Investigating the E-Readiness of Informal Sector Operators to Utilize Web Technology Portal

Abstract

Information and Communication Technology (ICT) has been acknowledged to be an enabler of small businesses, including those in the informal sector. However, determining the relationship between the perception of technology by informal service providers and the readiness to use technology is critical. By adopting a survey research design, this study investigates how the perception of a web technology portal (WTP) by informal sector service providers in the Cape Town metropolitan area in South Africa affects their readiness to use WTP to support their businesses. The study involved a sampled population of 419 informal sector service providers within the Cape Town metropolis. A conceptual framework consisting of constructs from the self-efficacy theory (SET), the unified theory of acceptance and use of technology (UTAUT) and the technology readiness index (TRI) was used as the theoretical reference for the study. We used a semi-structured questionnaire based on a five-point Likert scale to collect data, which we analyzed using partial least squares structural equation modelling (PLS-SEM). The results showed that discomfort (p = 0.330), effort expectancy (p = 0.630), innovativeness (p = 0.620), optimism (p = 0.740), insecurity (p = 0.110), facilitating conditions (p = 0.160), and internal factors (p = 0.180) all had a non-significant positive influence on e-readiness. On the other hand, self-efficacy (p < 0.001), performance expectancy (p < 0.030), social influence (p < 0.001), and external factors (p < 0.001) had a significant positive influence on e-readiness. Gender and business type, the moderating variables for respondents’ e-readiness, were found to be insignificant (p > 0.005) for e-readiness for web portal technology utilization. This study identifies the key variables that could influence the readiness of informal sector operators to utilize web portal technology. It also provides a guide for designers and developers of digital platforms and government policymakers on critical factors germane to providing technology support for the informal business sector.

1. Introduction

In a globalized world where access to information plays a critical role in the behavioural intent of individuals and organizations, the use of information and communication technology (ICT) tools has become a major enabler for big and small businesses. Therefore, e-readiness determines how any society or individual is strategically positioned to utilize the available opportunities when Information and Communication Technology (ICT) is leveraged [1]. In this context, technology readiness represents the favourable disposition to use innovation or new technologies [2], while technology adoption and acceptance represents an experience after technology use [2].

Several theoretical models and frameworks have been proposed to explain users’ adoption of innovation or new technologies. These include the technology acceptance model [3,4], theory of planned behaviour [5], diffusion of innovation theory [6], theory of reasoned action [7], model of PC utilization [8], motivational model [9], unified theory of acceptance and use of technology [10], social cognitive theory [11,12,13], and self-efficacy theory [12,13]. Recently the trend in the assessment of behavioural intention to use technology has been the use of pre-existing models and adding new constructs to increase their processing power for studying a phenomenon.

The consensus is that a combination of more Information System (IS) theories is required to improve the explanatory power of these models for a better understanding of the issues being investigated [14]. Several authors [15,16,17] have studied users’ adoption of new technologies. Still, acceptance factors generally play a critical and dominant role, including technology availability, convenience, users’ needs, and perceived built-in security [17]. The informal sector can leverage ICT adoption and utilization to drive their business objectives, which will alleviate poverty, better service delivery, competitiveness in the marketplace, and human capital development [18,19]. At the micro-level, ICT facilitates the rejuvenation of the environment, social networking, healthcare, production efficiencies and skills acquisition. However, to leverage new technologies, individuals, governments, societies, and organizations must be “e-ready” [20].

According to [1], e-readiness signifies the level to which a nation or entity is well-positioned to reap the full benefits of a digitalized world. From a UN perspective, “e-readiness” determines how any society or individual is strategically positioned to utilize the available opportunities when Information and Communication Technology (ICT) is leveraged. For effective policy formulation on the use of technology to support the informal sector and micro-business, a determination of the e-readiness of informal sector operators is essential.

The Research Problem

Compared to other developing economies, the percentage of participation in the informal sector in South Africa (SA) is only about 6% [21,22,23,24,25]. For reference, [24] posits that the informal sector of SA provides a source of livelihood, employment, and wages for 2.5 million South Africans.

The South African informal sector is plagued with numerous challenges, including low capitalization, absence of supporting technologies, poor infrastructure, lapses in security, external competitive pressures, zoning constraints, poor governmental policies, and lack of social protection [19,21,26,27,28,29,30]. Hence, micro-enterprises in SA cannot contribute meaningfully to the socio-economic advancement of South Africa [24]. Support through ICT has been identified as having the potential to solve some of the challenges of the SA informal sector [18,19,30]. In addressing some of these problems, [18,19] have proposed a technology-centric approach. However, the degree of e-readiness of the informal sector to leverage web technology portal utilization for improved visibility and profitability has yet to be discovered. Informal sector providers and customers need pertinent information on the potentials and opportunities available in the informal marketplace. The government needs an accurate picture of where the starting point should be for any attempt to design a viable digital platform to address the challenges of the informal sector providers, nor the level of sophistication required [31]. The consequences of these are that the status quo will remain in terms of the several problems of the informal sectors, such as lack of organization, lack of formalization, poor marketing visibility, low capitalization, low quality of service, and generally a low contribution to the national GDP. The level of e-readiness of the informal sector service providers to adopt web technology still needs to be discovered, which will make investments in the development of electronic portal technology solutions (digital platforms) that meet their needs to be difficult. Consequently, these constraints impeding the productivity and competitiveness of the SA informal sector service providers necessitated the development of the research question (RQ):

RQ 1: What is the impact of the beliefs and perceptions of technology by the South African informal sector service providers on their e-readiness to use a web technology portal?

Therefore, the rationale for this study was to determine the degree of e-readiness of the SA informal sector. None of the previous studies on technology adoption by informal sector or micro-enterprises in South Africa has examined their e-readiness to leverage web portal technology to advance their business objectives. This is an empirical gap that must filled in order to provide a basis for effective technology support for the SA informal sector.

2. Related Work

Several studies on IT adoption by small, medium, and micro enterprises (SMMEs) in South Africa have been reported in the literature.

In [32], the authors used a quantitative methodology to collect data from 300 respondents (owners/managers of SMEs tailored to the manufacturing and service sectors). The authors used Confirmatory Factor Analysis (CFA) and Covariant-based structural equation modelling with IBM SPSS AMOS vs. 22 (New York, NY, USA) for data analysis. Their findings showed that IT enhanced integration and collaborative operations that support customer services, inventory management, time management, task performance, and relationship-building. In [33], a low adoption rate among SMEs in South Africa for software as a service (SaaS) was observed despite its inherent benefits. By adopting the diffusion of innovation theory (DOI), the authors used a questionnaire instrument for data collection. Regression analysis was used to analyze the data. The study found that most SMEs needed to be aware of the potential of SaaS and would consider using it if it would help reduce costs.

Additionally, ref [34] explored factors that impacted the Bring Your Own Device (BYOD) concept by SMEs in South Africa. The qualitative study used the perceived e-readiness model (PERM) as the conceptual framework and adopted an interpretivist philosophical stance. The findings show that organizational e-readiness and environmental readiness are necessary to implement the BYOD concept successfully. The authors in [35] proposed a conceptual model that incorporates control indicators from the diffusion of innovation (DOI) theory, institutional theory, transactional cost theory, organization theory, information security theory and trust theories for the assessment of pertinent factors influencing SMEs in SA from adopting cloud computing services. The model was to be employed in future research.

Additionally, ref [36] employed the Publicise Interact and Transform (PIT) model in ICT adoption to study the barriers to integrating e-commerce by SME service providers engaged in tourism. It was a qualitative study that used thematic analysis for data analysis. The result showed that high costs, inadequate funding, and lack of technical skills impeded e-commerce adoption by SME service providers in tourism in Pretoria. However, it was also revealed that these SMEs experienced added benefits when these barriers were removed.

Authors in [37] observed that the high investment cost necessary for integrating ICT into businesses, and the absence of acquired skills for utilizing these tools, acted as an impediment to their use by SMEs. By applying a qualitative method, ref [38] observed that although ICT tools are acknowledged as drivers of the knowledge economy, SMEs in Gauteng still needed to adopt ICT tools fully. Therefore, they could not stay competitive locally, nationally, or globally. The study found that the SMEs in Gauteng were prone to adopting local technology in business operations. The authors identified limited funding, lack of skilled staff, and lack of tools as critical factors impeding ICT adoption. In [39], by comparing SMEs in SA to other advanced societies, the authors found that many SMEs were still interested in using local technologies and that the acceptance and adoption of e-commerce as drivers for the survivability of the SMEs were almost nonexistent.

Studies that focused on SMMEs include [40], who conveniently sampled 247 SMMEs quantitatively from two cities (Durban and Pietermaritzburg) in KwaZulu-Natal (KZN) to assess the relationship between inhibiting factors and e-commerce adoption. Results after descriptive statistics and Chi-square of independent test on the data from the two cities showed that factors including a low level of computerization, the exorbitant cost of computers and networking equipment, poor telecommunication services, internet security, legal problems, and liability and rigidity in contractual arrangements inhibited e-commerce adoption by SMMEs in KZN. In [41], the authors examined the factors that affect e-commerce adoption by SMMEs in Eastern Cape Province in South Africa using quantitative and qualitative data collection and analysis methods. The findings show that security, prestige, government support, vision, and the need to service niche markets were the key determinants of e-commerce adoption by SMMEs in the Eastern Cape province in South Africa. Based on the literature review, ref [42] developed a theoretical framework for e-commerce adoption by SMMEs. The theoretical framework was validated using a multiple case study strategy involving SMMEs in South Africa. The findings revealed ten e-commerce adoption factors for SMMEs. According to [43], introducing ICT into the daily operations of micro, small, and medium-scale enterprises (SMMEs) can help integrate them into the digital/knowledge economy. The authors also opined that ICT diffusion in SMMEs can lead to poverty reduction through income generation, more diverse opportunities in livelihood, employment for the poor, skills development and self-confidence, social protection, economic empowerment, and security against job losses.

Table 1. ICT diffusion in MSMEs in South Africa.

Study/ Year	Sector	Country	Focus
[34]	SMEs	Cape Town, WC, SA	BYOD conceptual framework in SMEs
[35]	SMEs	SA	Development of Conceptual framework for future research
[36]	SA Tourism service providers	Pretoria	E-commerce adoption
[33]	SA SMEs-IT decision-makers/and IT partners	Gauteng Province, SA	SaaS adoption in SA SMEs
[32]	SMEs	Gauteng and Free State	IT adoption in SMEs supply chain
[38]	SMEs	Gauteng, SA	ICT in businesses
[37]	SMEs	KZN, South Africa (SA)	e-commerce and rural development
[39]	SMEs	Western Cape	E-commerce adoption
[40]	SMMEs	KZN, SA	Inhibitors to the adoption of e-commerce
[41]	SMMEs	Eastern Cape, SA	e-commerce adoption
[42]	SMMEs	SA	e-commerce adoption
[43]	SMMEs	SA	ICT diffusion

summarises related work on ICT adoption by SMMEs in South Africa.

The findings from the literature reveal that none of the previous studies has examined the e-readiness of the informal sector (classified as Micro-enterprises) to leverage web portal technology to advance their business objectives. This study aims to fill this empirical gap.

3. Theoretical Framework and Hypotheses Development

A theoretical framework comprises concepts with appropriate explanations and references to important and relevant literature about theories embedded in the research. Additionally, it tries to limit the scope of the collected data to the study’s research questions (RQs). For example, the Technology Acceptance Model (TAM) [3,4] is the most widely employed theory for assessing the adoption of technology or innovations [44,45]. TAM tries to identify intrinsic and extrinsic drivers of the user’s motivation toward accepting and adopting technology. However, due to the parsimony of the TAM with only two constructs—Perceived Usefulness (PU) and Perceived Ease of Use (PEOU)—other models have been utilized to increase the processing power of the TAM model. Therefore, the conceptual framework for this study was composed primarily of three theories, which are the self-efficacy theory (SET), the unified theory of acceptance and use of technology (UTAUT), and the Technology Readiness Index (TRI), and key constructs extracted from the literature. An overview of the constructs used in the study and their theoretical sources is shown in

Table 2. Constructs used in this study.

Construct	Source	Description
Internal factors (INFs)	Review of the literature	Inherent factors which affect business operation
External factors (EXFs)	Review of the literature	Environmental/institutional factors which impede business activities
Self-efficacy (SE)	Self-efficacy Theory	Perceptions of an individual in his/her capabilities to attain the goal in job performance
Performance expectancy (PE)	UTAUT	The concept of a person being receptive to the idea that utilization of the system/technology will aid them in attaining the required job performance
Effort expectancy (EE)	UTAUT	Signifies the degree of ease that comes with the acceptance and adoption of a web technology portal
Social Influence (SI)	UTAUT	Signifies the degree/level an individual attaches to the importance others believe they should utilize the web technology portal
Facilitating conditions (FC)	UTAUT	Signifies consumers’ perceptions of the availability of resources (technical, institutional/governmental support) available for the use of technology
Optimism (OPM)	TRI	A positive opinion about technology and the perception that it enables people to be in control, efficient, and flexible in maintaining a productive life.
Innovativeness (INNO)	TRI	The tendency to be a technology pioneer and a thought leader among peers.
Discomfort (DISC)	TRI	The absence of discipline over technology and a feeling that one is overwhelmed by it
Insecurity (INSEC)	TRI	The skepticism about technology due to a lack of trust
Web Technology Portal (WTP)	Review of the literature	The use of a web technology portal

.

3.1. Self-Efficacy Theory (SET)

Self-efficacy emanated from the social cognitive theory (SCT) [46]. It represents the degree of certainty of an individual’s capabilities to perform a given task appropriately and successfully. Generally, SCT signifies the confidence of individuals in their skills and capabilities through the exertion of effort in performing a given task and how they would persist long-term in overcoming any hurdle better than those with less confidence [47]. On the premise of self-efficacy, ref [48] opined that computer self-efficacy is the certainty of an individual’s abilities to perform a computing-related task successfully. Hence, self-efficacy as a construct may assess an individual’s confidence in using technology or innovation, and it acts as an important component affecting high innovation adoption [49]. Researchers in the IS domain have concentrated on trying to understand the interrelation between computer self-efficacy and several related computer tasks [50,51,52]. Therefore, there is a need for more studies on the use of self-efficacy theory (SET) to assess the e-readiness of the informal sector practitioners to utilize web technology. Beliefs encompassing self-efficacy are driven by four pertinent information constructs [11,46,53,54]: Performance experience, which measures a better perspective on the present task to be accomplished due to prior success at something like the new behaviour; Vicarious experience, which involves learning by observing activities performed by someone else who is an associate/or colleague-become proficient; Social persuasion, encouragement others provide as an incentive to perform the task and Physiological/emotional states, The states (physical and emotional) caused by thinking about the requirement for executing the new behaviour.

Perceived usefulness (PU) can be predicted by computer self-efficacy because perceptions of being highly capable of using technology allows the inference the inherent ability of the SA informal sector operators to use a web technology portal for increased visibility, productivity, efficiency, and competitiveness in their business operations. Hence, we propose this hypothesis:

Hypothesis 1 (H1).

Self-efficacy (SE) has a positive effect on the use of web technology portal (WTP).

3.2. The Unified Theory of Acceptance and Use of Technology (UTAUT)

The UTAUT [10] resulted from a combination of models (technology acceptance model (TAM)4], theory of planned behaviour (TPB) [55], theory of reason action (TRA) [7], motivational model (MM) [9], diffusion innovation theory (DIT) [6], model of PC utilization (MPUC) [8], social cognitive theory (SCT) [46]). The summation of the constructs from these eight models resulted in four determinants used in predicting behavioral intention and use [10]. The UTAUT model is a combination of four exogenous constructs (performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating conditions (FC)) and two endogenous variables (user’s intention and user’s behavior). It is moderated by age, gender, experience, and voluntariness. Further addition of factors and influences on the original TAM model facilitated the expansion of the original TAM to UTAUT, which opines that an individual’s acceptance of technology and use depended on social and societal influences or perspectives. According to [10], the four main constructs of UTAUT are: Performance expectancy (PE), the concept of a person being receptive to the idea that utilization of the system/technology (represented here as web technology portal) will aid to attain the required job performance; Effort expectancy (EE), meaning the degree of ease that comes from using the system (web technology portal); Social influence (SI), signifying the degree a person places on the important beliefs others have on the rationale and need for using the innovation/technology (WTP); and Facilitating conditions (FC), the extent that an individual believes that there is an organizational and technical infrastructure to support using the new technology (web technology portal).

Thus, we used the four constructs in the unified theory of acceptance and use of technology (UTAUT) to propose the following hypotheses with respect to the SA informal sector practitioners:

Hypothesis 2 (H2).

Performance expectancy (PE) has a positive influence on the use of web technology portal (WTP);

Hypothesis 3 (H3).

Effort expectancy (EF) has a positive effect on the use of web technology portal (WTP);

Hypothesis 4 (H4).

Social influence (SI) has a positive influence on the use of a web technology portal (WTP); and

Hypothesis 5 (H5).

Facilitating conditions (FC) has a positive influence on the use of a web technology portal (WTP).

3.3. Technology Readiness Index (TRI)

The increasing diffusion of innovation for societal use has supported the premise that individual differences are observed in adopting and using technologies [56,57]. Other authors, including [58,59], have articulated factors such as ease of use and usefulness (regarding technology) and socio-demographics [60,61] as responsible for these differences. Recently, human trait as a variable to technology use has been given significant attention. According to [62,63,64], this trait is called technology readiness (TRI).

Similarly, refs [57,65] opined that TRI represents the propensity to accept and use newer innovations for goal accomplishment in home life or at work. The constructs have four dimensions (innovativeness, optimism, insecurity, and discomfort). The constructs from TRI, according to [57,65], are: Optimism (OPM), which signifies one’s positive opinion on technology and the perception that it establishes control, efficiency, and flexibility in one’s life leading to a more productive life; Innovativeness (INNO), meaning the tendency/or one’s ability to act as a technology pioneer and be a thought leader among peers or colleagues [65]; Discomfort (DISC), meaning the perception of a lack of control over technology, and a feeling of being overwhelmed by technology [65]; and Insecurity (INSEC), signifying one’s distrust of technology due to skepticism about its potential to work accordingly as well as having concerns from harm accruing from the use of technology [65]. The hypotheses that stem from TRI for SA informal sector providers are:

Hypothesis 6 (H6).

Optimism (OPM) influences users’ readiness to use web technology portal (WTP);

Hypothesis 7 (H7).

Innovativeness (INNO) influences users’ readiness to use a web technology portal (WTP);

Hypothesis 8 (H8).

Discomfort (DISC) significantly affects the user’s readiness to use a web technology portal (WTP); and

Hypothesis 9 (H9).

Insecurity (INSEC) significantly affects users’ readiness to use a web technology portal (WTP).

In summary, according to [65], motivators (innovativeness and optimism) are positive drivers necessary in improving one’s technology readiness (TR), while inhibitors (discomfort/insecurity) act negatively to decrease one’s technology readiness (TR).

3.4. The Effect of Internal Factors and External Factors on e-readiness of the SA Informal Sector

The contributory role of the informal sector to economic activities is widely known because it acts as a “conduit of employment” in most developing economies laden with under-employment, unemployment, low capital in-flows, and investments [19,23,24,25,26,27]. However, despite the contributory role of the informal sector to the SA economy, several problems beset the SA informal sector. These might include internal factors, such as the lack of adoption of innovation, absence of ICT skills, low capitalization, absence of adequate production facilities, copyrights infringements, non-standardization of goods, and external factors, such as security problems, external competitive pressures, lack of social protection, infrastructural constraints, governmental zoning constraints, and poor governmental regulations and control [19,23,24,25,26,27]. Generally, internal factors (INTF) signify constraints experienced by individual SA informal sector providers that affect their businesses’ daily operational activities. In contrast, external factors (EXTF) represent issues that stem from government regulations, zoning laws (restricting the usage of public space), infrastructural challenges, security problems, and lack of social protection that inherently limit the productivity and competitiveness of the informal sector. Thus,

Hypothesis 10 (H10).

Internal factors (INTF) significantly influence users’ readiness to use a web technology portal (WTP); and

Hypothesis 11 (H11).

External factors (EXTF) substantially influence users’ readiness to use a web technology portal (WTP).

3.5. The Dependent Variable (Web Technology Portal)

Web technology portal (WTP) is the study’s dependent variable to assess an informal sector’s disposition to use web technology. It was operationalized using (Q19: I am aware of colleagues who use a web technology portal; and Q20: I do have support from family/peers to facilitate the use of a web technology portal). The research model in Figure 1 shows the constructs of this study as synthesized from the selected IS theories.

3.6. Moderation Variable- Gender Type and Business Subsectors

Generally, moderation is found when the effect of an independent construct (ID VAR) on a dependent construct (DP VAR) depends on the value of another variable introduced into the model. These other variables moderate the relationship [66,67]. Likewise, [10] used UTAUT/UTAUT2 model to assess technology adoption and opined that UTAUT/UTAUT2 was moderated by age, gender, experience, and voluntariness of use. There are differences in the gender decision-making process regarding technology [68].

In the UTAUT model developed by [10], the important constructs of the model (PE, EE, SI, and FC) are moderated by age, gender, innovativeness of use, and experience. Therefore, it was important to introduce control variables of gender and business type into our analytical process. Entrepreneurial activities (represented by the SA informal sector) are driven by socio-demographic variables such as age, gender, experience, educational status, and financial constraints [69]. In developing countries, several authors have emphasized the role of gender in technology adoption [70,71,72]. Hence it is essential to assess the effect of gender differences on the readiness of the SA informal service providers to utilize web technology portal.

Usually, in statistical analysis, a moderating variable can be either qualitative (such as gender, business type, educational level (high school, bachelor’s, master’s degree), or marital status (single, married, divorced)) or quantitative (age, height, population size). This study, therefore, investigated the moderating effect of gender and business subsectors on the independent variable per the dependent variable’s e-readiness to use web technology portal support.

4. Research Methodology

4.1. Research Setting

The focus of this study was solely on the informal sector service providers in the Cape Town metropolitan area in the Western Cape Province. The languages spoken are Xhosa, Afrikaans, and English. The Cape Town metropolitan area was chosen as the study site because it is one of the most developed cities in South Africa (SA), with an extensive concentration of the informal sector in its urban and suburban areas. Additionally, the structure of its informal sector is representative of the structure of the informal sector in other cities in South Africa. Therefore, the minimum requirements to participate in the study were a minimum age of 18 years old, and SA citizenship or being another African citizen resident in SA. Additionally, the participant must have been active in the informal economy for at least one year.

We adopted a survey research strategy for the study. The main premise was assessing the degree of e-readiness of the SA informal sector providers to utilize web technology portal support to enhance the productivity and viability of their informal businesses. A questionnaire instrument was developed and administered through face-to-face contact after random stratification of the population. Before data collection was initiated, a pilot study in Cape Town was performed with 20 participants. Steps were taken for the internal consistency test using Cronbach’s alpha reliability test. The instrument was divided into sections covering demographics, factors (internal and external) affecting the daily operational activities of their businesses, as well as the core section on being “e-ready” to leverage web technology portal adoption. A five-point Likert scale question—(1) strongly disagree, (2) disagree, (3) neutral, (4) agree, and (5) strongly agree—was employed to obtain their perceptions on technology adoption. We followed peers’ and experts’ advice on the suitability of the questionnaire to ensure content validity [73]. A total of 43 items became part of the questionnaire after changing several wordings and scales.

4.2. Sample Size and Validation

An a priori sample size for a structural equation model can be calculated using Equation (1). Here, the number of observed variables (32), the effect size (0.3—moderate), the number of latent variables (14 from the conceptual research model), the desired probability levels (0.05), and the recommended statistical power level (0.8 or above) were inputted into the formula [74,75,76]. The structural equation model (SEM) lower bound sample size: n = max (n₁, n₂), where computed according to Equation (1).

The author [63] opines that Equation (1) can be expressed as:

$$\begin{gathered} n_1=\left[ 50 {\left(\frac{j}{k}\right)}^2-450 \left(\frac{j}{k}\right)+1100\right] \\ {n}_2=\left[\frac{1}{2H}\left( A \left(\frac{\pi }{6}-B+D\right)+H+\sqrt{A \left(\frac{\pi }{6}-B+D \right)+H{)}^2+4AH\left(\frac{\pi }{6}+\sqrt{A}+2B-C-2D \right)} \right)\right] \end{gathered}$$

(1)

where A = 1 − p², B = p arcsine (p/2); D = A/√3 − A; H = ${\left(\frac{ẟ}{z_{1-\alpha /2}-z_{1-\beta }}\right)}^2$

Variable definitions:

J = number of observed variables

K = number of latent variables

p = estimated Gini correlation for a bivariate normal random vector

ẟ = anticipated sample size

α = Sidak correlated Type I error rate

β = Type II error rate

z = a standard normal score.

A, B, D, and H are constants introduced into the equation.

The calculation showed that the maximum sample size required to detect the effect was 208, while the minimum sample size for the model structure was 333. Hence, the recommended minimum sample size is 333. This study utilized data from 419 respondents.

However, according to [77], using Mplus software, there was no formal agreement in the extant literature regarding the appropriate sample size for SEM analysis. Several instances are found or do exist where testing SEM with a small sample size has been reported [78,79,80]. However, [81,82] posit that N = 100 to 150 can be the minimum sample size. Others considered N = 200 for a large sample size [83,84]. In circumstances where normally distributed indicator variables (with no missing data is the norm) are considered, then a reasonable or acceptable sample size for Confirmatory Factor Analysis (CFA) model, according to [85], is about N = 150. Across multigroup modelling, the accepted rule of thumb, as recommended, is 1000 cases per group [84].

4.3. Research Instrument

A paper-based questionnaire that contains items on the e-readiness of the South African informal sector providers to use an electronic portal that can promote their businesses was chosen as the research instrument for the quantitative study. The survey instrument was translated into isiXhosa, the local spoken language, to collect data from study participants. The research did not employ the digital format (Google Forms) for data collection due to the unavailability of a database of informal sector providers in the Cape Town metropolis with names, e-mail addresses and micro-enterprise types. The absence of this vital information meant that there was no contact medium. In addition, the respondents were South Africans and other Africans that are residents of SA.

According to [86], the rationale for using the questionnaire is that it is a powerful medium for gathering information (as data) on the perceptions, opinions, attitudes, and actions of a large group of subjects. The questionnaire contains 43 items. It had three parts with distinctive sections on: (a) the demographics (age, gender, citizenship); (b) questions related to technology use, business location and type, rationale/frequency for using technology; and (c) factors that enhance or impede micro-enterprise operational activities, and questions on research constructs. The constructs were assessed using a five-point Likert scale (1 = strongly disagree. 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree). The indicators were scaled with continuous, string and categorical scales. Additionally, the instrument contained questions to determine the business type, description or specificity of job type and business location.

4.4. Instrument Administration and Response Rate

A total of 480 questionnaires were administered to informal sector service providers after the electronic portal artefact demonstration. First, the performance/functionality of a web portal artefact (available at: https://uvuyo-prod.firebaseapp.com/, accessed on 3 March 2020) was presented to each respondent, followed by administering the questionnaire. Second, we used two assistants to assist in the data collection process because some respondents could not speak/understand English. Here, the language of communication was isiXhosa. Where the informal service provider did not understand English, a translated version of a similar questionnaire in isiXhosa was provided to the informal service provider. Depending on the nature of their work, some informal service providers preferred each question read while the assistants would then fill the questionnaire based on the verbal response provided. After administering the questionnaire, 61 returned questionnaires were discarded because of several missing data, non-completion (some opted out of the exercise), or repetitions in the type of response. The response/return rate was 87.29% [(419/480) × 100%].

4.5. Demographic Characteristics of Respondents

Table 3. Demographic of the Respondents.

Demographic	Characteristics	Number	Percentage
Gender	Male	211	50.4
	Female	208	49.6
	Total	419	100.0
Age	18 to 24 years	24	5.7
	25 to 30 years	89	21.2
	31 to 40 years	144	34.4
	41 to 50 years	118	28.2
	51 to 60 years	41	9.8
	Above 60 years	3	0.7
	Total	419	100.0
Years in Informal Business	Less than 5 years	116	27.7
	5 to 10 years	150	37.9
	Above 10 years	144	34.4
	Total	419	100.0
Citizenship/Residency status	South Africans	341	81.5
	Other Africans	65	15.5
	SA residents	13	3.1
	Total	419	100.0
Frequency of Use of Online Technology	Daily	196	46.8
	Occasionally	97	23.2
	Never	126	30.1
	Total	419	100.0
Types of Devices	PC/Notebook	17	4.1
	Mobile phones	118	28.4
	Smartphones	236	56.3
	PC tablets	28	7.0
	Others	20	5.0
	Total	419	100.0
Purpose for using a device	Business activities	124	29.6
	Personal activities	176	42.0
	Social networking	119	28.4
	Total	419	100.0

gives a summary of the respondents’ pertinent information per gender, age, business subsector, the number of years of activity in the informal economy, citizenship/residency status, frequency of use of online technology, types of devices employed in online activities, and the purpose for using the device. Among the respondents, 50.4% were male, and 49.6% were female. The ages of the respondents ranged from 18 years to above 60. The class distribution was: between 18 and 25 years (24 or 5.7%); from 26 to 30 years (89 or 21.2%); from 31 to 40 years (144 or 34.4%); from 41 to 50 years (114 or 29.2%); 51 to 60 years (41 or 9.8%); and above 60 years (3 or 0.7%)

The types of informal businesses spread across sampled communities were retail (12.9%); trade (23.9%); construction (8.8%), manufacturing (5%); health and beauty (22.0%); waste recycling (1.7%); home maintenance/repairs (6.4%); short courses and training (1.9%); appliance repairs (4.3%); deep cleaning (4.3%); and others (13.1%). Within each of these business types were specific concentrations of informal sector providers in landscaping/plumbing (13.1%); mechanics/manufacturing (5.7%); hairdressers and barbers (23.0%); retail/trade (30.5%); recyclers/waste pickers (2.4%); short courses and training (1.2%); deep cleaning/domestic workers (6.4%); taxi drivers/security (4.5%) and other (12.4%).

5. Analysis of Data

5.1. Reliability

Reliability for a survey design represents the propensity that the responses or scores collected from the participants are stable and consistent over a specific timeframe [87].

In this study, Cronbach’s alpha coefficient [88,89] was used to test the reliability of the research instrument. The results of data analysis using IBM SPSS version 26 (New York, NY, USA) showed values of (α = 0.775) and McDonald’s value (ω = 0.752). Furthermore, item reliability statistics varied from a minimum of α = 0.728 to α = 0.786 as a maximum in circumstances where items/indicators in the questionnaire were either adopted or dropped, signifying good reliability values.

5.2. The Measurement Model

The research model’s validity and reliability indices were examined thoroughly for a reflective measurement model [90]. We used the composite reliability (CR) and Cronbach’s alpha coefficients to assess internal consistency/reliability in the study.

Using

Table 4. Composite reliability (CR), Average variance extracted (AVE).

Variable	Indicator	Loading	CR	AVE
INTF	Q12	0.95	0.78	0.61
	Q13	0.95
	Q14	0.16
EXTF	Q15	1.00	0.42	0.34
	Q16	0.11
	Q17	0.10
SE	Q18	0.85	0.85	0.74
	Q21	0.87
PE	Q22	0.23	0.72	0.5
	Q23	0.85
	Q24	0.85
EE	Q25	0.82	0.9	0.75
	Q26	0.90
	Q27	0.87
SI	Q28	0.90	0.9	0.82
	Q29	0.91
FC	Q30	0.70	0.85	0.58
	Q31	0.74
	Q32	0.82
	Q33	0.77
INNO	Q34	0.98	0.83	0.72
	Q35	0.69
DISC	Q36	0.68	0.86	0.68
	Q37	0.88
	Q38	0.89
INSEC	Q39	0.91	0.89	0.81
	Q40	0.88
OPM	Q41	0.91	0.77	0.55
	Q42	0.79
	Q43	0.46
WTP	Q19	0.89	0.9	0.82
	Q20	0.92

, both Composite Reliability (CR) and Cronbach’s alpha exceed the recommended thresholds, 0.60 [91] as well as 0.60 to 0.70 [66,92]. Thus, most constructs have achieved internal construct reliabilities except external factors in which items (RQ16 and RQ17) loaded below 0.11. For internal factors (INTF), RQ 14 loaded at about 0.16. Equally, for performance expectancy, RQ 22 loaded at 0.23. (These questions were kept because too few items/indicators for the constructs would be obtained, and the items were deemed central to the overall analysis.)

Additionally, the validity of the reflective model is dependent on convergent validity and discriminant validity. According to [93], convergent validity occurs when two measures act on a common construct. We used average variance extracted (AVE) for each measured latent variable to assess convergent validity. Table 4 shows that all constructs have their AVE (except External factor = 0.34) greater than 0.50, the minimum recommended threshold [84]. Interpretatively, RQ14 did not converge effectively on the construct Internal Factor; furthermore, RQ16 and RQ17 did not converge properly on the construct External Factor (we did not delete these items for a few indicators that would have been responsible for these constructs (INTF and EXTF)).

5.2.1. Discriminant Validity

The factor loading (

Table 5. Discriminant Validity of Research Model Construct (Fornell-Larcker Criterion).

	DISC	EE	EXTF	FC	INTF	INNO	INSEC	OPM	PE	SE	SI	WTP
DISC	0.82
EE	−0.36	0.86
EXTF	−0.19	0.27	0.58
FC	−0.44	0.65	0.38	0.76
INTF	0.37	−0.62	−0.36	−0.60	0.78
INNO	−0.14	0.39	0.08	0.41	−0.30	0.85
INSEC	0.60	−0.35	−0.14	−0.36	0.40	−0.31	0.90
OPM	0.20	−0.21	−0.07	−0.21	0.24	−0.15	0.45	0.74
PE	−0.36	0.64	0.27	0.66	−0.57	0.34	−0.34	−0.17	0.71
SE	−0.42	0.62	0.36	0.66	−0.67	0.36	−0.39	−0.22	0.61	0.86
SI	−0.33	0.64	0.36	0.65	−0.61	0.27	−0.31	−0.20	0.57	0.57	0.90
WTP	−0.32	0.54	0.43	0.59	−0.57	0.26	−0.33	−0.18	0.56	0.62	0.59	0.91

and

Table 6. Heterotrait-Monotrait Ratio.

	DISC	EE	EXTF	FC	INTF	INNO	INSEC	OPM	PE	SE	SI	WTP
DISC
EE	0.42
EXTF	0.27	0.29
FC	0.55	0.82	0.48
INTF	0.43	0.78	0.46	0.78
INNO	0.27	0.47	0.10	0.44	0.47
INSEC	0.72	0.44	0.18	0.47	0.50	0.36
OPM	0.27	0.27	0.18	0.26	0.31	0.22	0.60
PE	0.57	1.01	0.41	1.08	0.96	0.54	0.52	0.28
SE	0.56	0.84	0.45	0.94	0.95	0.45	0.56	0.30	1.09
SI	0.40	0.80	0.35	0.85	0.77	0.30	0.40	0.24	0.93	0.79
WTP	0.38	0.66	0.39	0.76	0.68	0.28	0.42	0.21	0.92	0.86	0.75

), the Fornell-Larcker test and the application of the heterotrait-monotrait ratio of correlations (HTMT) have been assessed and considered for discriminant validity determination. Based on the Fornell and Larcker criterion, each construct’s average variance extractor (AVE) must be greater than its corresponding squared correlation with another construct [94]. For example, at a factor loading of 0.05 significant level, AVE should be equal to or greater than 0.70 [95]. When AVE is between 0.40 and 0.70, attention should be paid to it with possible deletion. According to [96], the cut-off range is 0.50 to 0.70.

Additionally, [97], utilizing simulation study, indicates that the Fornell-Larcker criterion and cross-loadings need to be sufficiently robust to provide substantial evidence for the lack of discriminant validity in most research situations. The authors then opine that the appropriateness of an alternate approach is based on a multitrait-multimethod matrix for the assessment of discriminate validity, which is based on the employment of a heterotrait-monotrait ratio of correlations (HTMT). According to [87], the HTMT approach was far superior and more robust than the Fornell-Larker criterion or factor loadings after performing a Monte Carlo simulation. The pertinent guidelines for variance-based structural equation modelling have been suggested by [97]. However, discriminant validity is established if the HTMT value is less than 0.90.

The results after the HTMT analysis were relatively low (

Table 7. Goodness of Fit Values.

	Saturated Model	Estimated Model
SRMR	0.07	0.08
Chi-square	2311.97	2504.15
NFI	0.65	0.62

), and contributions from this measure was less important. However, the results showed distinctiveness between the various constructs in the synthesized research model (HTMT< 0.9). Therefore, the study relied more on the Fornell-Larcker criterion and factor loadings.

5.2.2. Good-of-Fit (GoF)

The goodness-of-fit (GoF) measure has been advanced for models employed in PLS-SEM analysis. However, GoF has a weakness because it cannot provide a reliable distinction between valid and non-valid models, resulting in limitations to its applicability [98]. Nevertheless, GoF measures are used frequently in multigroup analysis (PLS-MGA) [99].

There is no optimized global scalar function for PLS-SEM, and the absence of globally accepted goodness-of-fit measures places obstacles in the way of the applicability of PLS-SEM. Therefore, the goodness-of-Fit (GoF) statistic was used in this study because it represents the discrepancy between observed or approximated (latent variables-LVs) values obtained for the dependent variables and values. The authors of [100] proposed global goodness-of-fit measures, which have been discounted by [98] because these measures need to be more suitable for misspecified models. Similarly, refs [98,101] opined that the GoF measure could be employed for the model’s quality when PLS-SEM is the chosen analytical tool.

Following the consensus of [98,101], we placed little emphasis on GoF results obtained after the analysis with SmartPLS-SEM (3.3.5). The rationale is that this is a synthesized research model that still needs to be established. This study is the first attempt at predicting the quality of the research model with a dataset obtained from informal sector practitioners in the Cape Town metropolitan area. It will be necessary to validate the model with several datasets before generalizations can be made.

The results of the goodness-of-fit (GoF) test are presented in Table 7.

The scientific experimental fit value for SRMR is 0.07/0.08, while NFI is 0.65/0.62. According to [102], the recommended NFI value should be above 0.95 (or 0.90). The RMS_theta value was 0.15. The authors in [103] recommended RMS_theta value ≤ 0.12–0.14 for a well-fitting model [102]. Therefore, in this study, Goodness-of-Fit statistics was moderately satisfied.

In the extant literature, the approximate fit index for SRMR is less than 0.10 or 0.08 [104], and NFI > 0.90. Thus, using SmartPLS3, after performing the Bollen-Stine bootstrapping procedure (complete bootstrap option) in SmartPLS3, the critical threshold values are SRMR < 0.08 and NFI > 0.90 [105,106,107], and RMS_theta ≤ 0.12–0.14 [102,103].

5.3. Analysis of the Structural Model

PLS-SEM can evaluate the latent inner model (structural model), and model quality is assessed with the coefficient of determination (R²), cross-validation redundancy (Q²), path coefficients, and effect size (f²).

In

Table 8. Coefficient of determination (R2) of the research model.

	R Square	R Square Adjusted
WTP	0.52	0.51

, the results show that the independent variables (internal factors, external factors, self-efficacy, performance expectancy, effort expectancy, social influence, facilitating conditions, optimism, innovativeness, discomfort, and insecurity) accounted for the variance of 51.0% observed in the use of web technology portal by the SA informal practitioners. Since R² applies to several disciplines, a rough ‘rule of thumb’ for R² is 0.75 (substantial), 0.50 (moderate), and 0.25 (weak levels), respectively [108,109]. Therefore, according to Table 8, this study exhibited a moderate R² value.

The model’s blindfolding-based cross-validated redundancy measure or predictive relevance (Q²) is high (

Table 9. Predictive Relevance of the Constructs in the research model.

	SSO	SSE	Q2 (=1 − SSE/SSO)
DISC	1257	1257
EE	1257	1257
EXTF	1257	1257
FC	1676	1676
INTF	1257	1257
INNO	838	838
INSEC	838	838
OPM	1257	1257
PE	1257	1257
SE	838	838
SI	838	838
WTP	838	497.24	0.41

). The application of blindfolding [110,111] on the sample reuse technique allowed the determination of the Stone-Geisser’s Q value. If PLS-SEM exhibits predictive relevance (Q² > 0), the data point indicators would be predicted. A Q² > 0 for several endogenous latent variables shows the PLS path model can predict relevance for this construct [101]. Q² values are classified into three levels: (a) 0.02 (small); (b) 0.15 (medium); and (c) 0.35 (large) [112]. According to Table 9, this study had a Q² value of 0.41 (large).

Effect Size (f²)

The effect size for each path model is elucidated by calculating Cohen’s f². The computation of f² is possible by evaluating changes in R² by eliminating a specific construct from the model. Two PLS path models must be estimated thus: (a) calculate the path model for the full model with specified hypotheses, yielding R² (i.e., R² included) and (b) estimate the second model excluding selected independent/exogenous variables (R² excluded). Thus:

$${\mathit{f}}^{\mathbf{2}}=\frac{{\mathit{R}}^{\mathbf{2}} included-{\mathit{R}}^{\mathbf{2}} excluded}{1-{\mathit{R}}^{\mathbf{2}} included}$$

(2)

Based on f² values, Cohen [75] stipulated 0.02 (small), 0.15 (medium), and 0.35 (large) effects, respectively.

Hence, the size effect on Web Technology Portal is shown in

Table 10. The Size Effect on WTP.

Predictor	WTP
DISC	0.00
EE	0.00
EXTF	0.05
FC	0.01
INTF	0.01
INNO	0.00
INSEC	0.01
OPM	0.00
PE	0.02
SE	0.04
SI	0.03

.

Table 10 shows the results for effect sizes and f² for the respective path coefficient. According to [66], a path with a high value of f² does imply that the exogenous construct explains the endogenous construct. Established as well as acceptable size effects f² values follow similar levels of predictive relevance (Q2) with 0.02 (small), 0.15 (medium), and 0.35 (large) effect size [66]. The size effect for this study exhibits small f2 values.

5.4. Moderation of Web Technology Portal

Multiple group analysis was conducted to assess the effects of the independent variables on using a web technology portal in the subsectors of SA’s informal sectors.

Table 11. Effect of Independent Variables on WTP with respect to Gender (Moderating variable).

Path	Path Coefficients-Diff (MALE vs. FEMALE)	t-Value (|MALE vs. FEMALE|)	p-Value (MALE vs. FEMALE)
SE → WTP	−0.15	1.08	0.280
PE → WTP	0.01	0.09	0.930
EE → WTP	0.20	1.51	0.130
SI → WTP	−0.03	0.18	0.850
FC → WTP	−0.21	1.56	0.120
OPM → WTP	−0.14	1.81	0.070
INNO → WTP	0.01	0.11	0.910
DISC → WTP	−0.04	0.50	0.620
INSEC → WTP	0.00	0.04	0.970
INTF → WTP	−0.06	0.45	0.650
EXTF → WTP	0.05	0.56	0.580

and

Table 12. Effect of Independent Variables on WTP with respect to Business Type (Moderating Variable).

Path	Path Coefficients-Diff (TRADER vs. SERVICE_PRO)	t-Value (|TRADER vs. SERVICE_PRO|)	p-Value (TRADER vs. SERVICE_PRO)
DISC → WTP	−0.14	1.50	0.13
EE → WTP	−0.04	0.32	0.75
EXTF → WTP	0.07	0.62	0.53
FC → WTP	−0.05	0.40	0.69
INTF → WTP	−0.07	0.58	0.57
INNO → WTP	−0.01	0.06	0.95
INSEC → WTP	0.04	0.39	0.70
OPM → WTP	−0.06	0.73	0.47
PE → WTP	−0.18	1.37	0.17
SE → WTP	0.03	0.24	0.81
SI → WTP	0.08	0.55	0.58

show no significant differences in the predictive power pattern of the independent variables on the use of a web technology portal by the SA informal sector concerning gender/or informal business subsectors.

5.4.1. Gender

We found that gender classification of the SA informal service providers, regarding all independent constructs (INTF, EXTF, SE, PE, EE, SI. FC, OPM, INNO, DISC, INSEC), did not affect their tendency to leverage web technology portal support.

The result is presented in Table 11 after multigroup analysis.

Here the p values are not significant after multigroup analysis. The p values are greater than 0.05.

5.4.2. Business Type

We classified business type into two main subsectors: (a) services and (b) trade/retail for multigroup analysis to be performed. All the independent constructs (ID VAR) were operationalized on e-readiness for web technology portal support. The results are presented in Table 12. The results showed that p values were not significant after multigroup analysis.

The results obtained after testing all the formulated hypotheses are shown in

Table 13. Results of Hypotheses Testing.

Hypothesis	Path	Original Sample (O)	Sample Mean (M)	Standard Deviation (STDEV)	t -Value (|O/STDEV|)	p-Value	Hypothesis Outcome
H1	SE →WTP	0.23	0.22	0.06	3.53	0.000	Supported
H2	PE → WTP	0.14	0.15	0.06	2.22	0.030	Supported
H3	EE → WTP	0.03	0.02	0.06	0.48	0.630	Not supported
H4	SI → WTP	0.19	0.18	0.07	2.84	0.000	Supported
H5	FC→ WTP	0.09	0.09	0.07	1.41	0.160	Not supported
H6	OPM→WTP	0.01	0.00	0.04	0.33	0.740	Not supported
H7	INNO→WTP	−0.02	−0.02	0.05	0.49	0.620	Not supported
H8	DISC→WTP	0.04	0.04	0.04	0.98	0.330	Not supported
H9	INSEC→ WTP	−0.07	−0.06	0.04	1.60	0.110	Not supported
H10	INTF→ WTP	−0.09	−0.09	0.06	1.35	0.180	Not supported
H11	EXTF→ WTP	0.17	0.17	0.04	3.84	0.000	Supported

, while the structural model generated by PLS-SEM is shown in Figure 2.

6. Discussion

This study showed that in trying to predict the e-readiness of the South African informal sector practitioners to leverage a web technology portal for business activities, the constructs of self-efficacy (SE), performance expectancy (PE), social innovation (SI), and internal factors had a significant positive influence on e-readiness to use of web technology portal. In contrast, other constructs—effort expectance (EE), facilitating conditions (FC), optimism (OPM), innovativeness (INNO), discomfort (DISC) and insecurity (INSEC)—have a non-significant positive influence on e-readiness to use a web technology portal. These findings support previous studies in extant literature [113,114], where most of the constructs in the Technology Acceptance Model (TAM)—PU, PEOU—and a combination of UTAUT/TRI (PE, EE, FC, SI, OPM, INNO, DISC, and INSEC) all had a significant positive influence on the behavioral intention to use technology (BIUS).

The result for H1 shows that self-efficacy positively affects the use of a web technology portal (H1: t = 3.53, p < 0.001). Since p < 0.05 (Table 13) indicates a positive effect (positive influence) on the use of a web technology portal, therefore the hypothesis is supported. According to Bandura [11,12,13,46,65,66], self-efficacy signifies the personal beliefs of an individual relative to their ability to execute behavioural tendencies required to complete specific tasks.

Using a TAM model and LISREL for structural equation modelling, [113] opined that computer self-efficacy had a minimal but negative effect on perceived usefulness (PU) and no significant effect on perceived ease of use (PEOU). The rationale for including computer attitude and self-efficacy in TAM, in the view of [113], was to significantly improve the explanatory power of the hypothesized model on the variance governing perceived usefulness (PU) to enhance the behavioural intention of using information systems. Usually, perceived usefulness (PU) can be predicted by computer self-efficacy because perceptions of competence to use technology infers the ability to use it for increased productivity and effectiveness at work and in life.

The result for H2 shows that performance expectancy positively influences the use of a web technology portal (H2: t = 2.22, p = 0.030). Since p > 0.05, therefore, H2 is supported (Table 13). Performance expectancy represents the personal belief that actions will lead to intended performance goals. This belief is driven by past experiences, self-confidence, and perceived difficulty in attaining the goal. Competence, goals, difficulty, and control affect an individual’s perception of performance expectations [114,115,116]. Regarding SA informal service providers, moderate levels of self-efficacy enhance performance expectancy and can lead to beneficial outcomes from using a web technology portal.

The result of H3 shows that effort expectancy has a non-significant positive effect on the use of a web technology portal (H3: t = 0.48, p = 0.630) because p > 0.05. Therefore, H3 is not supported (Table 13). This result indicates that SA informal sector service providers believe using a web technology portal will be relatively easy. However, a lack of skill in technology use and interest in using technology negatively affected the degree of ease associated with using a web technology portal.

The result of H4 reveals that social influence (SI) has a significant positive influence on the use of web technology portal (H4: t = 2.84, p < 0.001). Hence, H4 is supported (p <.05) (see Table 13). This suggests that social influence, including persuasion from friends, colleagues, family members, and peers, helped enhance their propensity to use a web technology portal.

The result of H5 shows that the construct, facilitating conditions, has a positive non-significant influence on the use of a web technology portal (H5: t = 1.41, p = 0.160). Because p > 0.05, H5 is not supported (Table 13). The result indicates that the individual beliefs of SA informal sector practitioners about the supporting organizational and technical infrastructure are relatively independent of their intention to use the web technology portal.

According to [10], the UTAUT model shows the relationship between behavioral intention and user behavior towards technology. These are typically influenced by the performance expectations (represented by performance expectancy) during business operations (effort expectancy), social influence, and what supporting conditions (facilitating conditions) are available. All these constructs are moderated by age, gender, experience, and voluntariness of use. The UTAUT model is a combination of eight other models, which are: the theory of reasoned action (TRA), technology adoption model (TAM), motivational model (MM), theory of planned behaviour (TPB), combined TAM and TPB, model of PC utilization (MPCU), innovation of diffusion theory (IDT), and social cognitive theory (SCT). Using UTAUT is more adaptable to each of the eight theories and explains 70% of the variance [10]. An initial evaluation of the (UTAUT) model by [10] indicated that constructs such as performance expectancy (PE), effort expectancy (EE), social influence (SI), facilitating conditions (FC), attitude toward the use of technology, and self-efficacy, are significant direct determinants of behavioural intentions. A further evaluation presented four main constructs—PE, EE, SI and FC—as primary determinants of behavioural intention, while others might not be significant.

This study showed that facilitating conditions (FC) had a non-significant positive influence on using a web technology portal. Thus, the SA informal service providers were unaware of possible access to technical, institutional, and governmental support that might facilitate the perceived ease of use (PEOU) of a web technology portal, which could influence their e-readiness.

The result of H6 reveals that optimism does not significantly influence users’ readiness to use a web technology portal (H6: t = 0.33, p = 0.740). Since p > 0.05, H6 is not supported. The result suggests no optimism about using a web technology portal despite the added advantage of bestowing more control on the user and enhancing the overall quality of life. The authors of [57,65] devised the Technology Readiness Index (TRI) to determine a person’s predisposition to use newer technologies/innovations to achieve specific life goals. People’s tendency to embrace and use new technologies to accomplish goals in either home life or work represents technology readiness [117]. People’s perceptions of technology have positive and negative sides. Optimism dimensionally signifies a positive view of technology and the belief that it will be advantageous to use technology to improve an individual’s efficiency and performance at work or home. As a construct, optimism portrays control, flexibility, and efficiency in people’s lives. Optimism looks at technology holistically per its usefulness and ease of use with little emphasis on the negative aspect of its use [118]. Consequently, optimists are more predisposed to the use of new technologies. Our result indicates that the SA informal providers are not optimistic about using a web technology portal. Compelling factors such as discomfort and unawareness of the availability of facilitating conditions impede their intention to use a web technology portal.

The result of H7 shows that innovativeness does not significantly influence users’ readiness to use a web technology portal (H7: t = 0.49, p = 0.620) since p > 0.05, thus H7 is not supported. Innovativeness, as a construct, refers to being a pioneer and an important determinant of the perceptions of convenience to be derived and perceived usefulness [118]. People with very high technological innovativeness experience thrills in using new technologies and exhibit intrinsic motivation to try fresh innovations. The result implies that they are early adopters [119]. The wage earnings of 78% of the SA informal sector service provider is about ZAR 79,000 per annum (USD 4647/per annum) [27]. According to the SA Internal Revenue Board [27], this is below the base threshold of taxable income for South Africans. Typical hourly wage earnings for a one-account worker in the informal sector are R 18 (USD 1.20) for men compared to R13 (USD 0.88) for women. Therefore, insufficient wage earnings affect their innovativeness. Granted that the informal sector service providers are “digital immigrants” having exposure to technology during their productive lives, as opposed to “digital natives” born with technology [120], it is understandable why they have a lower tendency to innovativeness in web technology portal usage. This is compounded by the fact that they must adapt to new technology and feel comfortable around technology. More so than usual, the SA informal worker needs to gain the necessary skills in using technology.

This finding could be explained further using Roger’s Diffusion of Innovation Model (Figure 3) [6,121]. Informal sector providers in the Cape Town metropolis do not belong to the 2.5% (innovators) or 13.5% early adopters but more likely fall into the 34% (late majority) and 16% (laggards). This is due to their wage earnings. Laggards are older individuals or older women with low earnings, with a lower standard of education, and with fewer employable skills.

The result of H8 shows that discomfort does not significantly affects the user’s readiness to use a web technology portal (H8: t = 0.98, p = 0.330). Thus, since p > 0.05, H8 is not supported. This result means that discomfort does not significantly influence micro-entrepreneurs’ ability to use a web technology portal. Discomfort gives the perception of a lack of control over web technology portals because individuals are overwhelmed by technology [118]. People who show discomfort in using web technology portals believe that technology is complicated for an end-user, is not user-friendly, and is not meant for the public good. They prefer to use traditional modes of micro-enterprise operation with crude technologies.

The result of H9 shows that insecurity does not significantly affects users’ readiness to use a web technology portal (H9: t = 1.60, p = 0.110). Since p > 0.05, H9 is not supported. The result indicates that when a feeling of insecurity that stems from notions such as a reliance on technology can be harmful to the business or can lead to problems/disruptions in business operations does exist, it still has a non-significant positive effect on users’ readiness to use a web technology portal amongst SA informal sector practitioners.

In the views of [65], the TRI model utilizes four psychological variables: optimism, innovativeness, discomfort, and insecurity, to assess users’ readiness for technological innovation.

(a)

Optimism: reflects a positive attitude to technology with the perception that using technology would improve control and enhance flexibility and life efficiency;

(b)

Innovativeness: depicts the tendency of an individual to be the first to employ technology when newly introduced to the marketplace. Additionally, they are excited and are forerunners in experimenting with newer innovations;

(c)

Discomfort: connotes when individuals are having difficulty controlling technology and seem overwhelmed by it;

(d)

Insecurity: signifies problems regarding technology security and questions about personal data, which introduces an element of distrust of technology-based applications, transactions, and workability [122].

Generally, optimism and innovativeness are observed as “contributors” to technology readiness, while discomfort and insecurity are “barriers” to technology readiness.

According to [57,65], TRI determines one’s perceptions or beliefs about technology. It considers the innovative mastery capabilities of an individual. The technology-readiness score classifies users into explorers, pioneers, sceptics, paranoids, and laggards. Optimism and innovativeness drive the highest score (contributors), while discomfort and insecurity lead to the lowest score (as inhibitors). Explorers are interested in new technologies, becoming the first to explore and try newer technologies. Socio-economically, explorers are young, male, highly educated with high salaries. Comparatively, laggards are the last adoptive of new technologies. They generally exhibit the highest scores in the measure of inhibitors and the lowest scores in the dimensions of contributors. Laggards are typically older segments of society, women with lower educational and income levels. The other three groups—pioneers, sceptics, and paranoids—have rather complex perceptions of technology [57,65]. This study showed that discomfort, optimism, innovativeness, and insecurity have a non-significant positive influence on technology readiness. This supports the argument that most informal sector service providers are either paranoid/late majority (34%) or laggards (16%).

Based on the hypothesized research model (Figure 2), the SEM analysis showed that four constructs, external factors (EXTF), performance expectancy (PE), self-efficacy (SE), and social influence (SI), had positive relationships with e-readiness to use a web technology portal (WTP). The other seven, effort expectancy (EE), facilitating conditions (FC), optimism (OP), innovativeness (INNO), insecurity (INSEC), and internal factors (INTF), all had a non-significant positive influence on e-readiness to use a web technology portal (WTP).

Comparatively, [123] utilized the unified theory of acceptance and use of technology (UTAUT) and technology readiness index (TRI) as a theoretical model concerning e-government on a target population in Indonesia with a sample size of 225. The authors observed that citizens living in Jakarta (SCR citizens) could still be characterized as possessing low TRI (considered belonging to the Low Technology Readiness Group, with a TRI value equal to 2.7). In the study, data analysis was done by using descriptive statistics and multiple linear regression. However, the authors observed that the constructs, PE, EE, SI, and FC, showed significant positive effects on behavioural intention to use the system (BIUS) (which was the dependent variable).

Generally, the beliefs and perceptions of South African informal sector service providers to use a web technology portal did not significantly affect their intention to use technology. Hence, only three (SE→WTP, PE→WTP, and SI→WTP) out of nine constructs significantly influenced the e-readiness of the South African informal sector service providers to use a web technology portal.

7. Conclusions

The beliefs and perceptions of informal service providers in the Cape Town metropolis concerning using a web technology portal were elucidated through the constructs adapted for this study. The hypothesized research model, synthesized from self-efficacy theory (SET), unified theory of acceptance and use of technology (UTAUT), and technology readiness index (TRI), contained nine constructs. Of these, discomfort, effort expectancy, innovativeness, optimism, insecurity, facilitating conditions, and internal factors had a non-significant positive influence on the e-readiness of the SA informal sector providers.

This study observed a significant positive relationship between self-efficacy, performance expectancy, social influence, and external factors in using web technology portals. This implies that informal sector service providers who intend to adopt web technology portals exhibited significant positive behaviour despite inhibiting internal factors, discomfort in using technology, lack of innovativeness, absence of optimism, and feelings of insecurity towards technology. Their significant others (family members, peers, colleagues, friends) provide persuasive psychological motivation for them to overcome barriers to all internal factors: discomfort, optimism, innovativeness, effort expectancy, insecurity, and facilitating conditions. This propels them to take decisive actions toward using a web technology portal. Performance expectancy, social influence, self-efficacy, and external factors positively and significantly influenced their intention to use technology and, subsequently, technology readiness. The result implies a strong relationship between the exogenous variables (PE, SI, SE, EXTF) and the endogenous variable (e-readiness to use web technology portals). Interpretatively, although informal sector workers showed significant positive intent to use web technology portals, with others (friends, family members, peers, etc.) playing a significant role in stimulating this intention, generally these workers were not optimistic about technology, innovative enough per technology use, and did not feel secure with technology. More so, they did not trust available facilitating conditions sufficient to warrant the need to leverage web technology portal support. Thus, their intention to use web technology portals is impacted negatively due to the perceived difficulties in using technology to control, maintain a flexible life, and obtain more productive outcomes at home and work. Thus, informal sector service providers do not control all the internal factors, ease of use, or infrastructural challenges that impede the use of web technology portals for business venturing. These perspectives require more research and hypothesis-building. However, it could be argued that the exogenous variables of performance expectancy (PE), self-efficacy (SE), social influence (SI), and external factors (EXTF) all showed significant positive relationships with the intent to use web technology portals and technology readiness since the hypotheses that are based on these constructs were supported. These imply direct relationships between exogenous variables (PE, SE, SI, EXTF) and the use of web technology portals (endogenous variable). Therefore, informal sector service providers who showed a significant positive intent to use web technology portals for business operations are impacted positively by performance expectations, significant levels of self-efficacy, social influence, and external factors. Therefore, perceived performance expectations, enhanced social influence, significant levels of self-efficacy, and external factors could be accepted as the dominant drivers of using web technology portals by SA informal sector service providers.

7.1. Managerial and Design Implications

The findings have identified determinants that the SA informal sector practitioners emphasize: self-efficacy, performance expectancy, social influence, and external factors. All four constructs significantly positively influenced the degree of e-readiness to use web technology portals. Hence, the hypotheses related to these constructs were supported. The findings convey valuable insights for mobile software designers/developers, mobile sales personnel, mobile commerce merchants, online retailers, and other related stakeholders. Regarding these valuable insights, designers must factor in performance expectations, self-efficacy, and social influence on web technology portals in the future. The dominant determinant is performance expectancy. Web technology portals must satisfy the core interests of SA informal sector service providers in meeting all expectations in business operations. When this phenomenon is satisfied, their self-efficacy per technology use would be enhanced, and social influence will come into play to incentivize them to leverage web technology portal support.

From a managerial perspective, effort expectancy, facilitating conditions, optimism, innovativeness, discomfort, and insecurity are non-significant determinants of their employees’ readiness to use web technology portals. Therefore, the most important requirement is for management to facilitate high self-efficacy and performance expectations by stressing the potential benefits to be obtained from the performance of the web technology portal. When this is achieved, all other non-significant constructs could be minimized, hence the eventual leveraging of a web technology portal for daily business operations.

7.2. Direction for Future Work

Firstly, the sampling and scope were limited to informal sector hotspots in the Cape Town metropolitan area. More studies should be undertaken in other SA provinces, such as KwaZulu Natal, Limpopo, and Eastern Cape, with relatively high informal sector concentration, to enable better generalization over the informal sector population of South Africa. Additionally, demographics about the respondents, such as gender, age, residency status, business type, frequency, and reasons for using online technology, were skewed toward the older population from age 31 to over 60. These were mainly shop owners and persons whose businesses were in informal hotspots.

Secondly, the collected data on web technology portal support came from the present disposition of informal sector service providers. Therefore, the degree of e-readiness alignment with their intent to utilize web portal technologies in the future might change, depending on circumstances and financial constraints.

Thirdly, to ensure a deeper understanding of the e-readiness of the SA informal sector, it will be necessary to conduct studies focusing on specific groups and sub-sectors of the SA informal sector. For example, this type of study could include female micro-entrepreneurs, young micro-entrepreneurs, and older people in particular sub-sectors.