1 Introduction

Currently, organizations are developing a growing interest in promoting socially friendly activities. Michelon et al. (2013) identify advantages of an organization deciding to promote these activities, such as improvement in its legitimation and reputation, a better relationship with its stakeholders and the promotion of skills, processes and systems that increase the organization’s competitiveness. These advantages are translated into the ability to generate social and economic performance. As a consequence, one of the most interesting topics studied in the literature is the relationship between an organization’s economic and social performance.

The aim of this paper is to determine the existence and nature of the relationship between economic and social performance in the organizational context. Although the concepts of social and economic performance were originated in socioeconomic research more than 20 years ago, there are no generally accepted definitions, measurements or descriptions of the interactions between them (Felício et al. 2013; Bellostas et al. 2016).

This paper develops a meta-analysis of the relationship between economic and social performance. Meta-analysis is an appropriate statistical approach to use when multiple individual studies have yielded inconclusive or conflicting results, as in the case of this relationship (Waddock and Graves 1997; Rosenthal and DiMatteo 2002; Orlitzky et al. 2003; Wu 2006). We propose the treatment of the measurement criteria of economic and social performance and the characteristics of the organization as elements that can condition this relationship. Orlitzky et al. (2003) and Margolis et al. (2007) analyzed some of these aspects approximately 10 years ago. However, in the last few years, there has been a strong progress in this research field with the creation of new measurement criteria or indicators of economic and social performance. This paper introduces these new criteria and analyses its influence in the relationship between economic and social performance in different types of organizations. The influence of these new elements has not been studied by the previous economic literature up to date. As our main contribution, we statistically aggregate extant evidence concerning the claim that social performance interacts with the economic performance of an organization. Second, we test a central assertion of instrumental stakeholder theory, i.e., that there is a positive interaction between the two types of performance. Moreover, we investigate whether the relationship varies based on the distance between performance measures and characteristics of the organization. In particular, those measurements of social performance that include the degree of satisfaction of stakeholders promote higher interaction between both types of performance. Something similar happens when the organization is oriented to service delivery or belongs to an intensive technology sector. Finally, we note that organizations must design and integrate relevant definite indicators in their strategic management practices and that researchers should be careful in drawing conclusions because they could be influenced by the abovementioned elements.

This paper is organized into five sections: The first section is the introduction. The second section defines the various research questions posed in this paper. The third and fourth sections introduce the methodology and the results, respectively, to answer the proposed research questions. In the fifth section, we discuss the results. The last section provides conclusions based on the results obtained.

2 Theoretical Background

2.1 Interaction Between Economic and Social Performance

The link between economic and social performance has been a core topic in the management literature for years (Schaltegger and Synnestvedt 2002). Corporate social responsibility and socially friendly activities have been understood as an alternative way of generating economic and social welfare (Godfrey and Hatch 2007). These practices imply the creation of social value from different initiatives. Traditionally, business companies, cooperatives and mutuals created social value through the market, whereas other types of nonprofits, such as foundations or associations, created social value outside the market system (Sanzo et al. 2015; Costa and Carini 2016). Nowadays, all these organizations have an active role in markets, competing between them to obtain users and financial resources, although with a different social orientation or strategy (Chaves and Monzón 2012).

Despite the large number of academic contributions, the links between social and economic performance remain unclear (Brammer and Millington 2008; Hahn and Figge 2011; Orlitzky et al. 2003; Waddock and Graves 1997; Wu 2006). Aupperle et al. (1985) and, more recently, McWilliams et al. (1999) and McWilliams and Siegel (2001) find no empirical relationship between economic and social performance in companies with a social orientation. By contrast, Waddock and Graves (1997), Kinnell and McDougall (1997), Blois (1999) and Sargeant (1999) detect a positive relationship between a proxy of social value and accounting measurements of economic value, whereas Abiodun (2012) detects a negative relationship between investment in social activities and economic return. Taking into account the conflicting results reached by previous studies, we propose the following research question:

RQ 1 : Is there a significant relationship between economic and social performance?

If there is a significant relationship, the results will be in line with Preston (1978), Freedman and Stagliano (1991), Graves and Waddock (2000), Berman et al. (1999), Van de Velde et al. (2005) and Wu (2006). These authors all find a relationship between economic and social performance. The sign of this relationship could be influenced by the measurement criteria and the indicators used by different authors to analyze this relationship. In the context of corporate social responsibility, Orlitzky et al. (2003) study the importance of measurement criteria as influential variables. Bellostas et al. (2016) also detect a lack of agreement among academic researchers concerning the composition and measurement of both types of performance. Moreover, in the last few years, both new types of organizations with social orientation and measurement criteria of performance have emerged in the economic arena (López-Arceiz et al. 2016). However, the impact of these criteria and the behavior of new hybrid organizations have not been studied yet. Thus, correlations between the economic and social performance constructs can be influenced by these new factors.

2.2 Measurement Strategies for Economic and Social Performance

The interaction between economic and social performance can be influenced by the measurement criteria adopted in each research project, being a lack of consensus about the operational level (Yang et al. 2014; Testi and Bellucci 2011).

In this sense, the measurement of economic performance is not free of challenges. Economic performance supposes that stable and continuous economic activities are being conducted. The question is how to measure an organization’s economic activity. Orlitzky et al. (2003) proposed three broad subdivisions of economic performance: market-based (investor returns), accounting-based (accounting returns), and perceptual (survey) measurements. Market-based and accounting-based measurements constitute a partial perspective because they recognize only the consumer and the producer or owner of a company as legitimate stakeholders (Payne et al. 2000; Johansen and Nielsen 2012; Nishimura 2007; Fontaine et al. 2006; Freeman 1984). In this case, traditionally the most used criterion has been the accounting return, but nowadays sales or asset growth are more important in some entities such as nonprofit organizations (Liu et al. 2012; Coombes et al. 2011; Bai 2013). Something similar happens with perceptual measures. These measures are based on the answers of a person who can give a subjective evaluation (Conine and Madden 1986; Reimann 1975). The perceptions of managers are being used as a source in the measurement of economic performance because managers have access to the entity’s economic targets (Brouthers 2002; Liu et al. 2012). Nevertheless, it is reasonable to assume that the measurement criteria of economic performance chosen by the researcher can influence the relationship between economic and social performance. For instance, Lu et al. (2014) evidenced a negative effect of the market measurements. These indicators tend to consider all the available information, while accounting indicators are the result of the organizational accounting policy. Then, currently, the traditional criteria, compiled by Orlitzky et al. (2003) and Margolis et al. (2007), live together with new measurement criteria, such as subjective measurements and growth or size criteria. These new measurements can be able to influence positively the interaction between economic and social performance according to Santos and Brito (2012) or Bai (2013) (Table 1). Therefore, we define the following research question:

RQ 2 : Does the relationship between economic and social performance depend on the measurement criterion of economic performance?

If there is no influence of the measurement criteria of economic performance, we can assume that although there is no consensus in the measurement criteria of economic performance, there is a general agreement about the meaning of economic performance (such as return, growth or perception). Conversely, if we observe an influence of these criteria, economic performance should be considered a multidimensional construct with different dimensions that the researcher must consider (Moneva and Ortas 2010).

Table 1 Expected signs related to the moderator variables

This idea is relevant when we analyze the measurement criteria of social performance. In general terms, social performance refers to the generated impact on stakeholders affected by the organization. Lu et al. (2014), Orlitzky et al. (2003) and Post (1991) identify four strategies for measuring social performance: (a) Social performance disclosure; (b) Social performance reputation ratings; (c) Social audits, social performance processes, and observable outcomes, and (d) Managerial social performance principles and values. Social performance disclosure is a criterion based on public information (annual reports, letters to shareholders, etc.). Although this is the most objective criterion, information disclosure by itself is only a proxy of social performance and may be insufficient to study this element in its entirety (Farneti and Guthrie 2009). The second and third approaches are related to systematic third-party efforts to assess a firm’s ‘objective’ social performance behaviors, such as community service, environmental programs, and corporate philanthropy. For this criterion, the main problem is the comparability of the information. If the initiative does not publish the social audit process, the comparison will not be feasible, and the usefulness of this criterion will be low (Gao and Zhang 2001, 2006). The fourth criterion assesses the values and principles inherent in an organization’s culture (Aupperle 1984; Carroll 1979). This criterion is a broad category with a high level of subjectivity because it is based on the perceptions of the individual who evaluates these values and principles.

Although these authors made an important effort when they studied these measurement strategies, additional criteria should be considered at present. For example, service quality can be an indicator of the level of integration of stakeholders’ needs into the organization (Mitchell et al. 1997; Sacchetti et al. 2016). Furthermore, community interests or regional development are proxies of this integration when the entity promotes higher levels of growth in that area (Borzaga and Fazzi 2000). Other authors have developed indicators, such as social return on investments that offer a specific vision of social performance (Rotheroe and Richards 2007). Finally, social auditing and social indexing are not available in all cases because some entities are easier to access than others. Table 1 shows negative influences when the measurement criterion uses the third-party assessments. Moreover, the new criteria would be able to change the interaction between economic and social performance. Millar and Hall (2013), in relation to the social return on investment, suggest a tendency to obtain positive relationships. Bai (2013) in relation to social auditing identify negative interactions in the context of nonprofit organization which are not table to participate in social indexing. All these particularities can modify the relationship between economic and social performance. These new aspects have not been studied by the previous meta-analysis. As a consequence, we propose the following research question:

RQ 3 : Does the relationship between economic and social performance depend on the measurement criterion of social performance?

Finally, some organizational characteristics, which can act as control variables, influence the relationship between economic and social performance. Deegan and Gordon (1996) and García-Ayuso and Larrinaga (2003) identify a strong influence of the type of developed activity on the relationship between economic and social performance. The social orientation of the organization is also a variable that can modify this relationship. According to Felício et al. (2013), entities that adopt a legal form closer to nonprofit organizations will have a stronger social orientation and will be able to create a more intense relationship. However, other authors, such as Bai (2013), Bouckaert and Vandenhove (1998) and Weisbrod (2009), propose that although nonprofit organizations have an explicit social aim, self-dealing and market competition can prevent these entities from reaching an optimal level of social performance. The level of technology required by the organization also determines this relationship. In this sense, Guadamillas et al. (2010) and Morfit (2014) state that entities belonging to technological sectors are the ones that provide more information to their stakeholders and, as a consequence, are able to create a more intensive relationship between economic and social performance. Other characteristics that can influence this relationship are the cultural environment of the organization. Defourny and Nyssens (2008), Kerlin (2006), Quintão (2007) and Hulgård (2010) show that the impact of socially friendly activities varies based on the diversity of experiences at a regional level and is affected by the prevailing cultural backgrounds. As a consequence, the prevalent sphere of values will promote the development of a more intense relationship between economic and social performance (López-Arceiz et al. 2016; Wang et al. 2016).

RQ 4 : Organizational characteristics are a moderator variable of the relationship between economic and social performance.

The previous three research questions allow the relationship between economic and social performance to be tested from different perspectives to determine the extent to which economic and social measurements and the characteristics of the entity influence the behavior of organizations that decide to develop a “double bottom” strategy.

3 Methodology

3.1 Sample and Indicators Used

Searches of the Web of Science, Scopus, and ABI/Inform, databases were conducted using the keyword ‘organizational performance’. Synonyms, which were searched separately, were ‘organizational performance’, ‘profitability’, ‘economic performance’, ‘financial performance’, and ‘economic value’. The keyword ‘social performance’ was alternately substituted with ‘(corporate) social responsibility’ and ‘social value’. Web of Science gives access to the full text and images of more than four million business and trade journal articles, with a coverage period of one hundred years. Scopus indexes abstracts of journal articles (approximately 57 million references) and books (approximately 100,000 references). To increase the scope of our search, cross-citations from previous reviews (for example, Orlitzky et al. 2003; Margolis et al. 2007) were also explored.

The relevant studies selected for the meta-analysis had the following characteristics, and these were the selection criteria:

  • The studies referred to concepts associated with socially responsible businesses, social enterprises and nonprofit organizations.

  • The analyzed studies quantitatively examined the relationship between economic and social performance. The reported effect size could be Pearson’s correlation r, a t test statistic or an effect size (Hunter and Schmidt 1990).

  • The studies were concerned with at least one aspect of a firm’s economic performance. To study the different aspects, we distinguished between five possible criteria based on the theoretical framework (Moneva and Ortas 2010): (a) Accounting measurements, (b) Market criteria, (c) Economic aim management or perceptual indicators, (d) Size or growth criteria, and (e) Other measurements. We identified indicators that had a frequency of one in our database search as ‘other’ (for instance, the level of intangible assets).

  • The same procedure was used for social performance.Footnote 1 According to the previous economic literature, we considered seven possible indicators: (a) Professional integral audit based on social performance disclosure (e.g., KLD), (b) Stakeholder integration (e.g., managerial social performance), (c) Service quality, (d) Social auditing/indexing (e.g., reputational measurements), (e) Regional development criteria, (f) Created social value criteria (e.g., social return on investments), and (g) Other criteria (Wood 1991; Moneva and Ortas 2010). In the ‘other’ category, we included indicators that had a frequency lower than one (for instance, volunteering or networking).

  • Finally, we considered organizational characteristics such as the organization’s activity (raw materials, production of goods or service delivery), its social orientation (based on its legal form), the intensity of its use of technology, and the cultural environment (Anglo-Saxon or continental) in which the organization was framed.

As consequence, we had access to 678 effect sizes from 83 papers.Footnote 2 The Appendix lists the most important study characteristics, such as author(s), date of study, study sample size Ni, observed r (or transformed and/or partially corrected r), number of correlations per study, organizational characteristics and the measurement criteria of economic and social performance.

3.2 Methodology

A meta-analysis integrates the quantitative findings of separate but similar studies and provides a numerical estimate of the overall effect of interest (Petrie et al. 2003).Footnote 3 This meta-analysis uses Hunter and Schmidt’s (1990) statistical aggregation techniques for cumulative correlations and to correct for various study artefacts to estimate the true score correlation (ρ) between economic and social performance. The meta-analysis arrives at a mean true-score correlation by correcting observed correlations for sampling error.Footnote 4 Because sampling error varies directly with sample size, all studies are weighted by sample size Ni (Schmidt and Hunter 1977). Studies with a smaller standard error and larger sample size are given more weight in the calculation of the pooled effect size.Footnote 5

Agreement or disagreement between the studies can be examined using a heterogeneity test. In this study, we use Cochran’s Q. This statistic is the weighted sum of squares on a standardized scale. It is reported with a p value, where low p values indicate the presence of heterogeneity (Higgins et al. 2003). To test the relationship between economic and social performance, we specify a meta-regression model to study the role of the measurement criteria of economic and social performance. In this model, we have added the influential variables, such as dummy variables, following this expression (1):

$$\rho_{ij} = \alpha_{i} + \beta_{i} D_{ij} + \varphi_{ij}$$
(1)

where ρij is the effect size, Dij represents each influential variable (economic and social performance measurement criteria and organizational characteristics), and \(\varphi_{ij}\) is the random error. Parameter \(\beta\) measures the effect of the moderator elements on the effect size. We use the software SPSS 22.0 and Stata 14.0 to estimate the different models.

Finally, we have implemented a bootstrap estimation (number of samples: 10,000) in order to robust the obtained results. A bootstrap estimation allows us to obtain a set of uniform subsamples based on the total sample, the original model being tested in each one. Moreover, we have re-estimated the model using the Bayesian estimator. The Bayesian estimator is suitable in a meta-analysis when we have finite sample sizes and we introduce prior information based on previous research. Both techniques enable us to robust the previous results.

4 Results

As shown in the first line of Table 2, the mean observed correlation for the total set of 678 correlations (k) and the total sample size (N) of 1,368,044 observations is 0.188, with an observed standard deviation of 0.289.

Table 2 Meta-analysis with sample error correction

As Table 2 shows, Cochran’s Q coefficient has a p value lower than 5%, which indicates the presence of heterogeneity in the studied sample. Therefore, we decide to use a random effects meta-regression model. Thus, the true (corrected) correlation score is 0.199, which is higher than the observed correlation with a confidence interval at 95% of (0.165–0.232). Therefore, there is positive and significant relationship between economic and social performance among the papers that discuss this relationship. However, this result could be affected by the measurement criteria employed for social and economic performance. Moreover, the control variables related to the characteristics of the studied entities could affect this relationship. For this reason, taking into account the presence of heterogeneity, we decide to include these elements as moderator variables.

In Table 3, we show the impact of the measurement criteria of economic performance on the relationship between economic and social performance. Taking into account the previous literature, we create five measurement sets to examine the moderator effects based on the measurement criteria of economic performance: (a) Accounting criteria, (b) Market criteria, (c) Perceptual criteria, (d) Size criteria, and (e) Other.Footnote 6

Table 3 Meta-analysis with sample error correction. Economic performance

Table 3 indicates that the association between economic and social performance depends on the type of measurement used by the researcher to measure economic performance. The size criteria reveals the highest positive correlation between economic and social performance [r: 0.828, CI (0.687–0.908)], whereas other (related to subjective organizational aspects, such as self-values and utilitarian identity) presents the lowest correlation [r: − 0.054, CI (− 0.202 to 0.096)]. Accounting measures are more highly correlated with social performance than market-based measures [r: 0.167; CI (0.147–0.187) vs. r: 0.082; CI (0.071–0.093)]. Finally, perceptual criteria, related to management by targets, show an intermediate behavior [r: 0.129; CI (0.111–0.146)]. Therefore, the relationship between economic and social performance changes when we consider the measurement criteria of the economic dimension.

We also test whether the measurement criteria of social performance may affect the relationship between economic and social performance. The results are shown in Table 4.

Table 4 Meta-analysis with sample error correction. Social performance

To study the measurement of social performance, we distinguish between the following categories: (a) Professional integral audit criteria (e.g., KLD); (b) Stakeholder criteria; (c) Quality criteria; (d) Social auditing/indexing criteria; (e) Regional development criteria; (f) Created social value criteria; and (g) Other criteria.Footnote 7 The results show that the highest correlation occurs when the measurement criteria include the degree of satisfaction among stakeholders [r: 0.221, CI (0.163–0.278)]. By contrast, the lowest value is observed when the researcher decides to entrust in the measurement of a third party [r: 0.072; CI (0.062–0.082)]. In all cases, the correlations are positive, except when the created social value criteria are used [r: 0.215, CI (− 0.044 to 0.447)]. Therefore, the measurement criteria of social performance moderate the relationship between economic and social performance.

The obtained results are robust according to the meta-regression model (Table 5). In all cases, the indicators of each dimension determine the correlation between economic and social performance (p value < 0.05). However, the interpretation of each parameter is different because the β parameter is a measurement of the intensity of the change.

Table 5 Meta-regression. REML estimator

For example, in economic performance, when the paper uses a size criterion, the relationship between economic and social performance is higher (β: 0.158), whereas when the author uses the market criterion, the result is inverse (β: − 0.069). Although we are not able to determine the correlation using this methodology, we can approximate the change in magnitude. Thus, this method is complementary to the traditional meta-analysis. This methodology enables us to determine the influence of different variables. As we can observe, entities whose activity is related to service delivery are able to intensify the interaction between economic and social performance (β: 0.274, β: 0.296). This same pattern is revealed in high-technology organizations (β: 0.214, β: 0.239) in an Anglo-Saxon cultural environment (β: 0.071, β: 0.127). In contrast, socially oriented organizations are not able to promote a more intense relationship between economic and social performance because of the non-significant parameter achieved in the meta-regression (β: 0.019, β: − 0.021). Taking into account this result, a positive correlation between economic and social performance is detected, although this result is affected by the measurement criteria of economic and social performance and organizational characteristics.

This result is consistent with the alternative estimation techniques used in this study (Table 6). The first four columns show the parameters, p values and confidence intervals for the bootstrap estimation, while the second four show the same information for the Bayesian estimator.

Table 6 Meta-regression. Robustness

As we can observe, there is a positive interaction between economic and social performance, although the final sign depends on the measurement criteria and organizational characteristics. We only attract attention on the improvement in term of goodness-of-fit in the case in Bayesian estimation which robust the obtained results.

5 Discussion

The results of this meta-analysis demonstrate a positive association between social and economic performance across the studied papers. This result contradicts conclusions of McWilliams et al. (1999) and McWilliams and Siegel (2000), who state that economic and social performance are independent spheres in the organizational context. By contrast, our results support the conclusions of Waddock and Graves (1997), Kinnell and McDougall (1997), Blois (1999) and Sargeant (1999), who detected a positive relationship between economic and social performance.

However, this relationship may be influenced by the criteria used in the measurement of economic and social performance and by organizational characteristics. The measurement criteria for economic and social performance have been discussed in previous papers. Brown and Perry (1994, 1995) and Wood and Jones (1995) found that positive correlations may be artefactual functions of the measurement elements. Therefore, we distinguish different measurement indicators in the definition of both types of performance in our meta-analysis. In the analysis of the previous literature, we identified five measurement criteria. Differences in the correlation between economic and social performance are observed in the subjective criteria (other criteria), when the measurement adds elements such as self-behavior or a utilitarian identity. This measurement can cause illogical results because the relationship is based on the opinion of the manager who evaluates the level of economic performance in the entity. This result is also found by Orlitzky et al. (2003), who observe that when the economic performance measurement is based on a survey, the cross-study variation in correlation is removed, and the correlation becomes positive. Measurements based on perceptual criteria are associated with a stronger relationship between economic and social performance according to Santos and Brito (2012) or Peloza (2009). Thus, according Orlitzky et al. (2003), many of the negative findings in individual studies are artefactual, and if the researcher or the company uses a different criterion, positive relationships will appear (Jones and Wicks 1999; Pava and Krausz 1995; Wood and Jones 1995). The meta-regression shows that changes in the measurement criteria used tend to strengthen or weaken this relationship. Measurements that are not associated with efficiency, such as size measurements (sales or asset growth), are able to favor the relationship. However, market criteria introduce a penalization. This same result had been obtained by Goyal et al. (2013). Therefore, the use of a criterion can encourage or discourage the relationship between economic and social performance.

In relation to the measurement of social performance, we have grouped the indicators into seven categories and obtained different intensities in the function of each indicator. The weakest relationship is obtained when the created social value criteria are used. In the meta-regression, we observe that if the researcher decides to change the measurement strategy of social performance, it can influence the interaction between economic and social performance. In this sense, the indicators based on professional integral auditing and social auditing/indexing can decrease the strength of the relationship between economic and social performance. This result diverges from Chen et al. (2015), who obtain a positive interaction in the context of manufacture sector when these criteria are used. In contrast, taking into account the local impact and the regional development may improve this relationship. In any case, similar to the measurement of economic performance, some studies use one measurement and have small sample sizes; therefore, the conclusions in some papers may be biased (Orlitzky et al. 2003).

Finally, the control variables play an important role. The activity of the organization determines the relationship between economic and social performance. Those activities related to the services sector are able to promote a more intense interaction between the two types of performance. This result is obtained by Miles et al. (2014), who demonstrate a stronger relationship in the case of organizations in the sphere of social services. Other control variables also show a positive effect on this relationship. Then, when the entity develops high-technology activities, it is able to create a better interaction, according to Guadamillas et al. (2010) and Morfit (2014). The Anglo-Saxon environment also tends to promote greater interaction (Jackson and Apostolakou 2010). According to these authors, the differences in the institutional context and the level of involvement of stakeholders are the explanations for this behavior. In contrast, the social orientation of the organization does not influence this relationship. Costa et al. (2012) or Bellostas et al. (2016) detect a strong relationship between social and economic performance in Italian social cooperatives and Spanish sheltered workshops, respectively. This result can be explained based on the legal form of the organization, which drives this positive correlation. However, the meta-regression evidences that the social orientation does not influence the relationship between economic and social performance. In this sense, the adoption of professional management criteria in nonprofit organizations and the promotion of socially friendly activities in for-profit organizations has reduced the gap between both types of organizations according to .Chaves and Monzón (2012). So, social performance can be created by hybrid organizations in the market or in the nonmarket, independently of their legal form.

6 Conclusions

The objective of this paper has been to analyze the relationship between economic and social performance in the organizational context. The results show how those entities that develop socially friendly activities experience positive synergies between their social and economic performance. However, some singularities appear when we take into account the measurement criteria of economic and social performance and some characteristics of the organization, such as its activity, its technology and the cultural environment in which it operates. Although some of these indicators had been analyzed by previous studies (Orlitzky et al. 2003; Margolis et al. 2007), the impact of the new measurements of performance and organizational characteristics had not been considered as an influential variable.

Moreover, this paper contributes to the academic debate about the relationship between economic and social performance and shows how it is possible to foster social and economic performance from different strategic organizational models. In fact, a gradual process of convergence occurs in which some non-profit entities tend to develop the economic side in their management model. Similarly, some for-profit entities tend to develop their social side. Currently, there are emerging new models of hybrid organizations that pose a challenge for researchers and managers who need new theoretical frameworks that can explain these models. In any case, it is not possible to provide a universal set of indicators for the measurement of both types of performance due to the observed diversity among the different entities. Therefore, this paper also issues a warning about the use and design of different indicators. In this sense, managers of organizations must design specific indicators that take into account the singularities of the entity. Otherwise, if they follow general indicators, the measurement will be imprecise, and conclusions about the efficiency of the activity will be measured incorrectly.

Finally, this paper has some limitations that should be noted. The aggrupation in different categories of the indicators of economic and social performance is based on previous studies, and it could be different if we analyzed other papers. Moreover, in some selected studies, we have detected small sample sizes, which could influence the extracted conclusions. This fact and the lack of specific indicators are limitations that future research must address.