, University of Wollongong in Dubai Faculty of Business, United Arab Emirates

SUBJECTS:

• Management of Technology & Innovation
• Marketing Management
• Management & Organization

1. Introduction

Electronic commerce (e-commerce) refers to “digitally enabled commercial transactions between organizations and individuals” (Laudon & Guercio Traver, 2014). Although a developing technology, it has become a global trend due to its most salient features: ubiquity, transnational reach, interactivity, and social impact. Nevertheless, it also poses considerable challenges for organizations operating in developing countries—where internet penetration levels may be lower and the use of technology has not spread on a large scale.

E-commerce has become a crucial tool for the economic growth and competitiveness of small and medium-sized enterprises (SMEs) in several countries (Hasan & Mardhani, 2021). In addition, it has proven to be essential for reaching Generation Z, as found by Lestari (2019), who emphasized the importance of interacting with digitally connected population segments. Due to the global relevance that this form of digital commerce has gained in recent years, it is necessary to study its adoption in local contexts, for example, Andes (a rural municipality in Antioquia, Colombia). Understanding the factors that determine e-commerce adoption in Andes is crucial for driving the development of its local economy, improving the competitiveness of its local businesses, and meeting the ever-changing needs of consumers in the digital age.

E-commerce adoption not only boosts local economies but also has an impact on the national and international markets. As shown by Ariansyah et al. (2021), e-commerce adoption in Indonesia has been influenced by multiple factors (both drivers and barriers), which can offer valuable lessons for other developing countries. In an increasingly digitized world, understanding the determinants of e-commerce adoption in diverse contexts—such as the Municipality of Andes—not only contributes to the advancement of academic knowledge but also provides relevant information for the formulation of public policies and business strategies that promote economic growth and competitiveness.

In Colombia, few studies have investigated e-commerce and its level of acceptance as a technology. Nevertheless, it has been established that, on a national level, several demographic variables have a strong moderating influence on online shopping (Sánchez-Torres et al., 2017). On a departmental level, in Antioquia, most research has been focused on modeling e-commerce acceptance in Medellín, a large capital city (Tavera et al., 2011). However, no study thus far has established the factors involved in this phenomenon in the southwestern rural region of Antioquia or its most populous municipality, i.e. Andes (Departamento Administrativo de Planeación - Dirección Sistema de Indicadores de la Gobernación de Antioquia, 2016).

The Municipality of Andes (46,621 inhabitants) covers an area of 449 km² (Departamento Administrativo de Planeación - Dirección Sistema de Indicadores de la Gobernación de Antioquia, 2016). Its main economic activities are related to commerce and the agricultural industry, which is why it has been consolidated as a hub for development in both the region and the department (Cámara de Comercio de Medellín para Antioquia, 2015). Currently, Andes has an internet penetration rate of 5.30%, a percentage that has grown steadily since 2012 according to official information published by the Colombian Ministry of Information and Communication Technologies (MINTIC-SIUST 2015). This rate is significantly lower than the average in the department (17.2%) and the country (32.5%), as reported by the same ministry.

This penetration rate is also low considering that the internet facilitates communication, e-commerce as an effective mechanism to exchange of goods and services, and the development of the region and the country. Therefore, this study investigates the factors that drive individuals to adopt e-commerce in such precarious circumstances in order to understand the theoretical and practical background of their decision. This article examines new managerial realities that require novel and better commercial strategies.

The Municipality of Andes, in the southwest of Antioquia (Colombia), has become a good environment for economic and commercial development. Nevertheless, despite the growing penetration of the internet in this region, the factors that determine the adoption of e-commerce by its residents have not been researched. This lack of comprehension of the key elements driving technological adoption in the local context poses a fundamental problem for its sustainable economic development and the competitiveness of its businesses (Hendricks & Mwapwele, 2023). As e-commerce has become a vital channel for commercial interaction and service provision in the digital era, it is essential to understand how its residents perceive and adopt this technology (Akour et al., 2022).

Although previous research has examined the factors influencing e-commerce adoption in different contexts, such as in developing countries (Hendricks & Mwapwele, 2023), no study so far has addressed this issue in the Municipality of Andes. In the literature, several authors have identified variables that can influence e-commerce adoption, such as perceived usefulness, ease of use and attitude towards use (Akour et al., 2022; Amofah & Chai, 2022). However, these variables should be put into context in a local environment to understand how they apply to a particular case, e.g. Andes (Orji et al., 2022; Fonseka et al., 2022).

Following the scheme proposed in previous studies in different contexts, this article aims to establish the validity of the Technology Acceptance Model (TAM) as a mechanism to determine the set of variables and relationships that define the use of e-commerce among the inhabitants of The Municipality of Andes. The results obtained will make a significant contribution to the organizational improvement of companies in the area by offering them an up-to-date and well-defined overview of the commercial realities of their market.

This study on the adoption of e-commerce in Andes (Colombia) is a significant contribution to the existing literature because it focuses on a specific geographical and economic context that has been underexplored in previous research. Ocloo et al. (2020) investigated the adoption of e-commerce at small and medium-sized manufacturing enterprises in Ghana, and Su et al. (2021) examined farmers’ participation in the digital financial market in rural areas in China. In contrast, this article delves into the adoption of e-commerce in a specific municipality in Colombia, where socioeconomic and technological conditions may be different and challenging. This localized perspective provides a more comprehensive understanding of the factors influencing e-commerce adoption in diverse contexts and expands the knowledge of e-commerce by considering a less explored environment. Furthermore, by applying the Technology Acceptance Model (TAM) in this context, it contributes to the validation and adaptation of said model in different settings, thus enriching the theoretical foundation for future research in the fields of technology adoption and e-commerce.

1.1. Theoretical framework

1.1.1. Relevant implications in the study of E-commerce acceptance in specific contexts

E-commerce has been shown to reduce the income gap between urban and rural areas. In China, recent research on this topic, such as Wang’s (2023) paper, suggest that the government should promote the development of rural e-commerce through specific policies and measures. Said paper emphasizes the importance of technological innovation as a mediator in this process and claims that, to reduce this gap, it is necessary to improve the infrastructure, promote technological innovation, and enhance the business environment in rural areas.

Al-Adwan and Yaseen (2023) suggest that, to reduce customer uncertainty and attract more buyers, e-commerce firms can do several things, for example, implement flexible return policies, supply high-quality information, encourage positive feedback, improve seller’s reputation, and increase their popularity. These authors also emphasize the relevance of customer service and effective management of customer feedback and reviews in e-commerce, as these elements can significantly impact customers’ perception of sellers and products.

To analyze consumer behavior in e-commerce, cultural and contextual differences should be considered. What may generate trust and encourage purchase intentions on one platform may not apply in the same way on another. This is important for both academics and marketing and e-commerce professionals because it means that their strategies should be adapted to a specific culture and context (Han & Han, 2023). Thus—in a specific rural context in Colombia where access to the Internet is limited—a greater adoption of e-commerce should depend on improving the perceived usefulness of this technology and custom-made strategies.

In specific contexts and from a practical and managerial perspective, online businesses should consider some important implications of e-commerce, including social aspects in terms of service quality, customer trust, and satisfaction. Additionally, it has been suggested that reputation, security, responsiveness, and reliability are key factors that online businesses must address to foster customer loyalty (Al-Adwan & Al-Horani, 2019). All these aspects are fundamental to promote customer loyalty and satisfaction in e-commerce, and business strategies should be adapted to the unique characteristics of each environment.

Taking into account that e-service quality plays a crucial role in stimulating customer engagement behavior in the context of community e-commerce, Fan et al. (2022) demonstrated that improving e-service quality encompasses aspects such as system design, intelligent fulfillment, security assurance, and interactive service. These aspects have a significant positive impact on stimulating customer engagement. In particular, customer trust and risk perception are key factors for this engagement behavior. All these studies demonstrate that it is necessary to deepen the understanding of how specific factors can influence user behavior in different contexts and can inform the development of appropriate policies and strategies to drive local economic growth and digital transformation in different areas.

1.1.2. Technology acceptance model (TAM)

The Technology Acceptance Model (TAM) is, without a doubt, one of the most widely used mechanisms to explain the set of behavioral intentions associated with the adoption of technologies by individual users (Wu, 2009). It was outlined by Davis in 1989 (Davis, 1989) as the evolution of the Theory of Reasoned Action (TRA) developed by Fishbiein and Ajzen, which offered an explanation for the relationship between individuals’ intentions and behaviors (Fishbein & Ajzen, 1975; Ajzen & Fishbein, 1977). The TAM introduced Perceived Ease of Use and Perceived Usefulness as the main factors involved in the process of adopting a certain technology, influencing individuals’ Attitude toward its use (Davis et al., 1989; Davis, 1989) and, subsequently, their Behavioral Intention to use it (King & He, 2006).

The concepts of Attitude and Behavioral Intention to use stem from the TRA—where the former is defined as a learned predisposition that generates consistent responses to a given stimulus; and the latter, as the intention to show a certain behavior (Zhao et al., 2016).

Perceived Usefulness refers to users’ perception of improvement when they compare technologies to previous options (Awa et al., 2015). According to the TAM (Davis et al., 1989), it precedes Attitude and Behavioral Intention. In turn, Perceived Ease of Use is defined as users’ expectation of reduced effort when they use a new technology (Li et al., 2012). Therefore, in the TAM, it precedes Perceived Usefulness and Attitude (Davis et al., 1989).

This literature review shows that the TAM can explain technology adoption behaviors in different market segments and multiple markets (Yoon, 2009; Hong & Zhu, 2006; Chen & Tan, 2004; Molla & Licker, 2014; Liang & Huang 1998; Jia et al., 2018; Slade et al., 2015)—including e-commerce (Jamshidi & Hussin, 2016; Zhao et al., 2016; Basak et al., 2016; Awa et al., 2015; Li et al., 2012; Benamati et al., 2010; Qiu & Li, 2008; Shi, 2013). In Colombia, Tavera and Londoño (2014) examined e-commerce acceptance in Medellín. However, no study thus far has implemented the TAM to investigate e-commerce acceptance in the Municipality of Andes (Antioquia, Colombia).

1.1.3. Hypotheses (proposed model)

Based on the above, the TAM should be able to explain the adoption of e-commerce in the context examined in this study. Therefore, adopting the structure proposed in the TAM, the three following hypotheses were formulated:

H1: The Perceived Ease of Use of e-commerce has a positive influence on individuals’ Attitude toward it.

H2: The Perceived Usefulness of e-commerce has a positive influence on individuals’ Attitude toward it.

H3: Individuals’ Attitude toward e-commerce has a positive influence on their Behavioral Intention to use it.

Figure 1 is a graphic representation of the relationships between the three hypotheses:

Figure 1. Conceptual model and hypotheses represented in SmatPLS software.

Source: Adapted from Davis (1989).

The figure depicts a conceptual model and accompanying hypotheses related to the subject matter. The model is analyzed using the SmatPLS software, a statistical tool commonly used for structural equation modeling.

2. Materials and methods

This study adopted a cross-sectional descriptive approach to explain the phenomena associated with the hypotheses. The measurement instrument was a self-administered questionnaire. The form was shared with a sample of 400 individuals over 14 years old who lived in the Municipality of Andes and claimed to be internet users. These characteristics are highly relevant for understanding e-commerce adoption because these individuals are likely to be more familiar with modern technology. Convenience sampling was carried out using quotas, which proved suitable because it was easy to contact the target population, data collection was efficient, and the selected population was relevant. These factors were in line with the research objectives of this study, as discussed in previous papers on to this topic (Chen & Dubinsky, 2003; V Venkatesh & Davis, 2000; Fayad & Paper, 2015; Elahi & Hassanzadeh, 2009; B. J. Corbitt et al., 2003; Palvia, 2009; Bhattacherjee, 2000; Pavlou, 2003; Gefen et al., 2003; Chismar & Wiley-Patton, 2002). Table 1 details the characteristics of the study sample.

Table 1. Characteristics of the study sample.

Sampling unit	Internet users older than 14
Geographical area of the study	Urban area of the municipality of Andes (Antioquia, Colombia)
Data collection method	Self-administered questionnaire
Sampling procedure	Convenience sampling using quotas
Sample size	400
Date of the field work	September 2017

Source: Own work.

In demographic terms, the surveyed population is mostly made up of people younger than 35 years of age (83%); who are married (58%); with technical, associate, or bachelor’s (51%) degrees; employees or independent workers (39%); and with an income equal to or above the country’s minimum legal wage (36%). This high proportion of young and well-educated participants in the sample benefits this study because their characteristics are useful for research into online behavior (Hong & Zhu, 2006a; Mohapatra & Sahu, 2018; Lim et al., 2018; Mamonov & Benbunan-Fich, 2017; Mohtaramzadeh et al., 2018; Sohaib & Naderpour, 2017; Yoon, 2009).

The questionnaire used in this study is composed of a series of statements derived from the literature review and the support of experts who contributed to the content validity of each variable involved. Participants rated each statement (called item) on a 5-point Likert scale where (1) meant “strongly disagree”; and (5), “strongly agree.” Table 2 shows the English translations of the items for illustrative purposes; however, they were originally presented to participants in Spanish.

Table 2. Constructs and items in the self-administered questionnaire.

Construct	Code	Item	Source
Perceived Ease of Use	EU1	Learning how to use the internet to make purchases/transactions is easy for me.	Davis (1989); Moore and Benbasat (1991)
	EU2	Overall, I think that making online purchases/transactions is easy.
	EU3	It would be easy for me to acquire the skill to make online purchases/transactions.
	EU4	I don’t need to make a mental effort to make online purchases/transactions.
	EU5	I find it easy to make online purchases/transactions the way I want to.
	EU6	My interaction with the internet when I make online purchases/transactions is clear and understandable.
Perceived Usefulness	U1	It is faster to make purchases/transactions online than at a brick-and-mortar store.	Davis (1989); Moore and Benbasat (1991)
	U2	Online purchases/transactions save me time.
	U3	Online purchases/transactions have benefits.
	U4	Overall, online purchases/transactions are very useful.
	U5	It is more efficient to make purchases/transactions online than in person.
	U6	It is easier to make purchases/transactions online than in person.
Attitude	A1	I like the idea of making online purchases/transactions.	Taylor and Todd (1995)
	A2	Making online purchases/transactions is a great idea.
	A3	Making online purchases/transactions is a good idea.
	A4	Making online purchases/transactions is a positive idea.
Behavioral Intention	BI1	I have the intention to make online purchases/transactions.	Taylor and Todd (1995); Coyle and Thorson (2001); Kim and Biocca (2006)
	BI2	I think I will make online purchases/transactions in the next few months.
	BI3	I would recommend online purchases/transactions to others.
	BI4	I would use my credit/debit cards to make online purchases/transactions.
	BI5	I am going to use websites to make online purchases/transactions frequently.

Source: Own work.

3. Results

Data analysis was conducted employing the bootstrapping methodology and the sample of 400 questionnaire responses that were collected during the field work. Then, the data were made more pre-processed to run the model, and, subsequently, SPSS 16 and SmartPLS 3.0 were used to perform confirmatory factor analysis and structural equation modeling as proposed in Figure 1. These analyses were carried out to determine the validity and reliability of the model.

The results in Table 3 show that the measurement scales are reliable, that is, the items reflect the existence of the constructs. Strong relationships were found between each one of the four constructs and its respective variables. Additionally, there were weak relationships between each variable and the error. In all the cases, the loadings were above 0.6; and the p values, below 0.05. Therefore, it can be inferred that the model and its variables are consistent with the constructs they represent.

Table 3. External loadings. Own work.

Construct	Code	Original sample	Sample mean	Standard error	t-statistic	p value
Perceived Ease of Use	EU1	0.769	0.767	0.027	27.978	0.000
	EU2	0.842	0.842	0.018	45.833	0.000
	EU3	0.852	0.853	0.018	47.081	0.000
	EU4	0.751	0.753	0.035	21.460	0.000
	EU5	0.821	0.823	0.020	41.418	0.000
	EU6	0.829	0.830	0.020	42.109	0.000
Perceived Usefulness	U1	0.745	0.747	0.030	24.859	0.000
	U2	0.808	0.809	0.021	38.988	0.000
	U3	0.810	0.811	0.021	38.052	0.000
	U4	0.808	0.808	0.025	31.706	0.000
	U5	0.809	0.807	0.024	34.373	0.000
	U6	0.838	0.839	0.017	50.757	0.000
Attitude	A1	0.855	0.856	0.016	52.394	0.000
	A2	0.908	0.908	0.011	80.776	0.000
	A3	0.921	0.920	0.010	88.875	0.000
	A4	0.892	0.891	0.013	67.426	0.000
Behavioral Intention	BI1	0.908	0.908	0.011	80.776	0.000
	BI2	0.921	0.920	0.010	88.875	0.000
	BI3	0.892	0.891	0.013	67.426	0.000
	BI4	0.908	0.908	0.011	80.776	0.000
	BI5	0.921	0.920	0.010	88.875	0.000

Source: Own work.

According to Hair et al. (2014), in studies carried out with SmartPLS software, it is necessary to follow two successive and coordinated model validation stages: (1) the measurement model, which assesses the internal consistency, reliability, and convergent and divergent validity of the constructs with respect to their items; and (2) the structural model, which assesses the determination coefficients and the significance of the structural relationships between constructs (Table 4).

Table 4. Stages and minimum criteria employed in the validation of the proposed model.

Validation stage	Indicators	Minimum criteria
Measurement model	Internal consistency: Cronbach’s Alph (CA)	= >0.7 (Nunnally & Bernstein, 1994)
	Reliability: Composite Reliability (CR)	= >0.7 (Fornell & Larcker, 1981)
	Convergent validity: Average Variance Extracted (AVE)	= >0.5 (Fornell & Larcker, 1981)
	Convergent validity: loadings’ size and significance	= >0.6 (Bagozzi & Yi, 1988) and p < 0.001; p < 0.05; p < 0.01
	Discriminant validity: Heterotrait-Monotrait (HT/MT) ratio of correlations	= < 0.9 (Helgesen, 2008)
Structural model	Coefficient of determination (R^2 value)	= > 0.75 and = > 0.51, substantial; = < 0.50 and = > 0.26, moderate; and < 0.25, weak (Hair et al., 2014)
	Significance of the structural relationships	p < 0.001

Source: Own work.

Each statistical test was conducted according to the information above. Table 5 reports the calculated internal consistency, reliability, and convergent validity of the constructs in the model. Additionally, the loadings with p < 0.001 are marked with ***.

Table 5. Reliability and convergent validity of the constructs in the proposed model.

Construct	Items	Loadings	CA	CR	AVE
Perceived Ease of Use	EU1	0.769***	0.896	0.920	0.658
	EU2	0.842***
	EU3	0.852***
	EU4	0.751***
	EU5	0.821***
	EU6	0.829***
Perceived Usefulness	U1	0.745***	0.890	0.916	0.646
	U2	0.808***
	U3	0.810***
	U4	0.808***
	U5	0.809***
	U6	0.838***
Attitude	A1	0.855***	0.916	0.941	0.799
	A2	0.908***
	A3	0.921***
	A4	0.892***
	A5	0.908***
Behavioral Intention	BI1	0.921***	0.923	0.942	0.763
	BI2	0.892***
	BI3	0.908***
	BI4	0.921***
	BI5	0.769***

Source: Own work.

To complete the analyses conducted previously, the next step was to evaluate the discriminant validity of the proposed model. The results of this step are reported in Table 6, which shows the square of the AVE on the diagonal (Chin, 2014) calculated using SmartPLS. The results obtained, all under 0.9, represent a successful completion of the comprehensive validation of the model.

Table 6. Discriminant validity of the constructs.

	EU	U	A	BI
Perceived Ease of Use	0.811
Perceived Usefulness	0.697	0.804
Attitude	0.560	0.703	0.894
Behavioral Intention	0.584	0.560	0.528	0.874

Source: Own work.

The coefficient of determination (R²) was used to evaluate the predictive ability of the structural model. This coefficient indicates which latent variables are influential and what part of the variance of the dependent variables is explained (Aldás-Manzano, 2016). The results yielded by SmartPLS show that both dependent constructs, Attitude (R2 = 0.504) and Behavioral Intention (R2 = 0.279), are moderate indicators.

After having established the predictive ability of the proposed model, the hypotheses were tested using the bootstrapping function in SmartPLS. In Table 7, all the hypotheses in the model were supported due to the significance of the trajectory coefficients that were obtained. According to these results, Perceived Ease of Use has a positive effect on Attitude (H1: β = 0.135); Perceived Usefulness, on Attitude (H2: β = 0.609); and Attitude, on Behavioral Intention (H3: β = 0.528). In particular, Perceived Usefulness was found to be the most important construct to drive Behavioral Intention.

Table 7. Hypotheses testing results.

Hypothesis	Description	Trajectory coefficient	t value	Result
H1	Perceived Ease of Use → Attitude	0.135*	1.984	Supported
H2	Perceived Usefulness → Attitude	0.609***	9.311	Supported
H3	Attitude → Behavioral Intention	0.528***	13.804	Supported

Source: Own work.

Figure 2 details the results of the entire structural model, graphically showing the relationships obtained in the process.

Figure 2. Structural model results.

Source: Own work.

The figure presents the outcomes of the structural model analysis. It illustrates the relationships between variables and their corresponding statistical significance, demonstrating the findings of the study or analysis.

In general, the results obtained validate the use of the TAM to model the phenomenon of e-commerce acceptance in the Municipality of Andes—shedding light on the factors involved in said phenomenon and opening up the way for future studies that enable a better understanding of the role of users in the digital transformation of companies in this region.

4. Discussion and conclusions

Considering the aim of this study and the results obtained, it is possible to claim that the TAM is a good predictor of behaviors associated with e-commerce acceptance in the southwestern region of Antioquia, in particular in the Municipality of Andes. This region offers important business growth opportunities in terms of infrastructure, social and economic development, entrepreneurship, and public policies.

Beyond the scientific nature of this analysis, the technical implications of this research for the companies in said region are significant. In this study, the Perceived Usefulness of e-commerce was consolidated as the main antecedent of the Attitude towards its use, which represents an opportunity for new platforms, channels, and commercial offers that create value through direct interactions with users.

However, the municipal government should make public investments in the diffusion of e-commerce and similar technologies. Otherwise, the popularization of e-commerce in said municipality will be restricted by several factors: limited internet access, lack of trust in the security of transactions, and required technical support.

4.1. Practical implications

The findings of this study support the use of TAM as an effective tool for understanding the adoption of e-commerce in the region. This means that businesses and local authorities can use this model as a guide to promote and facilitate the adoption of e-commerce technologies in the area. This may include training programs to enhance the Perceived Ease of Use and Perceived Usefulness among potential users.

One of the key findings is that Perceived Usefulness is a crucial factor in shaping attitudes toward the use of e-commerce. Companies operating in this region should focus on demonstrating that e-commerce can provide real benefits to consumers, such as saving time and shopping efficiently. These can be highlighted in marketing strategies and personalized offers to attract more customers.

Given that Attitude toward e-commerce is positively influenced by Perceived Usefulness, companies can explore opportunities to develop new platforms and business offerings that enhance user experience. This may include user-friendly mobile applications and intuitive websites that facilitate online shopping. Additionally, personalized offers based on user preferences can increase the likelihood of customer conversion.

To ensure the continued success of e-commerce in the region, local authorities and businesses should invest in technological infrastructure. This involves improving internet connectivity, ensuring the security of online transactions, and providing adequate technical support. Without a robust infrastructure, the adoption of e-commerce could be hindered—overcoming this challenge is crucial.

To effectively drive e-commerce adoption in the southwest of Antioquia, local businesses and government authorities should work together. This could involve tax breaks for businesses that promote e-commerce, as well as active participation in training the local workforce in relevant technology skills. Together, they can harness the full potential of e-commerce in the region and contribute to its economic and social growth.

This study highlights the effectiveness of the TAM as a valuable tool to understand e-commerce adoption in the southwest of Antioquia. These results could inspire future researchers to delve deeper into the application of theoretical technology adoption models in diverse geographical and cultural contexts, thus contributing to a broader understanding of how users adopt and use new technologies. Moreover, it was found that Perceived Usefulness is a key predictor of user behavior, which suggests that further research is necessary to establish the factors influencing Perceived Usefulness and how communication and marketing strategies can enhance this perception among consumers.

In the educational realm, the implications of this study are equally significant. Educational institutions in the southwest of Antioquia can use these findings to develop training and educational programs that focus on the skills that are needed to use e-commerce effectively. This could include designing courses and workshops that address topics such as the ease of use of technologies and how to maximize the benefits of online commerce. These educational programs could be beneficial for both students and professionals looking to improve their digital skills. Furthermore, they could be specifically tailored to meet the region’s needs, providing locals with the necessary tools to successfully participate in the ever-evolving digital economy.

4.2. Limitations

This study presents several limitations that should be considered when its results are interpreted. First, convenience sampling was used to select a sample of 400 individuals over 14 years old in the Municipality of Andes. While this sample provided valuable data, its composition could limit the generalizability of the results to a broader population. Most participants were young individuals (under 35 years old) with relatively high educational levels, which may not fully represent the demographic diversity of their region. Future studies could obtain more representative samples to achieve a more comprehensive understanding of e-commerce adoption in the area.

Another limitation can be the use of self-administered questionnaires to measure psychometric variables. Although the questionnaire responses provided valuable information about participants’ perceptions and attitudes, the instrument may have limited the consideration of broader contextual factors influencing e-commerce adoption in the region. External factors, such as access to technological infrastructure or government policies, were not explored in depth and could be relevant to fully understand the adoption of this technology.

Furthermore, as this was a cross-sectional study (i.e. the data were collected at a specific point in time, 2017), it cannot capture the dynamics of e-commerce adoption over time and its changing trends in the region. Since users’ attitudes and behaviors can evolve gradually, future research may adopt longitudinal approaches to paint a more comprehensive and dynamic picture of technology adoption in the region.

A significant limitation was that cultural factors that could influence e-commerce adoption in the region were not explored. This study primarily focused on psychometric aspects and did not delve into cultural influences or social norms that may play a role in technology acceptance. Future research could address this gap and analyze cultural aspects in more detail to gain a more complete understanding of technology adoption in the southwest of Antioquia.

This study paves the way for future research that combines variables and constructs from different literature streams to establish potential relationships determining e-commerce adoption. The approach adopted in this work also has a limitation, i.e. the empirical measurement of Behavioral Intention as a predictor of effective use. Another constraint was the internet penetration rate in the region, which is remarkably low.

Nevertheless, this study makes a contribution to the understanding of e-commerce adoption in the Municipality of Andes in the southwest of Antioquia, Colombia. The results suggest that the TAM is an effective tool for explaining technological adoption behavior in this region. The Perceived Usefulness of e-commerce was identified as the most influential factor in forming positive attitudes toward this technology and the intention to use it. This finding highlights the importance of users’ perception of how a technology can benefit them in the specific context of their region.

In conclusion, this study provides a solid foundation for understanding e-commerce adoption in the southwest of Antioquia. Despite its limitations, its results underscore the importance of Perceived Usefulness in shaping technology adoption attitudes and behaviors. To drive e-commerce adoption in the region, businesses and educational institutions should focus on improving the Perceived Usefulness of this technology and adapting their strategies to the specific needs and characteristics of the local population.

Acknowledgments

We thank Fundación Universitaria CEIPA and Instituto Tecnológico Metropolitano for funding and constantly supporting the research project that resulted in this research paper.

Disclosure statement

The authors declare no conflict of interest.

Correction Statement

Present affiliation of Silvana Correa-Henao: Faculty of Economic Sciences, Universidad de Antioquia, Medellin, Colombia

Additional information

Notes on contributors

Jorge Andrés Vélez-Muñoz

Jorge Andrés Vélez-Muñoz Industrial Engineer, MSC in International Strategic Marketing, and Master in Business Administration. Holds complementary studies in Internationalization of Higher Education, marketing, and digital transformation. He has professional experience in Marketing Management, Digital Strategy Development, Market Intelligence, process improvement, and implementation of quality management systems. Additionally, serves as an advisor to Micro, Small, and Medium Enterprises on Comprehensive Strategic Marketing. With more than a decade in the education sector, has served as a faculty member for undergraduate and postgraduate programs, researcher in the field of e-commerce, communications leader, Director of Internationalization, and Academic Coordinator of the MBA program at the CEIPA University Foundation.

Sebastián Franco-Castaño

Sebastián Franco-Castaño is a versatile professional with a Master’s degree in Administrative Engineering from the National University of Colombia, complementing his Bachelor’s degree in the same field. He currently serves as a Link Teacher in the Master’s program in Organization Management and has been an Occasional Teacher at the Metropolitan Technological Institute. Previously, he worked as an Instructor at the Center for Innovation in Agribusiness and Regional Aviation in Antioquia. Sebastian’s expertise spans various areas including Market Research, Consumer Behavior, Marketing Management, and Change Management. He has contributed significantly to academic literature with articles and a book chapter, and has also created training programs, showcasing his commitment to academia and professional development.

Silvana Correa-Henao

Silvana Correa-Henao Business Administrator and Master of Science, Technology and Innovation Management. Her research interests include electronic commerce, strategic marketing, sports management, consumer behavior, knowledge transfer, and innovation, researcher at Imark, Marketing Group at the University of Antioquia, that focuses on consumer behavior and technological acceptance, which holds significant importance in marketing for formulating strategies aimed at introducing innovative technologies to specific target market, particularly in the context of emerging regions or countries.

Alejandro Valencia-Arias

Alejandro Valencia-Arias obtained his PhD in Management Engineering from the National University of Colombia in 2018, a Master of Sciences degree in Computer Sciences in 2013, and a Bs. Eng degree in Management Engineering in 2010. With 12 years of experience as a university professor, he has made significant contributions to his field. His academic achievements include the publication of books and over 85 journal articles in national and international indexed journals. He holds the prestigious title of Senior Researcher from the Ministry of Science, Technology, and Innovation in Colombia and is recognized as a distinguished researcher at RENACYT (Peru). His research encompasses entrepreneurship, simulation, marketing research, and statistical science. Proficient in agent-based modeling and system dynamics, he specializes in developing social models. His current research focus involves analyzing teachers’ perspectives on the use of virtual tools in emerging economies, a topic that greatly motivates the professor.