Abstract
Since the beginning of this century, Spain has become a host country for immigrants. In many cases, due to their living conditions, they are under pressure to demand different services and programs that are crucial for their social integration. This work is basically aimed at analyzing the immigrants’ satisfaction with social services. This article proposes an explanatory model of these users’ satisfaction with social services. The proposed model reports five significant factors, although the results obtained indicate that the three main factors influencing the satisfaction of users with social services are responsibility, assurance, and empathy. Satisfaction differences were also found according to users’ gender and place of residence.
The consolidation of the democratic state and the expansion of the social welfare system, as well as the speed of social changes in Spain, many of which are based on migratory flows, have led to growing demands not only in terms of number, but also of the quality of social services. This has led to paying more attention to the concept of quality, employing a number of models that allow an optimal combination of external and internal mechanisms which enable the analysis of the functional quality of the service itself and, along with it, the quality perceived by users.
In spite of the different existing models on the quality of a service, all of them are based on the assumption that any quality-focused organization should work with the support not only of the management, but also of all the people involved. A proper management of staff, resources, and all their daily processes should therefore be performed. This has an impact both on the satisfaction of the professional team and the recipients, and on the establishment of a positive social image (Fernández-Villarán Ara, 2003). In fact, according to Giménez-Bertomeu et al. (2012), the four components essential when establishing a system set up to ensure the efficient quality of social services are workers, team leaders, organization, and users. Specifically, the need to include users’ perspectives lies in two fundamental factors:
Users’ participation is recognized as a method that defines quality. The European Commission (2010) establishes that one of the governing principles of the relationship between service providers and users is participation (Pham, 2016).
The definitions of quality which service users emphasize are substantially different from those of the professionals or institutions providing them; for users, key components of quality are the types of services that may help them achieve improvements in their lives and the way in which those services are provided.
The methodology used to analyze satisfaction with public services is aimed, in general, at exploring in certain detail users’ attitudes toward the different facets of the service. Our study focuses on social services, which can be included in the category of ‘welfare policies’, which are usually the most relevant from the point of view of the citizen, whose tendency is to assess them as ‘very’ or ‘quite’ satisfactory (Agencia de Evaluación y Calidad (AEVAL), 2009). The negative perception always focuses on the resources that the administration allocates to them, which raises the problem of linking, on the users’ behalf, dissatisfaction and scarcity of resources. That is why studies focused on specific services are recommended. As indicated in the AEVAL (2009) report: A better understanding of the dimensions or characteristics of the Social Services that determine citizen satisfaction would contribute to deciding where to target improvement efforts, and perhaps where to concentrate resources, which would ultimately improve citizen perception. (p. 37)
If stereotypes are seen as the cognitive dimension that shapes the satisfactory or unsatisfactory attitude, and the experience of a person with a certain service is observed, this enables an accurate picture of reality (Van Ryzin, 2004). In this sense, when recording a user’s perception of a service, it increases the possibility of obtaining information that is moving away from the rigidity of stereotypes, which allows identifying its strengths and weaknesses (Del Pino, 2005).
The perspective assumed in this study is that of users, starting from the SERVQUAL (Service Quality) Model developed by Parasuraman et al. (1985, 1994), Berry et al. (1988), and Parasuraman et al. (1993). After analyzing the data obtained from their research and developing their conceptual model, they designed a measuring instrument which currently is the most widely used when assessing user satisfaction with the quality of the service received in the context of public services. It establishes how the user of the services perceives their quality, which dimensions form their perception, and, in cases where expectations play a key role, which are the factors shaping those expectations and perceptions of the service (Cabrera et al., 2010; Igami, 2005).
Initially, the authors identified 10 dimensions of quality of service to consider during assessments. After several revisions of the initial model, however, the dimension number was reduced to five, as it was found that several of them were overlapping (see Table 1).
Dimensions and statements of the perceived quality.
Source: Elaborated by the authors, based on the study conducted by Parasuraman et al. (1993).
Despite the experience amassed, and the quality plans developed by social services, in Spain no evaluation works have yet been carried out regarding the social services system, leading to possible difficulties of updating and improving efficiency (Cabrera et al., 2010; Sánchez Pérez, 2016). The lack of data about the relevance and scope of services, either from professionals or government, or from users, prevents the implementation of revitalization plans and continuous improvement, along with the resulting opportunities to enhance the quality of life of the population, which is the aim of social services.
Specifically, this article focuses on the analysis of immigrant users’ satisfaction with primary social services (dependent on the local government), starting from the scale proposed by Berry et al. (1988) and Parasuraman et al. (1993); second, the proposal involves an explanatory model of these users’ satisfaction with social services; and finally, the study determines whether there are differences in the satisfaction level according to users’ gender and locality of residence.
The model from which it originated is very simple (see Figure 1); adapted from the original, it is aimed at describing users’ perceptions of the elements or dimensions of quality they find satisfactory.
Adaptation of the model developed by Berry et al. (1988).
It should be noted that when referring to satisfaction, it is understood as an a posteriori reflection of the user’s experience with the service, and when referring to the perceived quality, it consists of the process in which the service provided and received is compared to the one the user expected to receive. For a better understanding of the results, which will be discussed subsequently, numerous differences between the three aspects deserve attention (Luo and Homburg, 2007; Martínez-Tur et al., 2005; Schneider et al., 1998; Szymanski and Henard, 2001):
Perceived quality is a stable and permanent attitude over time. Satisfaction is a transitional opinion about a service (or an aspect thereof) that can be modified at each transition.
Statements about perceived quality refer to the evaluation of the service attributes, whereas satisfaction refers to the overall service.
Statements about perceived quality are cognitive in nature, whereas those concerning satisfaction have a strong emotional component.
Method
Participants
The research involved a total of 795 immigrant-origin1 users of primary social services in 29 Spanish localities. They are predominantly women (66%), young (52.7% aged between 26 and 40 years), from South American countries (55.7%, mainly from Colombia and Brazil) and from Africa (24.3%, Morocco), married with children, with secondary education, and who do not work outside the home. They have lived in Spain for more than 4 years; 26 percent of them have lived here for over a decade, and they are the ones who have lived the longest in Spain. When they considered emigrating, they already had family and/or friends living here (79.5%), who were the main source of information about social services at the time. A total of 47 percent of them state they have work and residence permits, and only 7 percent say they have no permit; 62.2 percent of them live in localities of less than 20,000 inhabitants, and 37.8 percent in communities of more than 20,000 inhabitants.
The first time they used these services was to request information or guidance (42.2%), to look for a job (10.4%), or to ask for help and support at home (8.3%). They resort to these services on a monthly (23.5%) or quarterly (23.6%) basis.
Instrument
A scale adapted from the SERVQUAL Scale (Berry et al., 1988; Godás Otero, 2012) was used, consisting of 18 items divided into the following sections: Confidence (5 items), Responsibility (3 items), Assurance (3 items), Empathy (3 items), and Tangibles (4 items). These five dimensions explore users’ perceptions, assessing these individuals’ feelings about how the organization – local governments in this case – provides the service, as well as considering the evaluation made when they receive such service.
Responses were displayed in a Likert-type scale, with the following numerical correspondence: (5) strongly agree, (4) somewhat agree, (3) neither agree nor disagree, (2) somewhat disagree, and (1) strongly disagree.
The scale was applied to the same services, by a member of the research team, as a structured interview. The individuals in the sample were informed of the objectives of the study, how the data would be processed, and the confidentiality of the information. All of them gave their consent to participate in the study.
Results
Construct validity
In our sample, the exploratory factor analysis, after using the IBM-SPSS.20 package and following the principal component extraction method and Varimax rotation, yielded five factors that explained 76.66 percent of the variance (see Table 2). These factors are consistent with the dimensions proposed by Parasuraman et al. (1985, 1993, 1994). As observed, the factor that most explains satisfaction is responsibility, followed by assurance and empathy; on the other hand, the factor that least explains it refers to tangibles.
Exploratory factor analysis – percentage of variance explained according to each factor.
The data proved that the perception of responsibility arose from the fulfillment of promises to do things and keep appointments, meeting of schedules, and interest in solving users’ problems. The perception of assurance which service professionals convey to users is shown by the support offered, compliance with appointments, and the willingness to assist clients. Along the same line of research, Bueno and Pérez (2000) argued that, in the social realm, what emerges from a symbolic evaluation of human value is not necessarily based on respondents’ education or training, but more likely on their own personal experiences. Other studies, such as those conducted by Medina and Medina (2011) or De la Peña Salas (2014), reached similar conclusions, using the same scale with a sample of users of primary social services.
The reliability indexes recorded are summarized in Table 3 and, as shown, are suitable for both the entire scale and for each subscale individually (Rastgoo, 2014).
Reliability indices of the scale of users’ satisfaction.
In short, despite the cutbacks caused by the economic crisis and the ominous predictions regarding the development of the social services system (see Mänttäri-van der Kuip, 2014), the satisfaction of immigrant-origin users is based on responsibility, assurance, and empathy, dimensions which are focused more on professionals than on the material aspects of services (tangibles). This may also be explained by the nature of the instrument itself, since according to Kang and James (2004), the SERVQUAL model is focused on the process of service provision rather than on other characteristics of the service, such as its techniques.
Proposal of an explanatory model of satisfaction
Once the construct validity and reliability of the scale were analyzed, a confirmatory factor analysis (CFA) was performed using the SPSS software Amos 20, estimating the parameters of the original model under the criterion of maximum likelihood, in order to verify the adequacy of a five-factor model. This model is shown in Figure 2, where standardized regression weights can be observed, as well as covariance between factors (all are significant p < .01). Table 4 shows the fit indices.
Confirmatory factor analysis model for the scale.
Indicators of goodness of fit of the model of satisfaction of users with social services.
CI: confidence interval; df: degrees of freedom; GFI: goodness of fit index; CFI: comparative fit index; RMSEA: root mean square error approximation; SRMR: standardized root mean square residual.
In our opinion, the interpretation of these indices requires experts in the field, thus a brief description thereof is presented below: χ2 reports the distance between the sample and the hypothetical variance/covariance matrices (Bentler and Bonett, 1980). This value should not be significant if the intention is to conclude the non-existence of a discrepancy between the two matrices. However, this statistical value is very sensitive to the size of the sample (with n ⩾ 200 it is usually significant). That is why it should be accompanied by other indices, the most commonly used being the goodness of fit index (GFI) (Tanaka and Huba, 1985), which takes values between 0 and 1 and may be interpreted as a multivariate coefficient of determination. A value higher than .90 indicates a good fit; CFI (Bentler, 1990), which is a comparative fit index, whose values between .90 and .95, indicate a good fit; root mean square error approximation (RMSEA; Steiger, 1990) accounts for the difference between the population correlation matrix and the one proposed in the model of the sample used. In this case, values lower than .08 indicate a good fit; and standardized root mean square residual (SRMR; Hu and Bentler, 1999), which reports the standardized residuals and where, as in the previous case, a value lower than .08, indicates a good fit.
According to these data, the proposed model confirms the existence of five factors to predict users’ satisfaction with social services.
Users’ satisfaction according to gender and locality of residence
In general terms, one could state that users are satisfied with social services, since the means of the five factors exceed a value of 4. The lowest mean corresponds to tangibles (mean (M) = 4.08, SD = 99) and the highest one to responsibility (M = 4.45, SD = .78).
However, it is understood that gender and the size of the locality of residence (<20,000 and >20,000 inhabitants) could make a difference in these results. To determine whether there are differences based on these two variables, a factorial analysis of variance (ANOVA) 2 × 2 (gender × locality) was conducted, taking the five factors of the scale as dependent variables. The results are shown in Table 5 and Figures 3 to 6.
ANOVA gender × locality.
p < .05;
p < .01.
Gender versus locality interaction in responsibility.
Gender versus locality interaction in assurance.
Gender versus locality interaction in empathy.
Gender versus locality interaction in confidence.
The following should be noted:
There is a main effect of gender on all variables, always to the advantage of men, except for confidence and tangibles.
There is a significant gender × locality interaction in all variables except for tangibles.
There is no main effect of the ‘locality’ variable in any of the factors analyzed.
Given that the interactions were significant, post hoc contrasts were performed according to the size of the localities and it was observed that there were no significant differences between men and women living in localities with less than 20,000 inhabitants. However, the differences were significant in all four factors analyzed between those living in localities with more than 20,000 inhabitants (tResponsibility (335) = 2.29, p = .022; tAssurance (335) = 3.16, p = .002; tEmpathy (335) = 3.30, p = .003; tConfidence (335) = 2.18, p = .030). In all cases, the mean obtained by men is higher than that obtained by women.
This may have to do with the different motivations that lead men and women to turn to these services, but also with the dimensionality of the services. For men, the motivation to turn to these services is very specific and basically it is work-related, whereas women go looking for information or guidance on more heterogeneous issues. The programs and services on offer correspond to the size of the locality, and in smaller ones there are usually no more than a few professionals.
Conclusion
It is a well-known fact that, in advanced democratic societies, there has been a growing awareness of civic rights, affecting the quality of public services. As a result, the use of assessment processes about its effectiveness is necessary. In the case of social services, it is important to have a model to which to refer the continuous improvement of programs, actions, and resources, combining technical and relational quality. The SERVQUAL model serves this end insofar as it increases the role of the users and/or the people linked to the system as a whole (Buttle, 1996).
In addition to helping overcome an insufficient evaluation culture, these processes are basic if the aim is to move toward quality social services, essential for the achievement of a more cohesive society, committed to the well-being of citizens.
Specifically, we have developed a model for measuring satisfaction with social services among immigrant users, using a scale of five (5) factors and a total of 18 items: responsibility (3), assurance (3), empathy (3), confidence (5), and tangibles (4).
It was found that the satisfaction of these users depends on three dimensions: the perception of responsibility in the organization, the assurance that workers convey (professionalism), and the empathy that they display in relation to the problems faced by people of immigrant origin (Moshref Javadi et al., 2012; Witkowska and Lakstutiene, 2014). The tangible aspects have the least effect on the satisfaction of these users (Mañas et al., 2008; Purcarea et al., 2013).
The study confirms the prevalence of the human factor over other types of noteworthy facilities or aspects of the care provided by the social services to immigrants. The logic of well-being becomes visible in the relationship between user and social worker. It becomes clear that the perception of assurance that professionals convey to social security users derives from the extent to which they provide assistance, comply with their appointments, and show willingness to help (Pham, 2016).
On the other hand, users’ high perception of empathy when interacting with professionals depends on their courtesy and preparation, and the assurance they display when solving a problem. Moreover, the perception of responsibility is based on professionals’ compliance with appointments, adaptation to their schedule, and interest in their clients’ daily challenges (Pham, 2016).
However, our study has limitations, which include the exclusive use of self-report as a source of information, mitigated by the consideration of indicators about social services, published by local or national authorities (Sánchez Hernández, 2008). The cultural differences in the perception of users are not taken into account either, despite the fact that the individuals in the sample have different origins (Malley and Fernández, 2010; Oliver and DeSarbo, 1988; Szymanski and Henard, 2001).
There are obviously factors that affect the degree of satisfaction with a service, and a relevant one is the technical support that the measurement instrument may have. With its advantages, or strengths, the choice of SERVQUAL (Parasuraman et al., 1985) also involves the assumption of criticism, such as focusing on the use of divergence when assessing the service quality, or the use of a seven-point Likert-type scale, the excessive use of negative words, or the same vocabulary of the items (Buttle, 1996; Gimeno and Ruiz-Olalla, 2001). During the study we have certainly not blocked out all this criticism for the sake of an easily understandable measurement instrument.
Another aspect that undoubtedly affects the scope of satisfaction is the context of the service itself, in view, above all, of the relevance that the emotional aspects of the quality of life of the users acquire, which manifest themselves in the same assessment (Peiró et al., 2005; Price et al., 1995; Sánchez Hernández, 2008).
Finally, if satisfaction is focused on users’ perspectives, there is a risk that the assessment may overlap those related to the quality of the service, which refer to an external reality (Carman, 1990; Gotlieb et al., 1994). That is why, in our study, users’ satisfaction should be understood as a part of and complementary to other activities linked to the quality of service.
Footnotes
Funding
This work was supported by the research project subsidized through publication of a competitive call by the Government of Galicia-Xunta de Galicia (Spain), ‘The quality of social services seen by immigrant users and professional teams: evaluation and socio-educational consequences’ (10PXIB211120PR).
