Social value appraisal: cutting the Gordian knot

Abstract

Purpose

Social initiatives must disclose their results to access support. However, there is no theoretical consensus about how to do it. It is still necessary to understand the value creation in social initiatives because they may or not have economic goals. However, these goals serve to make the social ones feasible. This study aims to cut this Gordian knot by providing measures aligned to the value theory but developed by the social lens. It offers a non-economic-focused approach to dealing with assessment complexity and with multiple stakeholders’ needs for information.

Design/methodology/approach

A consolidation research path is suggested by three composed measures built upon tested and reliable scales. These value measures are discussed through narratives from Portuguese investors and social entrepreneurs in a mixed-methods design. Content analysis and online survey provided data for descriptive statistics, exploratory factor analysis and Cronbach's alpha test.

Findings

The analyses supported the value measures. Thus, they allow an effective way to assess and report the social value created. It also highlighted a potential use in preventive and corrective approaches helpful for several organizations that pursue social goals.

Research limitations/implications

The measures were tested based on social entrepreneurs' opinions. Future studies can include beneficiaries' opinions, allowing comparisons that can help to set more realistic goals and better investment criteria.

Originality/value

The relationship between investors and initiatives can improve, boosting their impact on society. The measures can highlight prioritization choices that influence the way value is created. Hence, they serve as a sensemaking from a holistic standpoint.

Citation

Ciccarino, I.D.M., Rodrigues, S.C.S.F. and Ferreira Da Silva, J. (2024), "Social value appraisal: cutting the Gordian knot", Innovation & Management Review, Vol. 21 No. 1, pp. 2-14. https://doi.org/10.1108/INMR-07-2021-0143

1. Introduction

There are many ways of doing business with a social purpose, which does not mean prioritizing the social goal (e.g. Corporate social responsibility). In addition, the economic goal is also a controversial issue among social initiatives. (Barki, Rodrigues, & Comini, 2020). The emphasis on social goals happens in a spectrum that goes from philanthropy and social movements to corporate social responsibility, crossing all the social economy and beyond it (Defourny & Nyssens, 2010; Salamon & Sokolowski, 2016).

The social economy initiatives, henceforth named SEIs, must seek a sustainable business model to ensure their value proposition continuity through a stable operation (European Commission, 2016; Hlady-Rispal & Servantie, 2018). The social goals should guide the strategic formulation and decision-making (Hlady-Rispal & Servantie, 2018). However, little is known about social strategy (European Commission, 2016; Shaw & de Bruin, 2013) and the social value process (Hlady-Rispal & Servantie, 2018). This study is a step toward strategic understanding by exploring social value creation alternative measures. Responsibility and transparency are fundamental elements of SEIs' legitimacy, especially in accountability and disclosure practices (Clark & Brennan, 2012; IMP - Impact Management Project, World Economic Forum, & Deloitte, 2020). There is no lack of methodologies. Nevertheless, social problems’ complexity, and constraints from context and organizational structure, make systematic evaluation difficult (André, Cho, & Laine, 2018).

Some frailties related to research infrastructure (Lee, Battilana, & Wang, 2014) prevent even basic consensus like establishing what defines an SEI or how the researchers must evaluate it (Hossain, Saleh, & Drennan, 2017; Rawhouser, Cummings, & Newbert, 2019). Meanwhile, SEIs are criticized for failing to report their value (Ormiston & Seymour, 2011), despite the lack of consensual recommendations (Bosma, Schøtt, Terjesen, & Kew, 2016). Moreover, techniques based on the social perspective are underdeveloped (European Commission, 2016; Khan, Yasir, Shah, & Majid, 2021; Rawhouser et al., 2019).

To make research feasible, many authors rely on perspectives derived from different research fields (Mair & Martí, 2006). It can represent a significant theoretical limitation because it is challenging to find representative synthesis from widely dispersed knowledge (Lüdeke-Freund, Carrouxb, Massad, & Breuer, 2018). Also, different research fields should employ different measures to accurately assess their results, since they seek answers to different questions (Venkatraman & Ramanujam, 1986). However, this multidimensionality is an opportunity to connect different approaches and disciplines, promoting academic diversity (Van der Have & Rubalcaba, 2016).

Concerned with this pre-paradigmatic issue (Hossain et al., 2017), this research relies on strategic theory for guidance to set an interpretation structure amenable to replication (Lee et al., 2014; Venkatraman & Ramanujam, 1986). Since social value is a common element in literature (Hossain et al., 2017; Ormiston & Seymour, 2011), research on it can provide consensus (Hlady-Rispal & Servantie, 2018). The proposed measures gauge the social value as the moving target (Snowden, 2002). As a result, the different stakeholders can understand each other, enhancing fundamental synergies for a positive impact on society (Barki et al., 2020). The authors established the theoretical ground by answering the hypotheses with a mixed-method approach to set up the value measures to achieve such a positive impact. The paper evolves by presenting the qualitative and quantitative analyses as a base for discussing theoretical and practical implications.

2. Literature review

2.1 Social value

Hlady-Rispal and Servantie (2018) define social capacity as a distinctive way to put the collective interests first toward meeting a social need, which involves different dynamics of value creation, capture and distribution (Brandenburger & Stuart, 1996). This social capacity allows reaching contexts and problems beyond commercial grasp (Tate & Bals, 2018) with distinct opportunity sources and resource mobilization. Thus, the performance cannot be assessed in a traditional way (Símon-Moya, Revuelto-Taboada, & Ribeiro, 2012), yet literature does not offer generalizable recommendations for measuring and reporting it (European Commission, 2016; Lee et al., 2014; Rawhouser et al., 2019). This theoretical gap increases the challenge to establish measures (Clark & Brennan, 2012; Lumpkin, Moss, Gras, Kato, & Amezcua, 2013).

Observing social effects is also not very enlightening, as there is a myriad of social needs that are met in many ways (Hossain et al., 2017). The social effects created by different initiatives cannot be directly compared (Defourny & Nyssens, 2010; Ormiston & Seymour, 2011). Some authors suggest performing a process analysis (El Ebrashi, 2013; Clark & Brennan, 2012). Although it is a practical and effective way of organizing information, the focus on the process fails both to establish criteria and detail the type of effect created. It can become idiosyncratic and needs complementation (Rawhouser et al., 2019). One can deliver value in the most effective way without the intended beneficiaries realizing the benefit (Clark & Brennan, 2012).

Rawhouser et al., (2019) propose the assessment in absolute and relative terms. The proposal aligns with how researchers study the value in the strategic field (Bowman & Ambrosini, 2000; Lepak, Smith, & Taylor, 2007). However, the construct operationalization is an entirely different matter, considering that the strategic field relies on competitive aspects. Thus, the aim is greater bargaining power to get more significant value appropriation. In contrast, social initiatives play with collaborative rules that are not totally understood (Bacq & Eddleston, 2018; Borzaga, Depedri, & Galera, 2012; Hlady-Rispal & Servantie, 2018).

Clark and Brennan (2012) argue that social assessment should be separated into (1) outputs, (2) outcomes and (3) ideal results with specific indicators for measuring each one. In line with them, this paper distinguishes the creation value process and the impact of that value on society. Researchers have analyzed the literature and noticed a blurred understanding of both concepts (Clark & Brennan, 2012; Hadad & Găucă, 2014). However, they have different meanings. The current research helps to clarify this understanding by focusing on the assessment of social value considering the immediate results of SEI's interventions. The social impact refers to the cumulative value contribution to society, whether concrete (i.e. a measurement) or ideal (i.e. a goal). (Khan et al., 2021; Símon-Moya et al., 2012).

Social impact is the assessment result of the social implications of a given initiative (Ormiston & Seymour, 2011) and can be either positive or negative (Rawhouser et al., 2019). It is a complex concept requiring multidimensional interpretation (i.e. individual, community, geographical and social assessment), according to the assisted social need (Rawhouser et al., 2019). Some impacts can be immediate while others are only possible with continuous and long-term actions (El Ebrashi, 2013). Some are identifiable only if contextualized; others are generic, such as the increase in social welfare and the productivity of social capital (Borzaga et al., 2012). It is often difficult to establish that a particular impact happened exclusively through SEI's action, so they cannot be assessed directly (Borzaga et al., 2012; Ormiston & Seymour, 2011).

It happens especially because social problems and their causes are poorly known and researched (Hadad & Găucă, 2014; Rawhouser et al., 2019), making their treatment difficult (Clark & Brennan, 2012; Lumpking et al., 2013). In general, ideal results are adopted as parameters and function as short-term and long-term goals (Rawhouser et al., 2019). Some social needs are conspicuous, while others depend on the restlessness of people who join social movements to make them evident and stimulate change (e.g. discrimination of race, gender and creed and domestic violence) (Mulgan, 2006). Researchers base most of the methods and guidelines to gauge social outcomes on impact assessment, despite the social value being a more objective and direct outcome. SEIs commit themselves directly to this result (Hlady-Rispal & Servantie, 2018; Khan et al., 2021).

The social value is noticeable by assessing social need and SEI’s direct action on it, being what SEI produces and delivers (Clark & Brennan, 2012; Hadad & Găucă, 2014; Khan et al., 2021). People involved can direct the analysis of the value creation process either from its sources or from its results (Hlady-Rispal & Servantie, 2018). Assuming a systematic view, measuring the value created to ensure that the organization appropriately employs its resources and capabilities to generate the greatest possible social impact. Furthermore, to clarify that society is better off with the SEI’s presence than without it (Tate & Bals, 2018). Therefore, it is a way to keep the social impact on track. Through social value measures, it is possible to check the prioritization of the elements that influence value creation and measure progress against goals creating paths that can lead to a global result (Rawhouser et al., 2019).

What is measured can be improved and, through this process, the value is better added and reported (Clark & Brennan, 2012; Hadad & Găucă, 2014; Rawhouser et al., 2019). Understanding that value relies on stakeholders' standpoint adoption (Myrah & Odinsky-Zec, 2013) also allows SEI disclosure to reinforce its collaboration ties, and these ties create the necessary synergy to achieve the expected social goals. Examining the theory of value in the strategy is useful and valid, given the absence of studies in the social field that have delved into the subject (Hlady-Rispal & Servantie, 2018).

Classical economists define value in terms of its usefulness. The use-value is established by consumer evaluation. It is a mixture of needs and expectations concerning the acquisition, experience and product use. It is a subjective and circumstantial perception, which depends on the qualities observed and the information available about what consumers will consume and concerning consumption alternatives (Bowman & Ambrosini, 2000; Lepak et al., 2007). As such, its measurement is complex and multidimensional.

Different values result from different operations with different strategies (Brandenburger & Stuart, 1996); they also reflect different assessments of use-value (Bowman & Ambrosini, 2000; Lepak et al., 2007). In line with a competitive paradigm, maximizing financial return is a practical simplification (Borzaga et al., 2012; Rawhouser et al., 2019). Often the profit functions as a generic parameter capable of representing the interests of most stakeholders (Hadad & Găucă, 2014). Therefore, the surplus of value occurs after realizing the exchange value and assumes a competitive relationship where agents strive for the most significant value appropriation possible (Bowman & Ambrosini, 2000; Lepak et al., 2007).

On the other hand, SEIs have more network ties complexity and manage greater stakeholder diversity with more expectations. Simplifying its complexity can mean not meeting some stakeholders' information needs and harming legitimacy (Borzaga et al., 2012; Rawhouser et al., 2019; Símon-Moya et al., 2012). The value is determined through internal and external relationships assuming the organization as a unit of analysis (Bowman & Ambrosini, 2000; Brandenburger & Stuart, 1996), the most popular unit in this social literature (Lee et al., 2014).

2.2 Measures of social value

The authors have taken exceptional care to ensure that the proposed measures avoided excessive economic and financial bias (El Ebrashi, 2013; Mair & Martí, 2006; Ormiston & Seymour, 2011; Rawhouser et al., 2019). For instance, this study avoids the transposition of the results (Hamann, Schiemann, Bellora, & Guenther, 2013), without associating growth measures (e.g. an increase in the number of beneficiaries) with financial measures (Combs, Crook, & Shook, 2005). The authors also considered that broad measures create measurement problems and restricted measures test fewer hypotheses. Thus, the interrelation between theory and its use was fundamental to outlining this proposal (Combs et al., 2005). These precautions aim to avoid excessive results variation, making complex further replications (Combs et al., 2005; Hamann et al., 2013).

The social value potential for representing a result dimension is assumed (Hlady-Rispal & Servantie, 2018). The value measure is objective when it reflects traditional quantifications (Leviner, Crutchfield, & Wells, 2006) and is subjective when it embodies its complexity (Hossain et al., 2017). This choice is in line with Rawhouser et al. (2019) orientation. All measures employed in the study were selected from the literature and seek to answer the call of Hlady-Rispal and Servantie (2018) for more works adopting the beneficiary's perspective concerning the use of value (Bowman & Ambrosini, 2000; Lepak et al., 2007).

Given the gap in the literature, the authors have developed the following hypotheses based on scales already tested and validated (Converse & Presser, 1986) embracing multidisciplinary (Van der Have & Rubalcaba, 2016). The first Hypothesis (H1) is that the objective social value can be reliably represented by (H1a) direct impact and (H1b) employees. “Direct impact” is the most basic way of measuring social value, widely used to measure the magnitude of SEI's performance (i.e. scale). It is trendy when the investment involves public capital (Leviner et al., 2006), and great social supporters worldwide take such public investments into high consideration (Ciccarino & Rodrigues, 2019). It refers to the quantification of people benefited by SEI in the last year. The employees refer to the quantification of the jobs created, as well as the differentiation between paid and voluntary work, which is fundamental for the perception of the organization's sustainability (Dees, 1998; Símon-Moya et al., 2012), although often represents the organization's size too (Bacq & Eddleston, 2018; Bosma et al., 2016).

The second Hypothesis (H2) is that the concept of subjective social value can be reliably represented by (H2a) uniqueness, (H2b) pain and (H2c) social relief. The concept developed in IMP (2019) was adopted to incorporate SEI's value offering uniqueness, isolating its contribution from other ecosystem agents. The IMP bases its conceptualization on the conjecture of what would happen to beneficiaries if SEI did not exist. However, Menter, Göcke, Zeeb, & Clauss (2020) argue that this uniqueness is related to innovation in the value proposition, and thus, it does not necessarily represent an outcome measure. Therefore, uniqueness can also be how value is created and distributed.

The subjective social value can also be associated with several psychological, social, economic, environmental and political effects (Hadad & Găucă, 2014). The result of a social problem can be considered a pain. The perception of pain is conditioned by the painful experience and the socially constructed idea of what that pain would be. Pains related to emotional and psychological processes are considered more potent than physical pains in some contexts (Biedma-Velázquez, García-Rodríguez, & Serrano-del-Rosal, 2018).

The beneficiary's perception of use-value may be due to the relief he/she feels concerning the pain caused by the problem he/she faces. There is a linear relationship between pain, relief and satisfaction with a significant correlation between changes in the perception of pain intensity and the perception of relief. For this reason, relief scales tend to reflect the therapeutic action better than the intensity of pain. Stahmer, Shofer, Marino, Shepherd, & Abbuhl (1998) found that to reduce high initial pain intensity (>5) is necessary a reduction of 84% of pain to provide some relief. That falls to a 29% reduction in the case of low initial pain intensity (≤5). The pain intensity's average is around 7, and 57% of this pain reduction is enough to offer relief.

Relieving social problems links to the intense perception of pain or discomfort that this situation generates, requiring treatment. The concept of social relief appears in some studies (Bacq & Eddleston, 2018), but it is not measured. Figure 1 shows the posed measures' conceptual model.

H1.

The concept of objective social value can be reliably represented by direct impact (H1a) and employees (H1b).

H1a.

The variable “direct impact” reliably represents the concept of “objective social value.”

H1b.

The variable “employees” reliably represents the concept of “objective social value.”

H2.

The concept of subjective social value can be reliably represented by (H2a) uniqueness, (H2b) pain, and (H2c) social relief.

H2a.

The variable “uniqueness” reliably represents the concept of “subjective social value.”

H2b.

The variable “pain” reliably represents the concept of “subjective social value.”

H2c.

The variable “social relief” reliably represents the concept of “subjective social value.”

The conceptual model provides sense-making interpretations, assuming that past information is a parameter used to perceive reality in each analysis instance. Reality is in motion, so the outcome cannot be measured by looking for constant patterns (Snowden, 2002). Results change over time (Combs et al., 2005; Hamann et al., 2013) but observing its evolution help to make better decisions (Snowden, 2002). The value created can also be comparatively gauged with scenarios, goals and competitors (Venkatraman & Prescott, 1990). Nevertheless, it is essential to have dependable and quickly systematized measures that guarantee the quality of information (Rawhouser et al., 2019). Accordingly, medical research points to a linear relationship between pain and relief, providing a measure of effectiveness (Stahmer et al., 1998).

3. Research method

This study embraces the multidimensionality of the social value (Mair & Martí, 2006) through the basic mixed-methods approach to produce balanced syntheses between literature and practice (Remenyi, Williams, Money, & Swartz, 1998). The research goal is to pose measures to evaluate the social value, simplifying their use by initiatives and social investors (André et al., 2018; Ciccarino & Rodrigues, 2019), providing a path for academic consensus (Venkatraman & Ramanujam, 1986).

We have conducted in-depth, semi-structured interviews from 19 June to 27 December 2019 with eight Portuguese investors and six social entrepreneurs. The selection criteria included having at least one investor and one entrepreneur from each of the five Portuguese geographic regions. We have analyzed the data by pattern matching using the Atlas.Ti software version 7.5. The research identifies the opinion about the value measures and integrates the stakeholder's opinions by examining divergences and convergences (Remenyi et al., 1998).

Then, the quantitative analysis assessed and tested the measures. We have collected the data through an online survey that created a probabilistic sample based on entrepreneurs' opinions (Hair, Anderson, Tatham, & Black, 2010). The survey went through an extensive pre-test process with the collaboration of 56 respondents. We have not used these responses as a study sample.

We collected the data between November 2019 and February 2020. The resulting database corresponds to 40.45% of the population of Portuguese SEIs which have received investment from 2015 to 2019. We treated the sample to ensure it is suitable for multivariate techniques resulting in eighty-nine valid cases (Hair et al., 2010). The sample also represents 43.63% of the response rate, higher than in benchmarking studies (e.g. 21% – Bacq & Eddleston, 2018).

Finally, we have conducted the exploratory factor analysis (EFA) to summarize variables keeping only the statistically significant ones (factor load >0.6). The EFA allowed us to verify if the variables were grouped according to the literature. Due to the research's exploratory nature, we considered the EFA dependable and significant when the KMO test is greater than 0.5, (α < 0.05), extracting factors with Cronbach's alpha higher than 0.6. From EFA we also considered how much the variable relates to the set of measures (MSA < 0.5) and how much it is represented by the factors extracted according to the values of the communalities (Hair et al., 2010).

4. Qualitative analysis

Investors emphasized that assessing social value is one of the main challenges to support SEI. Most investors argue that support for the investment must reflect a social demand (Investors 1, 3, 4, 5, 6 and 8). The investment choice in Portugal is based on the comparison with the competitors in each bid, with no theme guidance. It is worth noting that they use “value” and “impact” as synonyms. The Investor 1 testimonial represents that and points out that they intuitively analyze the division between objective and subjective values (Bosma et al., 2016; Rawhouser et al., 2019).

What we do in the bidding process is contract impacts and results expected from each project. Furthermore, what we are doing throughout the investment life cycle is to check whether these impacts happened or not […]. In general, we ask for in the applications is that they present result indicators. […] And therefore, we have two indicators. Those of accomplishment, which is: how many actions they did, how many people they served, and the results: the impact on the living conditions where the project operates. Realizing whether, after the intervention, the beneficiary's life has changed is the result indicator. The realization is whether the beneficiary has been served. There is an adjustment between the indicator proposed in the application and the one that is formalized later. (Investor 1)

Investors assume that SEIs can explain how they create and measure their value without offering guidance. However, all investors consulted recognize how hard it is for SEIs to do it. Investor 8 highlights the misuse of public resources due to the difficulty of assessing the social results: “The State pays to have some results in return, some of them exceeding 1,000 €, but, usually, we do not assess the impact on that person's life.” On the other hand, SEIs explain that assessing the social value is not problematic, but systematizing and reporting it. “We will never be able to measure every impact we generate” (Entrepreneur 1). They receive spontaneous and constant feedback from their beneficiaries: “I measure the impact practically every week with messages that I receive from beneficiaries.” (Entrepreneur 2). Entrepreneurs 2 and 6 have been using storytelling to complete indicators assessments.

Investor 2 admits the difficulty in implementing the available evaluation methods (André et al., 2018). “We already tried to make an initiative with SROI – Social Return on Investment, and it took us so long and it was so complex, that we ended up not implementing it.” The indicators selection limits the available knowledge about the social problem and can even represent a follow-up problem for investors who need the social result to justify their investment (André et al., 2018; Borzaga et al., 2012; Ormiston & Seymour, 2011). The subjective nature of social value represents the enormous challenge of supporting SEI and maintaining its sustainability. Therefore, limiting the scale of the social value created and making it more challenging to structure SEI properly to meet the requirements of the investment selection processes. This vicious cycle constrains the social impact (Entrepreneurs 1, 2, 3, 5 and 6). Investors 2, 3 and 4 recognize this difficulty and highlight the complexities of creating generalist indicators for subjective values.

But there are often issues that are not measurable, and not tangible. When we work with people, many things are not tangible. Possibly there are already metrics for the intangible; I do not know. However, it has several dimensions. […] it is an issue […] that we can solve through studies, like what you are doing. (Investor 2)

Sometimes social investors cannot even understand the social value proposition: “We are presenting a problem they did not have, but they know it exists” (Entrepreneur 1). Furthermore, there are taboos related to some initiatives with less potential to enhance the company's image and end up inhibiting support through corporate social responsibility. “And it is tough for anyone to finance an ugly activity, […] companies do not want to be associated with that, you know?” (Entrepreneur 1). As equity is generally rejected, those sensitive issues must rely only on public investment (European Commission, 2016): “[…] we do not have equity, and we do not want to share that equity […]” (Entrepreneur 6).

All the entrepreneurs interviewed measured value somehow and gave a practical importance to this process. “My feeling is that since we measure impact, it is easier to seek support and advice, because it allows us to show the results we have in a very succinct way. Thus, we gain the trust of social investors” (Entrepreneur 1). Quantification is a popular way to report impact (Entrepreneur 1, 3, 4, 5, 6). All SEIs must achieve some tangible goal (Leviner et al., 2006). At “[…] a quantitative level, it is exceedingly difficult to do. What we can do is estimate the service we offer, the number of activities we did (…) and the number of daily participants we had.” (Entrepreneur 3). Some interviewees claimed that they had improved their perception of the evaluation's importance after participating in the investment process and if such a process had helped them to organize previously dispersed information and methods. They argue that they had rethought their initiatives, better defined the problem they were solving, and identified their social value more precisely (Entrepreneurs 2, 3, 5 and 6).

When social value complements the government action, the complexity to prove it increases (European Commission, 2016; Stephan, Uhlaner, & Stride, 2015) because it turns challenging to show the real SEI's impact without capturing effects from other agents (Investor 4). This issue can damage the claim's credibility (Investor 6) as explained in the literature (Rawhouser et al., 2019). It is a concern present in all social entrepreneurs' testimonials. They always explain and contextualize their SEIs to prove how they create value that indirectly affects different stakeholders (Clark & Brennan, 2012; Hlady-Rispal & Servantie, 2018; Hadad & Găucă, 2014). Moreover, it is also challenging to dissociate the value created from a general benefit to society (Bacq & Eddleston, 2018; Tate & Bals, 2018), despite SEI being embedded in communities and contexts (Lumpkin et al., 2013).

We posed the pain and relief medicine scales (Stahmer et al., 1998) during all interviews with great acceptance. Only Investor 4 disagreed because he fears it would be too subjective. Investor 5 posed interesting reflections about who determines pain and relief. We have included them in the suggestions for future studies due to the importance of comparing these measures with other perspectives (Hlady-Rispal & Servantie, 2018; Myrah & Odinsky-Zec, 2013). It aligns with the non-static character of social value (Bowman & Ambrosini, 2000; Brandenburger & Stuart, 1996) and the measures' potential to reflect SEI's evolution (Venkatraman & Ramanujam, 1986). Investor 3 adds the preventive and corrective potential of the relationship between pain and social relief. It is in line with prognosis analysis (Stahmer et al., 1998).

5. Quantitative analysis

The retained factors' quality is significant and adequate for an exploratory study. The variables explained 64.536% of the sample variation. It is worth mentioning that the uniqueness variable is highly correlated with the other variables together (MSA = 0.641), but it has a slight correlation with the factors (Communality = 0.328). Although it can be part of the subjective social value construct statistically, it does not have a high factor load (0.531). Excluding it from the construct's composition increases reliability, as Cronbach's alpha rises from 0.593 to 0.699. The scree plot reinforces this decision. It corroborates the analysis by Menter et al. (2020) that put uniqueness as an innovation dimension in the value creation process. Table 1 summarizes the factors extracted.

Therefore, EFA suggests that the objective social value can be reliably represented by the variable direct impact (H1a) and employees (H1b), supporting the first Hypothesis (H1). The second Hypothesis (H2) is partially supported because the subjective social value concept can be reliably represented by the variable pain (H2b) and social relief (H2c). However, the uniqueness variable (H2a) forms another measure (Menter et al., 2020) that perhaps can be further studied in an ecosystem approach (Stephan et al., 2015).

Portuguese SEIs also deal with social problems that would not find treatment in other structures (Stephan et al., 2015). Emphasizing the subjective social value parameters (Stahmer et al., 1998): 81.8% of cases address severe social pains (>5) and 92% of cases produce an intense relief (>5). Social pain and relief are correlated (Pearson = 0.541 α < 0.05). On average, SEIs create 12.65% greater relief than the perceived pain. Thus, they effectively treat the problem. Only 24% of SEIs that work with severe pain do not create the necessary relief, and none of the 16 SEIs that work with less serious social problems fails to produce the necessary relief. Thus, the measures serve as parameters (Snowden, 2002) to assess the effectiveness of SEIs (Venkatraman & Prescott, 1990).

6. Conclusion

Due to the particularities of SEI, regarding their sources of opportunities and the way resources are mobilized, their performance cannot be assessed in the traditional way (Símon-Moya et al., 2012). Therefore, the analysis of value creation requires specific measures (European Commission, 2016; Hlady-Rispal & Servantie, 2018; Rawhouser et al., 2019). This study achieved its goal of presenting these measures by offering a new synthesis from the literature on strategy and medicine.

The study clearly defined its concepts and distinguishing social value from social impact (Venkatraman & Ramanujam, 1986) to provide a structure that can be interpreted and replicated (Lee et al., 2014). The results presented aim to favor the academic consensus by clarifying essential criteria based on empirical analysis (André et al., 2018; Lee et al., 2014; Rawhouser et al., 2019) and by emphasizing the more agreed-upon feature to differentiate the social from the conventional initiatives: the focus on social value (Barki et al., 2020; Hlady-Rispal & Servantie, 2018; Hossain et al., 2017).

The adoption of valid and reliable measures to express such a vital construct is fundamental to reducing the results variation and allowing knowledge consolidation (Combs et al., 2005; Hamann et al., 2013). It also allows SEIs to check their progress compared to a manageable parameter (Snowden, 2002). The measures are balanced criteria, capable of creating reliable assessments and testing hypotheses (Combs et al., 2005). They also pave the way for developing or improving existing evaluation methods, adopting an impact assessment lens (Rawhouser et al., 2019).

The qualitative analysis took advantage of the dynamic relationship between the Portuguese investors and the social entrepreneurs (Ciccarino & Rodrigues, 2019). The analysis supported important literature aspects such as the use of value and impact as synonyms and the need to measure social value (Bosma et al., 2016). It also highlighted the difficulty in implementing systematic value assessment, due to the complexity of the available methods (André et al., 2018). In addition, it showed the social effects of multidimensionality and the difficulty of dissociating the value created from a general benefit for society (Bacq & Eddleston, 2018; Tate & Bals, 2018).

The quantitative analysis followed to evaluate and test the proposed measures. The two measures initially proposed (i.e. objective and subjective social value) proved to be best used as three. The subjective social value is represented by (1) pain and relief scales (Biedma-Velázquez et al., 2018; Stahmer et al., 1998); (2) the objective social value by the number of beneficiaries and jobs created (Dees, 1998; Ciccarino & Rodrigues, 2019; Leviner et al., 2006) and (3) the SEI's uniqueness by the extent to which other available social structures can replace the SEI activity (IMP, 2019; Stephan et al., 2015).

7. Limitations and future studies suggestions

This paper provides evidence that can help improve social value measurement. However, the techniques used are exploratory (Hair et al., 2010). Therefore, we highlighted some research opportunities, which we could not explore, such as the potential of using pain and relief measures in preventive and corrective approaches. We tested the three measures based on social entrepreneurs' opinions, but they also serve to collect beneficiaries' opinions in a future study to compare the results. This versatility has immense potential for internal and external evaluation processes (Clark & Brennan, 2012). This kind of research can emphasize the beneficiary dimension, which is neglected in social evaluation studies (Hlady-Rispal & Servantie, 2018; Myrah & Odinsky-Zec, 2013) and has the potential to highlight the use-value (Bowman & Ambrosini, 2000; Lepak et al., 2007).

In a larger and more diverse sample, it would be interesting to identify relationships between business models' features and social value creation measures (Menter et al., 2020). In addition, it could study the social value creation relationship with the ecosystem (Stephan et al., 2015; Venkatraman & Prescott, 1990). Finally, the proposed measures can be a starting point for studies to understand the value appropriation in SEI, something fundamental for the sustainability of business models (Dwivedi & Weerawardena, 2018; Hlady-Rispal & Servantie, 2018).

Figures

Figure 1

Conceptual model

Table 1

Factor analysis

Factor	Variable	Factor loading	MSA	Communality	Cronbach's alpha
Objective social value	Employee	0.843	0.496a	0.720	0.652
	Direct impact	0.859	0.523a	0.739
Subjective social value	Social relief	0.870	0.506a	0.777	0.699
	Pain	0.810	0.520a	0.663
Uniqueness	Uniqueness	0.531	0.641a	0.328	–

Note(s): KMO = 0.522; Bartlett's test of sphericity, sig <0.01

The “a” only signal the diagonal MSA’s number to consider in an evaluation

Source(s): Authors