Model of Key Factors in the Sustainable Growth of Small and Medium-Sized Enterprises Belonging to the First Nations

Abstract

The search for better living conditions, hand in hand with economic development, is a desire shared by all peoples; the First Nations are no exception. In this sense, entrepreneurship is one of the ways to improve incomes and quality of life, both in industrialized and developing societies, which is considered a potential strategy for economic development. This exploratory research presents a model that explains, through causal relationships, the growth of Small and Medium-sized Enterprises (SMEs) belonging to the First Nations, based on the results coming from the analysis conducted within one of the most important First Nations in the Americas, the Mapuche people, located in South America (mainly Argentina and Chile). The framework was developed from interviews with entrepreneurs and owners of Mapuche SMEs, along with an exhaustive analysis carried out through the use of Partial Least Squares (PLS). The owners were consulted about their attitude towards variables that generate accelerated growth in entrepreneurship in different contexts around the world. Subsequently, a model of inter-relationships was generated that sought to explain which variables are determining factors in the growth of SMEs belonging to the First Nations. Through a process of evaluation and depuration, the model proposed here was arrived at, concluding that Constant Training and Commitment to Growth are the most relevant factors in the growth of these companies. Both of these factors are supported by Long-Term Customer Relationships, Differentiation by Quality, Business Skills, and Business Structure, with a special focus on sustainable development.

1. Introduction

Issues such as inequality of opportunity and quality of life are common among First Nations people as well as in the rest of the population [1]. In developing countries, the main factor is economic income [1,2,3]. In this sense, it is well known that economic development is crucial to improve the quality of life and reduce poverty in developing countries [4]. Although there are more than 500 million First Nations people worldwide, the vast majority of research focuses on the entrepreneurial attitude and entrepreneurship of those who are not First Nations [5].

Moreover, according to certain viewpoints found in the literature [6,7,8], technological and knowledge advancements attained through innovation processes can generate wealth and enhance living standards. Their opportunities are multiplied by technology, easy and affordable access to knowledge, and their participation in decisions that directly affect their lives, while creating new and better job opportunities, among many other possibilities. The search for ways to develop entrepreneurship and innovation, while not ensuring an improvement in living conditions, does expand the possibilities for development, use of technology, generation of wealth, and better employment opportunities [7,9].

As a result, there is an unfavorable situation of great inequality and low economic development for First Nations. An opportunity to improve this situation is found in the development of fast-growing enterprises and innovations; as of the date of this study, it has not been possible to determine the existence of research that has studied the development of fast-growing and/or innovative enterprises that consider the particularities, idiosyncrasies, cosmogony, social relations, and other differentiating factors of the First Nations. Considering the historical context in which we live, it is increasingly necessary to make progress on this issue, especially if we consider that the entrepreneurial attitude is closely related to the culture in which entrepreneurs operate [2].

If studies on this subject are not carried out abundantly, the current critical situation of the First Nations will likely continue or worsen, which hinders the full development of their people.

Thus, the main objective of this research is to determine and evaluate the most relevant factors for the rapid growth of SMEs belonging to the First Nations, based on the findings coming from the Mapuche people, one of the most relevant First Nations in the Americas [10]. Since no previous research has been found on the subject, all the variables detected in the bibliographic study are considered in an exploratory manner [11], to construct a model capable of representing their interactions and causal relationships in a predictive manner. This may help answer questions, such as the following: is it possible to create a model that explains the most influential factors that may lead SMEs belonging to the First Nations to achieve higher levels of growth? What are the main variables that determine such growth within the First Nations SMEs?

2. Literature Review

2.1. Entrepreneurship, Innovation, and First Nations

It is generally believed entrepreneurship and innovation are matters alien to the First Nations [5]. The truth is that in the Mapuche case, there are records dating back from the fifteenth century. In an example of what Tidd and Bessant [12] call Process Innovation, modifications are developed in the way of carrying out productive processes from the introduction of technology from the Inca Empire and later from the Spanish Empire to increase production [13]. Later, in the seventeenth century, with the introduction of the horse, a livestock regime was developed, allowing the development of an abundant market economy that extended until the nineteenth century when the invasion of Chile and Argentina occurred [13,14]. Innovation, entrepreneurship, and commerce, in general, are factors present for centuries in First Nations such as the Mapuche people.

Although it is not easy to address the issues related to entrepreneurship and innovation from the perspective of the First Nations, some authors have made significant efforts to address them from different perspectives. In this sense, a brief discussion is presented below regarding what entrepreneurship and innovation (also associated with rapid-growing companies) mean for First Nations and how the culture, society, and public policies can be analyzed to better understand this complex relationship between entrepreneurship, innovation, and the First Nations.

First, Lindsay [5] carried out research that sought to explain the impact of cultural factors on the attitude of First Nations entrepreneurs from two perspectives: cultural dimensions and entrepreneurial attitude, contrasting with the fact that most research on entrepreneurial attitudes has focused on Western, non-original, society. The author highlights the importance of culture for First Nations with a strong sense of self-determination, attachment to the land, affinity with nature and ecosystems, affection for their elders and ancestors, and preservation of their descendants, topics which are in tune with those discussed by Anderson [15] who, in his study among First Nations of Canada, adds the need for the creation of profitable enterprises capable of competing in the capitalist world economy, to which is added a strong motivation for economic self-sufficiency to achieve self-government.

Lindsay [5] concludes that due to the strong roots between these people and their culture, the last one should appear as a contextual variable in the theory of entrepreneurial attitude in First Nations; the findings also show that from an original perspective, business opportunities encompass economic and non-economic objectives, which differentiates First Nations people from businessmen who do not belong to the First Nations, in addition to the complexities of the social and family structure that is brought to the ventures. The author emphasizes the importance of considering these elements, especially when designing public policies related to promoting the entrepreneurial spirit in First Nations. In his opinion, if these do not address the cultural context, they are doomed to fail.

Swinney and Runyan [16] suggest that probably for originating business owners the strength of the owner’s stated purposes and objectives, as well as the owner’s emotional attachment to the business, are more clearly evident than for non-originating business owners. They also evidence the existence of high collectivism over individualism as part of the differences with the non-indigenous world, in addition to that social norms strongly condition the original start-ups. Another difference detected is that indigenous business values are more holistic; the success of the venture is measured in terms of various economic and non-economic dimensions.

Curry et al. [17] conducted an empirical study that examined how indigenous values are embedded in the mission of community development corporations in the First Nations communities in British Columbia, Canada, and how closely these development corporations align with the concepts and mission of social entrepreneurship. From this perspective, social entrepreneurship can be defined as a process that respects traditional cultural elements while getting rid of the old economic ways of doing business, and through creative processes develops new combinations of resources that enable innovation and change, with social, economic, and environmental impacts.

On the other hand, Macpherson et al. [18] present a study on indigenous entrepreneurship in Chile and New Zealand that explores the impact of national political and economic contexts on indigenous entrepreneurial behavior and business models. Evidence from the study supports the argument that the differences between the entrepreneurial activity of the Mapuche in Chile and the Maori in New Zealand are mainly due to the political and economic contexts in which the entrepreneurs operate, the degree of government support, and the connection with the global economy. The study also recommends that if the political and economic contexts for Mapuche entrepreneurs in Chile align more closely with those pursued by Maori entrepreneurs in New Zealand, it is likely to generate a broader range of business models and commercial objectives (beyond the focus on subsistence). Furthermore, it suggests that if this happens, it will need to be managed proactively through government policies or indigenous institutions.

Tretiakov et al. [19] establish that indigenous entrepreneurs depend on indigenous communities to carry out their businesses and often include indigenous culture as a differentiating point in their offers, such as indigenous tourism or indigenous handicrafts. In addition, they operate within the legal and institutional framework associated with the predominant non-indigenous culture. The study focuses on Wayuu entrepreneurs in Colombia and indicates that these indigenous entrepreneurs have been able to integrate values from more than one culture, where the family and the local indigenous community are also essential for the success of their businesses. Thus, indigenous businessmen, by integrating the values of indigenous culture and the predominant culture that surrounds them, become a contribution to their regions.

Mika et al. [20] carried out a qualitative study that compares indigenous entrepreneurship in Chile and New Zealand, defining and characterizing indigenous business ecosystems and their evolution. They found that indigenous business ecosystems evolve with the economic and social environments of their countries, due to an internal imperative towards cultural continuity and cultural resilience. On the other hand, it is highlighted that mature indigenous business ecosystems are associated with higher states of economic and social development, where national political and economic contexts have an impact on indigenous business conduct and business models.

From the cultural point of view of business, Daher et al. [21] carried out a study where they highlight the importance of distinguishing between “being Mapuche” and “Mapuche culture”, since the study participants see their ethnic identity as a dynamic process that combines elements of traditional culture with the Western culture of the business. However, the importance of protecting and recovering Mapuche culture as a coherent whole is also recognized, in part due to the historical marginalization of the community. In this sense, it is argued that Mapuche enterprises offer opportunities to revitalize culture and improve living conditions, while also fostering intercultural relations and cultural recognition. It is concluded that entrepreneurship programs must include aspects of Mapuche culture and take into account the cultural and contextual conditions of the participants. In addition, in the context of the First Nation considered in this study (Mapuche Nation), Soto Hernández et al. [22] explore how innovation in the context of indigenous entrepreneurship occurs as a process of cultural translation rather than homogenization. Through an analysis of two Mapuche companies, it is suggested that innovation focuses on the transformation of daily indigenous practices into valuable products within a market dominated by non-indigenous clients.

Thus, based on the literature review conducted, the main key factors in the sustainable growth of SMEs belonging to the First Nations were found (

Table 1. Key factors in First Nations entrepreneurship.

	Author
Concept	Jaafar et al. [23]	Berkes and Adhikari [24]	Lindsay [5]	Turner et al. [25]	Zhu and Ou [26]	Anderson [15]	Swinney and Runyan [16]	Dana and Anderson [27]	Foley and O’Connor [28]
Collective ownership						*
Alliances between national and non-native entrepreneurs						*
Focused on global markets						*
Motivated by economic self-sufficiency for purposes of self-government		*				*
Motivated to improve the socio-economic conditions of their people		*				*
Motivation for the preservation and strengthening of ethnic factors		*				*
Economic and non-economic objectives			*				*		*
Involves the family and/or community			*
Commitment to the objectives							*
Emotional attachment of the owner to the business							*
Family and community support								*	*
Adapting rather than relying on rules								*
Use of networks with a dominant culture									*
Education as a means of access to social capital									*
Approval of networks by the community									*
Proactivity	*
Ability to learn from mistakes	*
Tolerance to ambiguity	*
Protection of the environment					*
In their endeavor, they make use of his ancestral lands		*
Balance between personal and community integrity and ecosystems				*
Minimize negative impacts on the environment				*
Seek opportunities that distribute benefits in the community				*
Respect the role of traditional authorities in decision-making				*

related to entrepreneurship and Table 2 related to Rapid-Growth firms), which were used as variables for the model developed in this study.

Consequently, Table 1 summarizes the characteristic factors concerning First Nations entrepreneurship found in the literature. As examples, some of the factors present in this table correspond to the following: (a) tolerance to ambiguity, which is the ability to function successfully in uncertain, unpredictable, and unknown environments; (b) use of networks with the dominant culture, which corresponds to the networks of contacts with non-indigenous businessmen to promote the business; (c) adapting rather than relying on rules, which means that instead of relying on the rules established by the non-indigenous market, this factor corresponds to the entrepreneur’s ability to adapt, i.e., to pass from an indigenous cultural context to a non-indigenous one.

Considering that the objective of this research is to study the most relevant factors in the growth of SMEs belonging to the First Nations, the type of enterprises that essentially achieve the highest levels of growth, Rapid-Growth Firms, were studied.

2.2. Rapid-Growth Firms

For Kantis et al. [29], Rapid-Growth firms are those that, as a result of their accelerated growth, leave the world of micro-enterprise in a few years and become competitive SMEs, with the projection of continuing their growth through differentiation and innovation, while at the same time diversifying the productive structure of a region. They are the main creators of new jobs, increased incomes, and a dynamic and competitive economy [30]. Fischer and Reuber [31] help clarify their definition as those whose growth rate increases by at least 20% annually for five consecutive years.

Table 2 summarizes the key factors that characterize Rapid-Growth Firms according to different authors.

Table 2. Key Factors in Rapid-Growth Firms.

	Author
Concept	Cancino et al. [32]	Mueller and Thomas [33]	Kantis and Díaz [34]	Benavente [35]	Hidalgo et al. [36]	Awais and Manzoor [37]	Capelleras and Kantis [38]	Kantis et al. [39]	Barringer et al. [40]	Pšeničny et al. [30]	Segarra and Teruel [41]	Kantis et al. [29]	Kantis et al. [42]	Goedhuys and Sleuwaegen [43]
Share capital				*	*							*	*
Sci-tech platform												*	*	*
Educational system				*	*							*	*	*
Conditions of the complaint												*	*
Culture					*					*		*	*
Social conditions	*				*								*
Entrepreneurial human capital					*								*
Business structure					*		*					*	*
Financing					*					*		*	*	*
Government and public policy	*				*					*		*	*
Team initiated company			*	*			*		*
Mission (focused on growth)			*						*
Commitment to growth	*		*				*		*			*
Emphasis on planning	*								*
Low initial investment			*				*	*	*
Bootstrapping			*				*	*			*	*
Innovation in products and/or services	*								*
International relations	*								*
Export orientation							*
Investment in R&D	*										*
Investment in innovation	*		*								*
Logistics and transport costs														*
Located in metropolitan area	*							*
Creating unique value	*								*
Learning to learn						*
Knowledge of the customer									*
Motivation				*	*			*				*
Role models							*					*
Start before age 35								*
Starting after age 35	*		*	*				*
Contact networks	*					*	*	*
Linking with business mentors												*
Internal control locus		*
Experience as an employee	*			*				*				*
Undergraduate university education	*		*	*			*	*	*
Postgraduate university education				*				*	*
Experience as an entrepreneur	*			*			*	*	*			*
Relevant industry experience	*		*						*			*
Children of entrepreneurs				*	*			*
Male gender			*	*				*
Selectivity in recruitment	*					*			*
Empowerment of employees						*			*
Training and coaching of employees						*			*
Employee development						*			*
Non-economic incentives to employees	*					*			*
Economic incentives to employees	*					*			*
Stock options to employees	*								*

Table 2. Key Factors in Rapid-Growth Firms.

	Author
Concept	Cancino et al. [32]	Mueller and Thomas [33]	Kantis and Díaz [34]	Benavente [35]	Hidalgo et al. [36]	Awais and Manzoor [37]	Capelleras and Kantis [38]	Kantis et al. [39]	Barringer et al. [40]	Pšeničny et al. [30]	Segarra and Teruel [41]	Kantis et al. [29]	Kantis et al. [42]	Goedhuys and Sleuwaegen [43]
Share capital				*	*							*	*
Sci-tech platform												*	*	*
Educational system				*	*							*	*	*
Conditions of the complaint												*	*
Culture					*					*		*	*
Social conditions	*				*								*
Entrepreneurial human capital					*								*
Business structure					*		*					*	*
Financing					*					*		*	*	*
Government and public policy	*				*					*		*	*
Team initiated company			*	*			*		*
Mission (focused on growth)			*						*
Commitment to growth	*		*				*		*			*
Emphasis on planning	*								*
Low initial investment			*				*	*	*
Bootstrapping			*				*	*			*	*
Innovation in products and/or services	*								*
International relations	*								*
Export orientation							*
Investment in R&D	*										*
Investment in innovation	*		*								*
Logistics and transport costs														*
Located in metropolitan area	*							*
Creating unique value	*								*
Learning to learn						*
Knowledge of the customer									*
Motivation				*	*			*				*
Role models							*					*
Start before age 35								*
Starting after age 35	*		*	*				*
Contact networks	*					*	*	*
Linking with business mentors												*
Internal control locus		*
Experience as an employee	*			*				*				*
Undergraduate university education	*		*	*			*	*	*
Postgraduate university education				*				*	*
Experience as an entrepreneur	*			*			*	*	*			*
Relevant industry experience	*		*						*			*
Children of entrepreneurs				*	*			*
Male gender			*	*				*
Selectivity in recruitment	*					*			*
Empowerment of employees						*			*
Training and coaching of employees						*			*
Employee development						*			*
Non-economic incentives to employees	*					*			*
Economic incentives to employees	*					*			*
Stock options to employees	*								*

2.3. Mapuche Context as a Relevant Case Study of the First Nations

As already mentioned, research on the growth of Mapuche businesses is not abundant. However, it is necessary to mention two studies that approach the subject from a descriptive point of view.

In his study, Di-Giminiani [44] accentuates the spirit of seeking forms of autonomy in micro-enterprises in rural Mapuche areas, as it is rooted in political debates on self-government and indigenous rights, coinciding with Anderson [15].

3. Methodology

This research is a quantitative study and of exploratory type. Thus, the methodology to be followed is consistent with this approach [11], for the following reasons:

• Exploratory studies are carried out when the aim is to examine a subject that has not been studied before, or when there are many questions.
• The usefulness of exploratory research lies in the possibility of becoming familiar with unknown phenomena.

Because the scope of this research is exploratory, no hypotheses have been formulated.

Through a survey, it aims to find out the attitude of Mapuche entrepreneurs when they are confronted with statements that show the presence of the factors defined as key factors in the presented undertakings. The aforementioned are listed in Table 1 and Table 2.

The factors are then modeled as causal relationships for the growth of Mapuche businesses and an attempt is made to explain them through the Partial Least Squares algorithm (PLS), with the work being supported by SmartPLS 3 software [45]. The use of data mining techniques was also initially considered, but it was ruled out due to the need for large volumes of data to achieve satisfactory results, an element that is not possible to count on in this experiment.

On the other hand, the use of interviews (survey) as a data collection technique is justified because the attitudes of SME businessmen belonging to First Nations were analyzed based on variables of the same parameter, which in this case was entrepreneurship. The measurement of these variables was taken to numerical values, which were then analyzed through the algorithm described in Section 4. Finally, the result of the data analysis is found in Section 5.

3.1. Partial Least Squares

The Partial Least Square Models (PLS) is a type of Structural Equation Model (SEM) [46,47,48], which in turn corresponds to a set of statistical models that allow the estimation of relationships and effects between multiple variables [49].

PLS uses known elements to determine unknown elements; for this, the algorithm needs to determine the score of the Y (or dependent) constructs, which is used to determine partial regressions from models following the routes defined in the model [47,48].

For illustrative purposes, Figure 1 presents a model of PLS where all the elements of which it is composed can be recognized and commented on:

• The observable variables (or indicators) are in yellow rectangles.
• Latent variables (or constructs) are found in light blue circles.
• Observable variables are variables that can be measured.
• Latent variables that point to observable variables are called reflective variables. These are the model-dependent variables.
• Latent variables that are noted by their indicators are called formative variables. These are the model-independent variables.
• The values that accompany the arrows of the indicator variables toward their constructs are called weights.
• The values that accompany the arrows of the construct toward its indicator variables are called loads.
• The values that accompany the arrows between constructs are called path coefficients. They represent the relevance of the relationship between the variables.
• The set of observable variables and their relationships with their formative constructs is called the Formative Model.
• The set of observable variables and their relationships with their reflective constructs is called the Reflective Model.
• Arrows between latent variables are called paths (or latent variable relationships).
• Variables from which the route departs are called exogenous.
• Variables that receive routes are called endogenous.
• The value within each endogenous latent variable corresponds to the indicator and implies the amount of variance explained by the combination of exogenous latent variables that point to it.
• The set of latent variables (or constructs) and their paths is called the Structural Model.

Partial Least Squares models are characterized by achieving high levels of statistical power or predictive level from small sample sizes, although larger sizes increase their accuracy, while not assuming data distribution. They work with all types of data scales and can work seamlessly with missing values and outliers. In addition, they can work with complex models of many structural relationships [48].

There are known uses of this technique in contexts such as medicine [50], chemistry [51], epidemiology [52], sustainability [53], and mathematics [54], among others.

3.2. Survey

The development of the survey follows the suggestions of Oppenheim [55] and Hernández Sampieri et al. [11], and it is defined as follows:

• Closed questions, i.e., possible answers have been delimited and they must be limited to these possibilities.
• A Likert scale was developed for possible answers for statements related to the variables to be measured, with the following options: “Strongly agree”, “Agree”, “Doesn’t care”, “Disagrees”, “Strongly disagrees” (for which the criterion suggested by Hair et al. [48] was also considered). It was coded from 4 to 0, where “Strongly agree” equals 4 and “Strongly disagree” equals 0.
• In addition, questions were considered with only two options, True and False, taking advantage of the PLS possibility to work with different measuring scales.
• For survey validation, expert judgment was employed. To achieve this, the support of four experts was enlisted, selected for their experience in the study and development of public policies for innovation and entrepreneurship: (1) a Ph.D. with expertise in entrepreneurship and innovation, (2) a business incubator director, (3) a representative from the public sector in charge of innovation studies and programs, who works for the Government of Chile; (4) an entrepreneur with over 15 years of experience. The study objectives and methodology were explained to them, including a question-by-question review and their relationships with the variables under study. Based on their experience, they suggested improvements to the survey to enhance the comprehensibility of the phenomenon being measured.
• The strategy of asking one question for each variable observed was adopted.
• We worked so that the questions fulfilled the following characteristics:

  °

  Clear, precise, and understandable questions.

  °

  Brief questions.

  °

  Use of simple vocabulary

  °

  Avoidance of any questions that might seem uncomfortable.

  °

  Precise questions in terms of the aspect to be measured.

  °

  Avoidance of inducing responses.

4. Model Development

4.1. Selection of the Sample

4.1.1. Description of the Sample to Be Studied

The sample studied includes business owners who are also members of the Mapuche nation, who have initiated activities, applied for a program to support entrepreneurship, and/or belong to a business association of entrepreneurs.

As of the date of this study, there are no registries on the number of Mapuche entrepreneurs in Chile, and this factor generates the difficulty of not being able to count with accurate data the size of the population. Considering this, the decision was made to use the only information that is available and that corresponds to the data provided by CORFO (CORFO (Corporación de Fomento de la Producción) is the Chilean government agency under the Ministry of Economy, Development, and Tourism in charge of supporting entrepreneurship, innovation, and competitiveness in the country): 83 entrepreneurs.

4.1.2. The Sample

For the selection of the sample size, two criteria were used, the first consisting of the minimum data required to work with the PLS algorithm and consisting of the rule indicating the highest value among the following [56]:

• Ten times the number of relationships of the latent variable or construct with the most connections.
• Ten times the number of causal relationships of the latent variable with more structural pathways.

For PLS models, the largest number of connections of a construct is expected to be three, so the minimum sample size is 30.

The second criterion used is statistical, for which Equations (1) and (2) are shown.

$$n=\frac{z_{\propto /2}^2·p·q·N}{d^2·\left(N-1\right)+z_{\propto /2}^2·p·q}$$

(1)

where:

• N: Population size, in this case, 83 entrepreneurs;
• $z_{\alpha /2}$: For a confidence level of 90%, 1.65;
• d: 10% error;
• p = q = 0.5.

$$n=\frac{{1.65}^2\times 0.5\times 0.5\times 83}{{0.1}^2\times \left(83-1\right)+{1.65}^2\times 0.5\times 0.5}=37.64$$

(2)

These values are consistent with the sample size of 37 respondents. This meets the statistical criterion of the PLS 10 rules; in addition, the 37 respondents correspond to 45% of the population, which is considered a good sample [57,58].

The individuals in the sample, considered randomly, were contacted through the Mapuche Chamber of Commerce of La Araucanía, the Mapuche Chamber of Commerce of Santiago, and CORFO users in the Los Ríos Region, as well as contacts from the author of this research; the interviewees were asked to comply with the characteristics that define the population.

There was no discrimination by sales or profit amounts, gender, location (rural or urban), economic activity, professional training, or any other variable, to make the sample as representative as possible.

4.2. Data Collection

To collect the data, three channels of communication with respondents were established: face-to-face interviews, online surveys, or telephone interviews. Their contact information was provided by the business organizations mentioned in the previous section. For the face-to-face interviews, the researchers traveled to the La Araucanía Region for a week, where Mapuche entrepreneurs were visited to talk about this study and then conduct the interviews personally, following all the ancestral protocols and codes. The telephone interviews consisted of a conversation—also based on the Mapuche codes and protocols—providing context for the study and its usefulness, as well as how the interviewee had been contacted. Those who responded to online surveys were contacted beforehand by telephone to explain to them the evaluation tool to be applied and, as in the previous interviewing methods, according to the traditional and ancestral codes and protocols of the Mapuche people. This procedure made it possible to ensure that, regardless of the channel of communication used to interview (face-to-face, online, or by telephone), the interview was the same, that is, the same questions, and different ways of applying the questionnaire. The set of questions applied can be found in Appendix A.

Out of the 37 respondents, 15 responded to the online questionnaire, 12 were interviewed face-to-face, and 10 were interviewed by phone.

In addition, no pre-selection was made that could guide the results in any direction, and all those who had the availability to participate in the study were interviewed.

4.3. Description of Interviewees

The interviewees are, to a greater extent, male, corresponding to 64.86% of the sample, as compared to 35.14% women. The same proportion is maintained concerning its population, with 64.86% urban versus 35.14% rural. The territorial distribution includes 40.54% in the Araucanía Region, 13.51% in the Biobío Region, 13.51% in the Los Ríos Region, 27.3% in the Metropolitan Region, and 5.4% in other regions. Schooling is heterogeneous, with a marked presence of university professionals: 2.7% only have primary education, 21.62% have reached secondary education, 16.22% have postgraduate studies, 13.51% have professional technical studies, 43.24% have a university education, and 2.7% have postgraduate studies in addition to their university education. Regarding the economic sector, the sample of interviewees encompasses a diverse range of industries, with 37 interviewees representing 27 different economic sectors. They began their entrepreneurship mostly between the ages of 26 and 40, an age that could explain why most of them have previous experiences as entrepreneurs, although not necessarily around their current entrepreneurship. Most of them do not belong to families of parents with independent economic activity. Their main motivations for entrepreneurship are positive, with a desire for personal fulfillment and being their own boss; motivations based on negative factors (being unemployed) were not frequent. The variables consulted and their results can be found in

Table 3. Descriptive variables of the interviewees.

Variable	Categories	Percent (%)
Gender	Male	64.86%
	Female	35.14%
Demographical distribution	Urban	64.86%
	Rural	35.14%
Territorial distribution	La Araucanía	40.54%
	Biobío	13.51%
	Los Ríos	13.51%
	Santiago	27.3%
	Others	5.4%
Schooling	Postgraduate education	16.22%
	University education	43.24%
	Technical studies	13.51%
	Other studies	2.7%
	Secondary education	21.62%
	Primary education	2.7%

.

A noteworthy fact is that 100% declare to protect the environment and almost 90% minimize their impact on the environment; 70% involve their family in the venture. More than 86% are concerned about looking for business opportunities that benefit their community (understanding community not only as the concept of the rural community but also the set of people with whom entrepreneurs surround themselves daily). Almost 90% of those interviewed maintained a balance between their personal integrity, their relationship with their community, and the different ecosystems present in the area, through sustainable development, in a clear reference to the concept of Küme Mongen (“Good Life” in Mapuche language). In addition, 95% state that when analyzing business opportunities, they consider economic and non-economic objectives; 92% are in favor of part of their motivation for entrepreneurship being to control resources for self-determination purposes. These latest data give a clear perspective that Mapuche entrepreneurs are different from those of Chilean society, have other motivations, and have more comprehensive objectives. This perspective is in line with Anderson [15], Foley and O’Connor [28], Lindsay [5], and Swinney and Runyan [16] on First Nations entrepreneurship.

4.4. Definition of the Dependent Variable

In the present investigation, the factors that generate the growth of Mapuche companies are sought; the growth is measured mainly in terms of their sales income and to a lesser extent by their growth in size (employment). To this end, the questionnaire incorporated questions regarding their sales income for the first and last year. The number of employees in these periods was also considered. This led to the creation of three growth rates, which are detailed in Equations (3) and (4) and

Table 4. Summary of conglomeration.

Conglomerate	Members	Percentage
1	5	13.51
2	8	21.62
3	11	29.73
4	13	35.14

and

Table 5. Descriptive statistics for Growth Rate by Sales and Growth Rate in Size.

	Growth Rate by Sales	Growth Rate in Size
Average	0.308727	0.238068
Standard division	0.431265	0.675961
Coefficient of variation	139.691%	283.936%
Minimum	−0.119527	0
Maximum	1.47574	4
Range	1.59526	4
Standard bias	3.55783	12.7219
Standard kurtosis	1.25125	35.1776

:

• Growth rate by sales:

$$\frac{\mathrm{Revenue} \mathrm{Sales} \mathrm{In} \mathrm{Last} \mathrm{Year}-\mathrm{Revenue} \mathrm{Sales} \mathrm{In} \mathrm{First} \mathrm{Year}}{\mathrm{Age} \mathrm{Of} \mathrm{The} \mathrm{Venture} \mathrm{In} \mathrm{Years}}$$

(3)

2.

Growth rate in size:

$$\frac{\mathrm{Number} \mathrm{Of} \mathrm{Employees} \mathrm{In} \mathrm{Last} \mathrm{Year}-\mathrm{Number} \mathrm{Of} \mathrm{Employees} \mathrm{In} \mathrm{First} \mathrm{Year}}{\mathrm{Age} \mathrm{Of} \mathrm{The} \mathrm{Venture} \mathrm{In} \mathrm{Years}}$$

(4)

3.

Due to the diversity of economic activities, and the age of the companies, among others, there is a great disparity in the income from sales between the enterprises, which brings about a high variability. To mitigate the effect of the high variability of the data, it was decided to stratify the enterprises based on their annual income, generating four conglomerates with the following criteria:

• Number of cases completed: 37;
• Conglomeration method: k-Average;
• Distance metrics: Euclidian;
• Conglomeration: observations;
• Standardize: yes.

The conglomerates are listed in Table 4.

Table 5 shows the summary statistics for Growth Rate by Sales and Growth Rate in Size. It includes measures of central tendency, measures of variability, and measures of shape. The bias and standardized kurtosis are of particular interest here, which can be used to determine whether the sample comes from a normal distribution. Values of these statistics outside the range of −2 to +2 indicate significant deviations from normality, which would tend to invalidate any statistical test concerning the standard deviation. In this case, only except for the standardized kurtosis of Growth Rate by Sales, parameter values are not within the expected range for data from a normal distribution, which is consistent with the scatter plot in Figure 2. Where the high concentration is at low values, it suggests an F-Fisher distribution.

4.5. Proposed Model

Based on the identification of key factors in the development of dynamic entrepreneurship described in Table 1, as well as in First Nations entrepreneurship summarized in Table 2, the set of latent variables (or constructs) and observable variables (or indicators) were defined. Interviewees were consulted on the observable variables that form part of the analysis.

4.5.1. Evaluation of the Reflective Measurement Model

A PLS model should be constructed initially from the knowledge provided by bibliographic sources and the criteria coming from the researcher’s experience [46,47,48]. Determining the sequence of the constructs is not an easy task; as shown by Hair et al. [48], in large and complex models (such as this one) it is common for researchers to adjust the sequence several times until they find the right one. In this research, it was necessary to iterate the process of designing a causal relationships model and its complete evaluation on several occasions.

After iteratively running several models (with almost 40 variables involved), the first debugged model shown in Figure 3 displays the main constructs that explain the growth of the companies surveyed and their relationship with the observable variables.

A good reflective model must be valid and reliable [48], with its validity is determined through the following:

• Cronbach’s Alpha measures the internal consistency and provides a reliability value based on the intercorrelations of the observed variables of the reflective construct. It has the disadvantage of assuming that all indicators are equally reliable; on the other hand, PLS favors reflective indicators according to their individual reliability, so the following indicator is used in combination.
• Composite Reliability Indicator

$$\mathrm{Composite} \mathrm{Reliability}=\frac{{\left({{\displaystyle \sum }}_i{l}_i\right)}^2}{{\left({{\displaystyle \sum }}_i{l}_i\right)}^2+{{\displaystyle \sum }}_{i}var\left(e_i\right)}$$

(5)

Values between 0.7 and 0.9 are considered satisfactory for both parameters, although for exploratory investigations a value of 0.6 or higher is acceptable [48,59].

For both indicators,

Table 6. First evaluation of the reflective measurement model.

	Cronbach’s Alpha	Composite Reliability	Average Variance Extracted
Current value	0.776	0.930	0.705
Minimum expected	0.7	0.7	0.7

presents the results obtained after the complete debugging process in the row “Current value”.

Discriminant Validity is the extent to which a construct is significantly different and captures a phenomenon that is not present in any of the other constructs of the model. There are two ways to measure it: the first is to review all the loads of the construct and contrast them with the other route coefficients of the construct, and the loads of the indicators must be greater than any of the other route coefficients, a fact that is met in this case [48,59]. The second criterion is that of Fornell and Larcker [60], who postulate that any construct should share more variance with its indicators than with any other construct in the model, which in practical terms implies that in all cases inequality is met: $\sqrt{AVE\left(y_i\right)}>Corr\left(y_n,y_i\right)$, where $\sqrt{AVE\left(y_i\right)}=0.84$ is greater than the correlation between the latent endogenous variable SME Growth. The resulting values are as follows: Collaboration With Close Partners (0.325); Commitment to Growth (0.537); Constant Learning (0.268); Focus on Differentiation (0.220); Focus on Innovation (0.598); Focus on Management (0.361); Learning Skills in Entrepreneurship (0.417); Located in Metropolitan Area (−0.300); Previous Experience (0.127); and Third-party Collaboration (0.498).

4.5.2. Evaluation of the Formative Measurement Model

Unlike reflective indicators, training indicators are not expected to be correlated, as they can cause methodological and interpretation problems; this phenomenon is called collinearity, which implies that one indicator is a linear combination of another. The cases of collinearity were solved in previous depurations.

The VIF (variance inflation factor) is a measure that indicates the magnitude of collinearity among variables in a statistical model. The VIF is commonly used in the context of regression analysis and Partial Least Squares (PLS) models. In PLS, the VIF is calculated for each predictor variable or independent variable in the model, for both the Formative Model (Inner Model) and the Reflective Model (Outer Model). The VIF is a measure of the correlation between each independent variable and the other independent variables in the model and can therefore be used as a measure of collinearity among these variables. A high VIF indicates that the corresponding variable has a high correlation with at least one of the other predictor variables in the model, suggesting that this variable may not be very informative to explain the variance in the response variable and may be causing instability in the model. In general, a VIF equal to 1 indicates that there is no collinearity in the variable, while a VIF greater than 5 indicates that the variable may have collinearity problems and should be examined more closely [48,61,62]. Consequently,

Table 7. First evaluation of collinearity.

Latent Variable	Commitment to Growth	SME Growth	Third-Party Collaboration
Collaboration with Close Partners			1.127
Commitment to Growth		1.032
Constant Learning		1.061
Focus on Differentiation	1.020
Focus on Innovation			1.095
Focus on Management	1.483
Learning Skills in Entrepreneurship	1.480
Located in Metropolitan Area		1.072
Previous Experience			1.067
Third-party Collaboration	1.134

shows the VIF values for the collinearity analysis of the variables in the model.

Low weight and loading values are observed in the Outer Model, which could imply considering these formative variables as candidates to be eliminated from the model due to their low relevance, based on what is indicated by Hair et al. [48]. However, it was decided to keep them in the model to first consider the analysis result presented in the next step.

The significance and relevance of the weights consider the contribution of each observable variable to its corresponding latent variable, which can be established through the bootstrapping process, which provides the t-values (Student’s T distribution) for each of the weights and path coefficients. In general, if the number of observations exceeds 30, the T distribution can be closely approximated by the normal (Gaussian) distribution. In such cases, normal quantiles may be employed to establish critical t-values (or theoretical t-values) for hypothesis testing [48]. The resulting values were expected to be higher than the requirement for the model; in this case, with $\propto =0.1$, the t-values are expected to be higher than $z_{\propto /2}=1.65$. Bootstrap analysis, in summary, aims to determine the relevance of each reflective model variable by assessing their statistical significance and non-zero distinctiveness, through comparison with t-values.

Figure 4 shows the results coming from the previous analyses. As observed, all parameters met the minimum value established to be considered significant and relevant in the predictive model.

Additionally, instead of reporting and eliminating variables that fail to reach a significant level, these variables are evaluated to validate if they are within the “bootstrap confidence interval”, which provides additional information about the stability of the estimated coefficient. The interval is constructed based on the standard error obtained from the bootstrapping process for each of the variables (as shown in Equation (6)), although only the formative model variables that have not reached the significance level for the alpha value (α) will be evaluated. If the value of zero is within this interval, it is possible to eliminate this variable from the model due to its low level of significance, as it could eventually reach a value of zero [48].

$$[w_1-z_{1-\raisebox{1ex}{$\propto $}\!\left/ \!\raisebox{-1ex}{$2$}\right.}·s{e}_{w_1}^{\ast }; w_1+z_{1-\raisebox{1ex}{$\propto $}\!\left/ \!\raisebox{-1ex}{$2$}\right.}·s{e}_{w_1}^{\ast }]$$

(6)

Accordingly, the following variables were then eliminated from the model:

• Logistics;
• Adaptative Business Practice;
• Innovation-Focused Production/Service Investment;
• Previous Employment Inspiration;
• Community Involvement and Support;
• Multiple-Ownership Business;
• Mapuche and Non-Mapuche Business Partnerships;
• Schooling;
• Learning from a Mentor.

The whole process was repeated for two more iterations until all the unnecessary variables were removed from the model. Consequently, Figure 5 shows the already refined model, while

Table 8. Final evaluation of the reflective measurement model.

	Cronbach’s Alpha	Composite Reliability	Average Variance Extracted
Current value	0.776	0.942	0.7
Minimum expected	0.7	0.7	0.7

shows the new evaluation of its reflective model.

It is possible to observe that all the values analyzed equal or exceed the minimum values expected for the validity of the reflective model.

Again following the criterium of Fornell and Larcker [60], it is necessary to compare if $\sqrt{AVE\left(y_i\right)}>Corr\left(y_n,y_i\right)$, where $\sqrt{AVE\left(y_i\right)}=0.809$ is greater than the correlation between the latent endogenous variable SME Growth, and the resulting values are as follows: Commitment to Growth (0.553); Constant Learning (0.260); Focus on Innovation (0.618); Focus on Management (0.372); Learning Skills in Entrepreneurship (0.426); Located in Metropolitan Area (−0.307); and Third-party Collaboration (0.522).

It is now necessary to evaluate the collinearity again, for which the VIF values obtained using the SmartPLS software are analyzed, as shown in

Table 9. Final evaluation of collinearity.

Latent Variable	Commitment to Growth	SME Growth	Third-Party Collaboration
Commitment to Growth		1.009
Constant Learning		1.021
Focus on Innovation			1.000
Focus on Management	1.491
Learning Skills in Entrepreneurship	1.556
Located in Metropolitan Area		1.030
Third-party Collaboration	1.312

.

All the values obtained are close to 1 and far from the critical value of 5, so it is possible to consider that the model complies with not having collinearity.

The significance and relevance of the weights are the next step in the evaluation of the proposed model. In this case, a model was reached that meets the load values required, and those that did not passed the bootstrapping confidence interval validity test, that is, the zero value is not within the potential values that could be achieved. The results obtained from the bootstrapping algorithm are shown in Figure 6.

4.5.3. Evaluation of the Structural Model

The first step in this evaluation is to test the significance and relevance of the route coefficients in the same way as with weights. The results are also in Figure 6. The criterion is the same as when evaluating the loads, where it is expected that for them to contribute to the model, they must be significantly different from zero. Thus, this criterion was applied with a significance level of 10%, where the only path coefficient that does not meet it is the one that goes from Focus on Management to Commitment to Growth; however, its confidence interval [0.018, 0.503] does not consider zero.

The coefficients of determination ($R^2$) show the predictive relevance of each of the latent variables of the model and are calculated as the square correlation between the current endogenous constructs and the predicted values. In other words, they represent the combined effect of exogenous variables on the endogenous variable. The value is provided by the software SmartPLS [45] for each algorithm run. The value of each $R^2$ is expected to be greater than or equal to 0.2. By the results displayed, the model can explain each of the endogenous variables with the required sufficiency.

Table 10. $R^2$ values for the latent variables of the model.

Latent Variable	${\mathit{R}}^2$	${\mathit{R}}^2 \mathit{A}\mathit{d}\mathit{j}\mathit{u}\mathit{s}\mathit{t}\mathit{e}\mathit{d}$
Commitment to Growth	0.466	0.417
SME Growth	0.468	0.420
Third-party Collaboration	0.377	0.359

shows the values for $R^2$.

• In addition to the coefficient of determination, $R^2$, it is possible to evaluate the size of the effect, $f^2$, which represents the effect of the change in the values of $R^2$ (Equation (7)) if any of the exogenous variables of the model are not found, and the values of 0.02, 0.15, and 0.35 reflect small, medium, or large effects, respectively, of an exogenous construct. A good quality of the values was obtained, except for the relationship between Focus on Management to Commitment to Growth and from Third-party Collaboration to Commitment to Growth, whose values are below what is considered an average effect, although further away from what is considered a small effect, as shown in

  Table 11. The $f^2$ values for the refined model.

  Latent Variable	Commitment to Growth	SME Growth	Third-Party Collaboration
  Commitment to Growth		0.524
  Constant Learning		0.181
  Focus on Innovation			0.605
  Focus on Management	0.062
  Learning Skills in Entrepreneurship	0.157
  Located in Metropolitan Area		0.168
  Third-party Collaboration	0.097

  .

$$f^2=\frac{R_{included}^2-R_{excluded}^2}{1-R_{included}^2}$$

(7)

5. Discussion of Results

Firstly, it is important to note that, as a study based on data collected from a survey measuring the perceptions of entrepreneurs belonging to a specific First Nation located in South America (the Mapuche people), the results may not necessarily be generalizable to other First Nations. However, the findings of this study can be a valuable initial input to trigger other similar studies that specifically focus on the particularities and peculiarities of different First Nations around the world. In terms of the implications for practitioners, making the necessary adjustments associated with different contexts (geographical location, culture, etc.), the results of this research could be beneficial to determine the main factors that define entrepreneurship and innovation in SMEs that belong to the First Nations, and based on this, to develop government plans that allow strengthening public policies that include the perspective of indigenous peoples.

Likewise, it is observed that the model manages to explain 46.8% of the latent variable Commitment to Growth; this implies that the combined effect of the exogenous variables with which it is related provides a very good predictive value.

The most determining latent variable is Commitment to Growth, whose score in its route coefficient towards the dependent variable is 0.531. Constant Learning is the second most relevant factor with 0.314. In contrast, there is a variable with a negative influence on the dependent variable, which is Located in Metropolitan Area, with −0.303. The negative relation implies an inverse influence on the dependent variable.

These data help to understand the phenomenon of growth in Mapuche companies, determined mainly by the Commitment to Growth, an important element that should be a decision made by the management of the organization involving the entire work team, which involves leadership, persuasive power, and a lot of teamwork, so that the decision to grow is a real commitment assumed by the team.

This Commitment to Growth is in turn influenced by other endogenous variables, such as Learning Skills in Entrepreneurship, which involves constant learning of the necessary skills to develop successful businesses, as well as by Focus on Management, which relates to elements of classical management. Additionally, the Commitment to Growth is influenced by the endogenous variable Third-party Collaboration, which includes the influence of exogenous variables related to collaboration with partners, family, and the Mapuche entrepreneur’s community, while also placing a concern on long-term relationships with customers. The Focus on Innovation plays an important role by strongly influencing Third-party Collaboration, which can be explained by the necessary interaction with third parties to achieve good innovation, as innovation must necessarily solve problems not only for oneself but also for others.

Located in a Metropolitan Area is the only factor of inverse relation with growth, which is equivalent to saying that the enterprises that show a higher level of growth are far from the metropolitan areas. It is necessary to study in greater depth the factor of location concerning metropolitan areas, even more so considering the relevance of the factor according to specialists [32,39].

6. Conclusions

At first glance, it may seem that entrepreneurship, innovation, and economic development are alien to First Nations knowledge. However, in the case of First Nations located in South America before facing foreign military invasions, it was possible to find rich nations with great economic development, especially in the livestock and textile industries. Witnesses to this period of prosperity, textiles, and silver jewelry can still be appreciated. Economic development has been a constant concern throughout the country’s history.

Understanding that economic development is the result of individual and collective efforts, the aim is to create a model that explains the most influential factors that have led SMEs from First Nations to achieve higher levels of growth and to identify what differentiates them from those who do not achieve the same level. This can be useful for entrepreneurs who have achieved advanced levels of growth to empower themselves even more, for entrepreneurs who have found it more difficult to grow their businesses, to know which factors to focus their efforts on, and for those who wish to start a business, to be clear from the outset about which elements have been important for entrepreneurs with similar characteristics to them in achieving their goals.

The proposed model provides insight into the most relevant variables in the growth of SMEs from First Nations based on the case study of the Mapuche people, where growth is measured both in terms of their sales revenue and their size in terms of the number of employees. The variables considered were obtained after a wide review of the theoretical framework and state of the art in undertakings, especially dynamic undertakings, as they are the ones with the highest growth rates.

This study highlights how important this model can be for the growth of a business, the leadership, and the commitment of the entire team to generate accelerated growth, especially considering that many of these companies started with very little and in disadvantageous social situations. All of this is enhanced by the entrepreneurial skills cultivated by the entrepreneurs, the long-lasting relationships with the clients, and the innovation, in addition to lifelong learning, whether with formal studies or other alternatives, to which these entrepreneurs submit themselves and the decision of collaboration between Mapuche entrepreneurs, but without prejudice to associate with other types of entrepreneurs.

This study is expected to have contributed to the knowledge of the First Nations business world and entrepreneurship for sustainable development from the perspective of the Mapuche people, understanding that despite being part of Chilean and Argentine society, they are another people, with their own values, ways of seeing the world and objectives, and characteristics that affect all areas, including business. The aim was to contribute to the knowledge of the most relevant factors so that other First Nations can take the path of entrepreneurship, as well as to enable those who have already started to improve the performance of their businesses. Specifically, this research showed that the main factors that define the growth of entrepreneurship and innovation in SMEs that belong to First Nations are Constant Learning and Commitment to Growth, where the last one is influenced by Learning Skills in Entrepreneurship, Focus on Management, and Third-party Collaboration. In contrast, the variable Located in Metropolitan Area was the only factor of inverse relation with the First Nations SME growth, showing that the enterprises with a higher level of growth might not need to be placed in metropolitan areas, a phenomenon that should be analyzed in future research.

At present, public policies on entrepreneurship do not consistently consider the development of indigenous entrepreneurship, where indigenous entrepreneurs continue to work in environments of discriminatory practice and marginal empowerment [63], and support for the indigenous economy can be seen just as a token gesture [18]. In this sense, although there are attempts (such as the Indigenous Development and Promotion Committee, implemented by the Chilean government), they have small budgets and support programs, especially when compared to those for the non-First Nations people, and are focused on rural communities with small businesses. More recently, there have been a few programs for sustainable entrepreneurship, although they are not linked to the development of First Nations entrepreneurship. This is paradoxical, given that an important number of First Nations people have professional training and, as this study has shown, are dedicated to multiple economic sectors. Consequently, it is recommended to carry out studies to diagnose the First Nations SME problems and promote the creation of support, training, and financing programs for First Nations entrepreneurs. Each initiative must be conducted with the understanding that these entrepreneurs are part of different people, operate under different values, have broader goals, and do not compete on equal terms with the rest of society. If the cultural context is not considered, any effort to support First Nations businesses will fail. It is essential to carry out research with these objectives in mind. Similarly, from the point of view of public policies, it is proposed to create government programs that promote cooperation between First Nations SMEs and larger and more experienced non-First Nations companies to generate learning in terms of processes, good practices, and know-how to support the further acceleration of the growth of First Nations firms, learning from those who have already made this journey.

In terms of limitations and future research, the main limitation of the present study is that the results may not necessarily be extrapolated to other First Nations and indigenous people. However, its findings can be a valuable initial input to trigger other similar studies that specifically focus on the realities of different First Nations around the world and, after some adjustments related to different contexts (geography, culture, etc.), the results and methodology of this research could be the basis to determine other factors that also may define entrepreneurship and innovation in the First Nations SMEs.