Validation of a participant selection method within a mixed sequential research design for case studies of sustainable supply chains*

: is research addresses the scarcity of literature on participant selection for the qualitative phase in a mixed Sequential Explanatory Design (DEXPLIS) in the eld of sustainable supply chain management (SSCM). e Pathway participant selection method is applied and validated within DEXPLIS to investigate the inuence of integration with secondary stakeholders on the implementation of advanced SSCM practices in Colombian Small and Medium-sized Enterprises (SMEs). e Pathway method selects “precise” cases (SMEs) with a greater inuence of the independent variable on the dependent variable. Experts validate and prioritize the selected cases based on dened criteria. e results demonstrate the applicability and relevance of DEXPLIS and Pathway for SSCM studies.

Introduction e incorporation of environmental concerns into supply chain management has been addressed in Green Supply Chain Management (GSCM) studies. With the additional incorporation of the social dimension of sustainability, this eld has been extended to Sustainable Supply Chain Management (SSCM), with a notable growth in SSCM-published articles since 2007 (Ansari & Kant, 2017). However, since SSCM can still be considered an emerging theory, appropriate methodological tools to address its interdisciplinary nature have yet to be fully developed (Min & Kim, 2012).
Previous works in the literature show the importance of multidisciplinary or interdisciplinary approaches in SSCM (Bag & Anand, 2016;D'Eusanio et al., 2019;Kaufman & Ülkü, 2018). Hence, there is a need to study SSCM problems with different tools from those commonly used in quantitative research (e.g., optimization), such as case studies (Gerring, 2017), which are complementary research approaches (Dubey et al., 2017a). In fact, case study research in SSCM has intensied recently (e.g., Prasad et al., 2018;Zhou et al., 2018).
Several authors suggest that the case study is one of the fundamental methods for research in GSCM or SSCM. For instance, Srivastava (2007) and Malviya and Kan (2015) propose the case study as one of the approaches to be used in GSCM. Similarly, Carter and Easton (2011) and Seuring (2011) suggest case study research as a fundamental approach for SSCM, supported by a broad body of literature (Khalid et al., 2015;Singh & Trivedi, 2016;Winter & Knemeyer, 2013).
Case studies can be immersed in several types of research designs, such as mixed designs (Guetterman et al., 2015) and, especially for the purposes of this paper, explanatory sequential designs (ESD) (Creswell & Plano Clark, 2017). In this type of mixed research design, knowing which participants should be involved in the study and how they are selected is relevant. However, few studies explain this process with some level of detail (e.g., Ivankova et al., 2006).
According to Ansari and Kant (2017), one of the most used methodologies in articles on SSCM is the case study (with 100 articles out of 286 analyzed), followed by conceptual models, survey-based studies, and mathematical modeling. Most case studies have no previous research hypothesis because they are qualitative exploratory or descriptive research (Levy, 2008;Yin, 2017), whose purpose is generally to use their results as a basis for future research (Ansari & Kant, 2017). In this regard, Seuring (2008) suggests that applying case studies in SSCM research oen lacks methodological rigor.
SSCM research that considers a mixed research design or empirical triangulation (e.g., data collection and analysis by qualitative and quantitative methods in the same study) is minimal (4,20 %) (Ansari & Kant, 2017). Likewise, the use of the triangulation approach in GSCM studies is limited; therefore, it is an area that requires signicant attention (Dubey et al., 2017b). ere is a growing need to apply mixed research methodologies (e.g., a mixture of case studies with analytical methods) that increase the advantages of research tools (Min & Kim, 2012) and encompass interdisciplinary sustainable supply chain approaches (Kaufman & Ülkü, 2018).
In response to the recognition of the need for more guidelines on how researchers should select cases within an ESD (i.e., Ivankova et al., 2006;Ivankova & Stick, 2007), the literature has been recently exploring the methodological aspects of ESD. For example, McCrudden and McTigue (2019) integrated both quantitative and qualitative approaches at different levels (method, study, interpretation, and reporting) while studying the judgments made by adolescents regarding scientic arguments related to beliefs and the underlying rationales for those judgments. In a similar vein, Draucker et al. (2020) integrated both approaches into a ve-step process to study lung cancer screening.
Other studies expanded the traditional two phases of the explanatory sequential designs (i.e., quantitative and qualitative) and added a third phase. Framed in the paradigm of community-based participatory research, Maleku et al. (2021) added a third phase in which voices of African refugee communities were considered to start a cycle of dissemination and action that brings about transformation in those communities. Similarly, Haynes-Brown (2022) added a phase to draw meta-inferences from the quantitative and qualitative ndings. In this way, the authors integrated this new phase into a theoretical model to better understand how beliefs inuence the utilization of technology in the classroom.
As these previous examples have shown, an opportunity to apply the exploratory sequential design in studies of GSCM and SSCM emerges. As mentioned by Ayati et al. (2022), mixed methodologies in supply chain management studies, such as exploratory (e.g., Faisal, 2023) or explanatory (e.g., Angeles et al., 2019) sequential designs, are potentially helpful in gaining insights into the participant selection process. However, in the framework of SSCM studies, the literature on case selection procedures for the qualitative phase within an ESD is scarce. From a methodological perspective, the level of detail provided needs to be increased.
As James et al. (2022) suggest, further research is required to explore the methodological considerations in which the quantitative ndings lead to the selection of qualitative sampling criteria in explanatory sequential designs. Following this advice, this research seeks better specication in the necessary procedures for an ESD to guarantee a more reliable method for participant selection in GSCM and SSCM case studies. In this way, this paper applies and validates a participant selection method in an ESD for Small and Medium-sized Enterprises (SMEs).

eoretical framework and problem statement
In the last two decades, research on sustainable supply chain management has grown and gone from being a fringe topic to one within the mainstream of supply chain management (Allen et al., 2021). However, there are still abundant research opportunities in various branches of SSCM (Carter et al., 2019).
Globally, more than 90 percent of the impacts on natural resources and more than 80 percent of greenhouse gas emissions are produced by supply chain management activities (Bové & Swartz, 2016). If these and other environmental and social issues are not adequately addressed, they can have serious global consequences (Ripple et al., 2020). Furthermore, this can damage a focal company's relationships with its stakeholders, inside or outside the supply chain (Esty & Winston, 2006;Pagell & Wu, 2009).
Sustainable Supply Chain Management (SSCM) was developed to counteract those issues. SSCM is dened as the management of information, materials, and capital ows, having the objectives of the three dimensions of sustainability in mind, through collaboration between companies in the supply chain, without forgetting the requirements of other interested parties (Seuring & Müller, 2008, p. 1700. e most applied organizational theories in SSCM studies are the natural-resource-based view of the rm (RBV), stakeholder theory, institutional theory, and transaction cost theory (Touboulic & Walker, 2015). Carter and Easton (2011) defend the combination of multiple theoretical perspectives to offer original ideas in the eld of SSCM.
It is crucial to note that implementing SSCM or GSCM practices can signicantly and positively affect the companies' performance and sustainability indicators (Hong et al., 2018). Especially the implementation of SSCM practices triggers greater efficiency and innovation (Ageron et al., 2012) and improves economic performance and business transaction costs (Um & Kim, 2018). Furthermore, SSCM social practices can also positively affect competitive advantage and corporate reputation in the market (Chacón-Vargas et al., 2018;Zailani et al., 2012).
However, most companies and their supply chain strategies do not attach the same level of importance to the three dimensions of sustainability (also called triple bottom line -TBL-). Besides, these three dimensions are not fully incorporated into the core business either (Carter & Rogers, 2008). is is mainly because most companies, and researchers on SSCM, have focused on making unsustainable supply chains more sustainable rather than pursuing a "truly" sustainable supply chain management (Pagell & Shevchenko, 2014).
Additionally, many studies in SSCM have neglected smaller companies, some of which could radically rethink their sustainability strategy. One of the critical organizational capabilities to achieve this rethinking is integration with external or non-economic stakeholders, called external secondary stakeholders (ESS). According to various researchers, such integration can inuence the development of social or environmental practices (Aschehoug et al., 2012;Beske et al., 2014;Buysse & Verbeke, 2003;Hart & Dowell, 2011;Hart, 1995), either with a focus on incremental innovation or disruptive innovation (Klewitz & Hansen, 2014). In this sense, disruptive innovation deals with radical practices that break the barriers of incremental innovations, cost reduction, or the search for eco-efficiencies to make a leap from existing practices within the organization that implements them.
A specic problem of recent interest in this literature is analyzing how stakeholders with non-economic objectives are enablers in deploying disruptive or -from now on-'advanced' SSCM practices in SMEs (Pagell & Shevchenko, 2014). For example, an advanced environmental practice is the design of a closed-loop supply chain (circular economy), while an advanced social practice is to guarantee all 'fair trade' conditions in the supply chain (Marshall et al., 2015). More recently, Koberg and Longoni (2019) have shown that collaboration with these third-party actors is emerging in the research on SSCM, focusing on sustainability outcomes in all three sustainability dimensions.
Finding a niche in this relevant eld of research, an application and validation of the participant selection method were conducted in this study. is is achieved through the use of the explanatory sequential design in a case study that investigates how the integration of Colombian SMEs with external secondary stakeholders (ESS) (e.g., NGOs, governmental organizations, the base of the pyramid, and universities) inuences the development of disruptive socio-environmental practices deployed inside and outside of their supply chains. e participant selection method was depicted within a multiple case study of a diagnostic, explanatory type (Gerring, 2017). As previously stated, selecting relevant cases during this rst stage of the mixed study is crucial to allow for a better understanding of the phenomenon under study during the qualitative stage of the mixed research.
On the other hand, based on the analysis of the extant literature on mixed research methodologies (e.g., Creswell & Plano Clark, 2017;Hlebec & Mrzel, 2012;Ivankova & Stick, 2007;Hernández-Sampieri & Mendoza, 2018), it was found that mixed methodologies are not sufficiently detailed in terms of the procedure for participant selection in the explanatory sequential design (ESD). at is, most of this methodological literature does not delve into the issue of how to choose the participants or dene the qualitative sample using data from a previous quantitative phase. Ivankova et al. (2006) suggest that better guidelines should be written on selecting cases within an ESD methodology. Some empirical studies applying ESD use deliberate sampling for the qualitative phase (e.g., Campbell & Profetto-Mcgrath, 2013;Schindler & Burkholder, 2016). For example, these studies chose participants with the highest scores on some variables from the quantitative results but without sufficient methodical treatment.
erefore, this study addresses a signicant research gap within the SSCM academic community by proposing a rigorous method for selecting cases in qualitative studies from quantitative data previously collected in the same mixed research.

Context for approaching the participant selection method
Since it is an area of interest in the literature, the integration between non-economic stakeholders and focal companies in SSCM research needs to be validated in the context of SMEs through a multiple case study in an emerging economy such as the Colombian one. Hence, a mixed research methodology called Explanatory Sequential Design (Participant Selection model) is proposed to address this research problem (Hernández-Sampieri et al., 2014).
e ESD consists of two major phases. e rst phase is the data collection through a quantitative, nonexperimental phase, with an explanatory scope and transversal temporality (Creswell, 2012). e second phase is the collection and evaluation of qualitative data (Creswell & Plano Clark, 2017). According to Creswell and colleagues (Creswell & Creswell, 2018;Creswell et al., 2003;Creswell & Plano Clark, 2017), this is a method characterized by successive stages so that the quantitative data collected and analyzed in the rst stage can be used to improve the qualitative data collection and analysis of the second stage. Figure 1 shows the eight stages involved in this method. e cases are characterized according to certain features of interest related to the problem (Creswell & Plano Clark, 2017). us, SMEs are selected quantitatively because they meet the two related variables of the problem at a high level: the extent of integration with external secondary stakeholders (X) and the extent of deployment of innovative SSCM practices (Y). ey are selected because there is quantitative evidence of greater inuence of X on Y in those companies than in the other surveyed companies. Seawright and Gerring (2008) offer seven techniques for case selection in situations where data are available in a large sample and associated with key variables: typical, diverse, extreme, deviant, inuential, most similar, and most different cases. Based on this typology, the Typical case is considered for this research since it intends to explore the causal mechanisms of the general relationship in depth. Gerring (2017) deepens this case selection technique and develops a more detailed classication. In his classication, a Pathway case is considered the typical case in the framework of a multiple case study of the causal and diagnostic type. In the Pathway case, the apparent impact of the causal relationship X → Y is adjusted to theoretical expectations and is stronger in magnitude. At the same time, background factors (Z) exert a "conservative" bias.
Consequently, a procedure for the participants' selection phase is established. is procedure can be classied as sequential and convenience sampling to achieve comparability (Tashakkori & Teddlie, 2010). Specically, the procedure is a typical case sampling using the Pathway strategy to choose the "ideal" participants for a multiple case study of causal and diagnostic types (Gerring, 2017).
Methodology is section shows how the quantitative and qualitative phases of the explanatory sequential design are conducted. en, the procedure for selecting participants is detailed.

Quantitative phase
Data were collected through an online survey strategy, with a questionnaire based on empirical survey research. Five experts pre-evaluated the survey's content validity, and it was pre-tested in a convenience sample taken from the Directory of Exporting and Importing SMEs in Colombia. Aerward, the number of items in the original questionnaire was reduced from 65 to 50. e two operationalized theoretical dimensions were 'advanced SSCM practices' (Marshall et al., 2015) and 'integration with stakeholders' (Plaza-Úbeda et al., 2010). Subsequently, the nal questionnaire was applied to managers of Colombian SMEs from a sampling frame of 1,300 companies. ese companies, which were involved in knowledge transfer activities with varied stakeholders, were identied through the National Association of Industrial, Administrative, and Production Engineering Students (ANEIAP).
From the collected data, missing cases and outliers were treated accordingly (Hair et al., 2010). en, an Exploratory Factor Analysis (EFA) was performed with the Varimax rotation method. Later, Conrmatory Factor Analysis (CFA) suggested eliminating nine items to achieve better goodness of t. is exercise provided a nal sample of 100 SMEs, corresponding to a 7,7 % response rate. e sample was composed of 60 small and 40 medium-sized companies. In terms of sectors, 43 SMEs belong to the manufacturing sector, 39 to the service sector, and the remaining 18 to other activities.
e EFA validated the multidimensional construct of integration of stakeholders, consisting of three dimensions: knowledge of interested parties and their demands (KNOW), interaction with external actors (INTER), and adaptive behavior (ADA_BEH). ese dimensions are crucial to integrating interested parties in corporate sustainable management.
Specically, the rst variable (KNOW) implies receiving and interpreting information and sustainability expectations from the stakeholder by comparing internal and external knowledge (Aschehoug et al., 2012). e second dimension (INTER) incorporates activities such as meetings with the stakeholders, consulting them before making decisions, and devoting time and resources to the relationship with them (Plaza-Úbeda et al., 2010). e third dimension, adaptive behavior (ADA_BEH), refers to applying changes in the behavior of the company to satisfy the demands of its interest groups (Maignan & Ferrell, 2004), either by adapting policies, strategies, or activities to further the integration with the stakeholder (Plaza-Úbeda et al., 2010).
Advanced sustainable practices (PRAC_ADV) involve a redenition of the supply chain strategy and a more radical than incremental, sustainable innovation. ese practices have been traditionally less studied in the empirical literature (Pagell & Shevchenko, 2014). For this reason, the results were rather different compared to the reference model predicted by Marshall et al. (2015). Consequently, the factors associated with advanced sustainable practices were modied into three factors. e rst one, 'advanced social practices' (ADV_SOC), implies designing practices that not only consider reducing costs or environmental impacts. is factor goes beyond incremental changes in their green strategies and proposes disruptive practices throughout the entire supply chain. e second factor, 'advanced environmental practices' (ADV_ENV), is based on previous literature (Kearins et al., 2010;Klewitz & Hansen, 2014;Marshall et al., 2015;Pagell & Wu, 2009;Rothenberg, 2007) and was operationalized through items that materialize the concepts of circular economy (both for the industrial cycle and the natural cycle), biomimicry, servicizing, and "decommoditization." Finally, the third factor is related to 'product responsibility practices' (PRAC_RP), which are divided into ve key elements for the focal company (Chacón-Vargas et al., 2018): First, establishing quality relationships and a fair price for its clients; second, providing quick and respectful responses to customers' concerns; third, offering customers transparent promotions; fourth, providing complete and accurate information about the company's products or services and their impacts; and the nal key element is striving for total customer satisfaction (Ağan et al., 2016;Govindan et al., 2018;Shaq et al., 2014). Figure 2 classies the latent factors of the integrated model according to whether they are rst or secondorder variables in the model, which then serve as the basis for selecting participants. e operationalized variables shown in Figure 2 serve as input for the participant selection procedure that will be detailed later. Observed variables are not shown in the gure. Factors used in the participant selection procedure Source: Own elaboration.

Qualitative phase
Aer the participant selection procedure (see in the following section), the qualitative phase consisted of a multiple case study (Yin, 2017). e case study was classied as explanatory, causal, and diagnostic. It aimed to investigate the causal mechanisms that connect the relationship of the phenomenon X → Y (Gerring, 2017), where X is the integration with stakeholders with non-economic objectives (INT_STK), and Y is the deployment of advanced sustainable practices (PRAC_ADV). ese advanced practices may involve a redenition of the supply chain sustainability strategy and more radical than incremental, sustainable innovation. e primary data collection tools were semi-structured interviews and nonparticipant observation. e data collected were systematized and studied through detailed content analysis (Saldaña, 2013). e analysis strategy consists of theoretical propositions, and the analysis technique is explanation building (Yin, 2017).

Participant selection method
In the rst place, the chosen method sought to select those SMEs that offer evidence of above-average performance in the study's main variables: 1) the deployment of sustainable advance practices of SSCM, and 2) the effective integration with external secondary stakeholders. e selection of these companies as cases in the case study should offer sufficient evidence within each case and represent importance for the academic community. Moreover, they should prove to be a Pathway case type (Gerring, 2017). In the Pathway case, the result score (Y) is strongly inuenced by the theoretical variable of interest (X), considering other factors (Z). erefore, it should be easier to "observe" the causal mechanisms between X and Y in this case type. e equations proposed by Gerring (2017) served to select the Pathway cases. ese cases have the greater absolute value for the difference between the residual for the minimum specication (Equation (1)) and the residual for the complete specication (Equation (2)), as follows: Using the resulting data from the quantitative phase and aer applying the modications derived from the EFA and CFA, LISREL® was used to calculate the Latent Variable Scores (LVS) for the rst-and secondorder variables (Marsh et al., 2013;Schumacker & Lomax, 2010). e values obtained were standardized (i.e., mean of zero and standard deviation of one). With these data, a column was created for the residuals of the minimum specication using Equation (1). For doing this, the variables dened as Z were: 1) the number of employees (LN_EMP), 2) if the company had any quality or environmental management certications (CERTIF), and 3) its economic sector (SECTOR). e standardized residual of this regression is dened as ZRE_1.
en, another column was created for the residuals of the complete specication using Equation (2). e standardized residual of this second regression is ZRE_2. Finally, a column was created for the Pathway variable using Equation (3) (i.e., the absolute value of the difference between standardized residuals), ordered from the highest to the lowest values and with a threshold greater or equal to 0.4.

Application and validation of the participant selection procedure
Applying the procedure proposed in Gerring (2017), and once the Δs of the residuals from the linear regressions were calculated, a scatter plot comparing the two variables of the model (X and Y) was depicted (Figure 3). Aer that, three additional linear regressions were generated, one for each rst-order dependent variable (ADV_SOC, ADV_ENV, PRAC_RP) vs. the second-order independent variable (INT_STK). e obtained graphs were analyzed to nd those cases that showed higher integration with ESS (INT_STK) and higher development of advanced practices (PRAC_ADV) simultaneously. ese cases were found in the upper right corner of the Scatter Plot (Figure 3). Additionally, cases were selected to exceed the predened Pathway threshold (≥ 0,4). In short, it is a requirement to be preselected that the case should appear 1) in the upper right corner (Figure 3), 2) in the second-order linear regression model, and 3) at least in one of the other three rst-order models. e last column of Table 1 shows the obtained preselected participants from the Pathway procedure. Aer this procedure was conducted, it was considered necessary to verify the ten pre-selected cases by contrasting them with the SME's primary or secondary data sources. Consequently, a report for each SME was written. ose reports have information from their website, social media, and other secondary sources. In addition, rst-hand data were obtained through telephone calls or emails sent to the company to verify or complement the report. No information was found for one company; hence, it was eliminated from the preselected cases.
Despite the detailed participant selection procedure, an expert panel review of the chosen cases also was considered necessary. Donaldson et al. (2013) suggest eight types of methods to select cases where the "consultation of an expert panel" is highlighted (as mentioned in Cimadamore et al., 2002). is method is a way to guarantee sampling validity. Additionally, the method can provide a solid basis for analysis and discussion since judgments come from expert participants (Donaldson et al., 2013). Criteria for the evaluation of preselected cases Note: Likert scale: 6 -Very strongly agree, 5 -Strongly agree, 4 -Agree, 3 -Disagree, 2 -Strongly Disagree, 1 -Very strongly disagree Source: Own elaboration based on Gerring (2017), Brown (2010), Pagell & Shevchenko (2014), Tidd et al. (2005), and Alegre & Chiva (2008).
From the different methods of consultation with experts, the Delphi method is discarded since it requires the experts to be anonymous and several rounds of evaluation (Taylor, 2020). Methods such as multicriteria decision-making or the analytical hierarchy process were also discarded (Triantaphyllou, 2000). Consequently, the team of experts composed of 4 people, including the main researcher and three university professors acquainted with innovation research, jointly designed a simple qualication and weighting procedure in Excel. is procedure is based on two key criteria and six sub-criteria adapted from Gerring (2017) and additional literature. Table 2 shows these criteria and their assigned weight. Additionally, a consensus was reached using a 1-to-6 Likert scale for rating each sub-criterion, where 1 was "very strongly disagree," and 6 was "very strongly agree" (Brown, 2010). Visual model for the explanatory sequential design with the participant selection procedure Source: Own elaboration based on Ivankova et al. (2006) Table 3 synthesizes the explanatory sequential design. It summarizes how the quantitative phase is linked to the qualitative phase through the participant selection procedure detailed in this study. is table is a synthesized model of the methods and techniques used in this work, along with the products of each procedure.
Aer an expert's panel discussion, the threshold for selecting the cases was set to 3,9 or greater, resulting in six cases being selected for the next qualitative phase. Finally, the cases identied with numbers 94, 102, 73, 101, 103, and 22 were selected by the panel of experts following the established procedure. e targeted cases are believed to be of importance to the scientic community, offering enough evidence and facilitating a deep understanding of the phenomenon of interest in the qualitative phase.

Discussion
is study shows how it is possible to apply the Pathway method of selecting participants for so-called typical cases proposed by Gerring (2017) in the research eld of GSCM and SSCM. In addition, it also shows the relevance of extending the techniques and the necessary methods for a thorough selection of cases in an explanatory sequential design. Table 4 compares various sampling or participant selection methods for case studies mentioned within the SSCM theory. Various methods, selection criteria, or forms of sampling are used, with several degrees of systematization and replicability. Sampling methods in SSCM case study literature  Jabbour et al. (2013), and Lee (2009) Interestingly, most of these case studies in SSCM are exploratory or descriptive. erefore, we suggest a different and rigorous alternative method for an explanatory case study. Henceforth, the Pathway method within ESD described and applied in this study is particularly suitable for causal, explanatory, and diagnostic studies (Gerring, 2017).
Next, it is also interesting to compare our adapted and extended method for selecting participants with other sampling approaches used by previous research that also applied the mixed methodology -explanatory sequential design (ESD). Table 5 makes a parallel between these procedures and ours.
As seen in Table 5, some of the studies that apply the ESD mixed methodology use a more intentional and expeditious sampling approach, while other samples are based on cases with high scores obtained from the measurements in the quantitative survey. Perhaps, the most thorough selection method is the one performed by Ivankova and Stick (2007), which selected the participants in two steps, using mean scores within a standard error of the mean and a maximum variation strategy. Our approach also considers high scores but draws rstly from the Pathway method (Gerring, 2017), which is applied to the causal relationship of interest in the study. Furthermore, our procedure for selecting participants goes beyond the Pathway with two additional steps: verication with primary or secondary data sources and an expert panel for nal validation.  Ivankova and Stick (2007), and Schindler and Burkholder (2016) Conclusions is research adapts the Pathway approach suggested by Gerring (2017) and implements a participant selection method within an explicative sequential design followed in a study of sustainable supply chains in Colombia. e method denes how the cases can be selected rigorously from a quantitative phase that proceeds to a later qualitative phase consisting of a multiple case study. Additionally, it is worth emphasizing that the selection method is extended using an expert judging procedure for ltering and validating the cases obtained.
is can be seen as a strength of our method compared to previous SSCM case studies and studies that apply ESD outside of SSCM theory. Another strength lies in the rigor employed when selecting participants, following an ideal procedure for explanatory causal studies, which is also reinforced by validation with the panel of experts.
Once the relevant cases have been selected, the case study will show how the integration with external secondary stakeholders inuences the development of advanced sustainable practices in Colombian SMEs. Although the literature concerning participant selection in the mixed exploratory sequential design is scarce, our study shows that it is possible to incorporate methodological strategies from this area in contemporary GSCM and SSCM research.
e method described in this research allows the selection of participants who meet the pre-established criteria in the literature, which experts then validate. For this reason, these cases are of interest, and the insights gained from their study are important for the academic community, providing enough evidence for the studied relationships.
In the context of the ongoing research in SSCM, the literature suggests applying mixed methods or empirical triangulation, whether in an explanatory, exploratory, or descriptive study. Studying how participants are selected or how the sampling frame is established are vital procedures to ensure greater knowledge transferability or analytic generalization. erefore, methods for rigorously selecting participants in case studies should be rened in future research, especially when framed in a mixed method sequential design. In summary, it is necessary to deepen the study of methods and techniques that guarantee greater credibility and auditability in GSCM and SSCM research.

Ethical considerations
e investigation did not require ethical endorsement of any kind.

Authors' contributions statement
Each of the authors of this article participated in the design and execution of the research, the literature review, and the writing and revision of this article.

Interest conflicts
e authors do not have any type of conict of interest associated with the development of the research or this article.