One of the principal governmental concerns in member countries of the OCDE over the last few years has been the improvement of public administration management in two ways: on the one hand, increasing effectiveness and efficiency of the public sector by reforming management processes; and on the other, improving relationships between administration and citizens. This reform has also become transformed into an economic demand for interdependence between the public and private sectors, and for policies of keeping down costs and maintaining social order in the face of growing pressure from the users of public services.

The public service concept is so wide that it may be convenient to begin this presentation by defining it. In general, we understand by a public service one that has the whole set of citizens as the final addressee, provided that public authority makes the commitment to supplying it, either directly (using its own resources) or indirectly (delegating the production of the service to private companies and reserving the right of control).

Hence, in this work we are dealing with social services as services in which different public administrations play a prominent role in the supply decision, setting the ambit and conditions under which provision is made, as well as its funding system.

During the last 15 years we have witnessed an unprecedented growth in social services in Spain. This growth can be explained by the leading role acquired by activities related to the development of social services as an axis of electoral programmes in the first years of the post-Franco democratic consolidation period.

In this period, numerous initiatives in this field were taken at all levels of administration, in many cases justified by the existence of inherited needs and by the desire to match our service levels with those of other European countries.

Consequently, during these recent years, the objectives of public authorities have had a quantitative, rather than qualitative, character, giving priority to policies oriented to extending the availability of services over those directed to improving their quality. Moreover, in many cases, the extension of services has been made without adequate planning of means, and to the detriment of other actions essential for attaining the projected goals. In summary, from a political point of view, social services quality has not been -nor is it at present- the priority objective.

It is necessary to recognise that -although structured from a theoretical point of view- in the last draft of the Plan Concertado de Prestaciones Básicas (Co-ordinated Plan for Basic Provisions) we find as one of its goals the will to provide citizens with quality social services suited to their needs. In another paper I referred to this matter, pointing out the incoherence this objective represents if it is not followed up by the corresponding decision to introduce a quality management plan in those institutions providing social services (Medina Tornero, 1994).

It can be said that, currently, there is a change -albeit slight- of trend towards the implantation of new and more effective managerial methods. However, what does effectiveness mean in the social services domain? It is said that an organisation is effective when the goals assigned to it are systematically achieved. In the social services context, and in the absence of a concept of profit, these objectives must be marked by the establishment of standard conditions for service - that is, effectiveness becomes the achievement of pre-established quality levels during the service production process. Hence, the effectiveness concept has a qualitative character, and must have a great deal to do with quality.

In contrast, efficiency relates services or goods obtained to production costs. Between the two, effectiveness and efficiency, we will reach the equivalent of the quality concept.

In recent years, there has been a slow and progressive change in citizens’ mentality with respect to their expectations about public services in general - a trend which is beginning to become evident in the users of social services: while previously there were nearly no complaints about the functioning of social services, we have now arrived at a situation in which the citizen’s demand for more quality services has become clear. In the opinion of Ruiz Expósito (1996), various circumstances have come together to foment this change of mentality:

There are multiple reasons why we should look into the subject of social services quality. We agree with Wells and Brook (1988, in Medina Tornero, 1996) on those below:

The conception of the citizen as a client of services provided by public administration represents a 360-degree turn as regards the state’s legitimating political base. Relations between citizens and the administration cannot be exclusively circumscribed to the moment of voting. That would not be sufficient to establish the public administration’s new political legitimacy; nor adequate or recommendable to properly estimate the impact of public actions or citizens’ degree of satisfaction with public services (López Camps, 1995). The aim is to develop and apply new managerial ideas allowing the abandonment of the notion of relationship between administrator and administered, in order to conceive of the administration as a service whose client is the people (OCDE 1991).

However, it should be taken into account -as López, Camps and Gadea Carrera (1992) correctly pointed out- that an important difficulty for advancing in this wide criterion of modernisation, combining interests of economic efficiency and social efficacy, is the lack in many, especially local, organisations of adequate instruments for carrying out the follow-up, control, and evaluation of public policies and services.

The current interest and concern about social services quality in Spain, and about the measurement of institutional effectiveness in services, has led to the existence of tentative evaluation projects which, despite many difficulties, are bearing their first fruit. No-one now denies the importance of evaluation, and its social function is recognised by those holding political or executive office. Evaluation is beginning to be something more than a merely academic matter, since it has succeeded in the fields of social policy and public administration (Medina Tornero, 1993a).

It is clear that methods for the evaluation of governmental action programmes are useful for facilitating better knowledge about them, both on the part of the public and on the part of those involved in public administration itself. However, the objectives of evaluation are often unclear, and it becomes important to determine, as far as possible, what is really intended to be found through the evaluation process (Palumbo, 1987).

From a managerial perspective, and starting from the hypothesis that any policy leaves traces in its implementation that allow the study of its impact, Subirats (1992) considers that the questions to be answered concern the policy’s effectiveness, the degree of satisfaction achieved and its level of efficiency.

Ballart’s (1993) thoughtful work points out some of the factors that may explain the difficulties our country encounters in introducing evaluation in institutions providing social services. These are related to a poor applied research tradition on the supply side and, on the demand side, to the characteristics of the political-administrative system.

As regards the first point, interest in applied social science and, particularly, in analysis and evaluation of public policies on administration problems, is recent, which means there is an important lack of qualified personnel capable of carrying out this kind of research.

As far as the second point is concerned, obstacles likely to hinder the development of evaluation are many:

Firstly, the ideology of general interest justifies any policy or programme legitimately decided, with no need to submit it to formal evaluation in order to ensure that its effects agree with its intentions (Nioche, 1982a).

Secondly, the political and, above all, bureaucratic élite have been trained in an institutional tradition dominated by administrative law, in which the main concern is control of the legality of administrative actions.

At this point, it should be noted that the little evaluation so far developed is mainly focused on quantitative aspects and dimensions, with a significant dearth of evaluations dealing with the quality of services. Only in the fields of health, traditionally, and education, incipiently but intensively, can assessments of service quality be found. It must be recognised that such concern has not yet extended to the area of social services.

Before looking into other aspects, it would be convenient to make some important things clear. What is quality? Do we all understand the same thing by quality?

Quality ....we know what it is, but we don’t know how to express it! What the hell is quality?, asked Pirsig (1984). Indeed, this is a question that has been debated and discussed since the beginning of time. However, there is no definite answer. The basic problem is the lack of theory on quality and the lack of agreement on the concept itself. Why is the quality concept so slippery? The main reason is that we differ in our concepts of quality. Quality is in the eye of the beholder, in the client’s and user’s experience; it has different meanings for different people.

The same thing happens with the term quality as with many other terms: its widespread diffusion has made it fuzzy, and it has taken on multiple shades of meaning (Gento Palacios, 1996).

The Spanish Espasa encyclopaedic dictionary (1992) defines quality as the inherent property or set of properties of a thing that allow its appreciation as equal, better or worse than the rest of its kind.

Focusing on the field of organisations, public institutions or companies, quality has been defined in several ways:

Juran, J.M. (1988) considers product quality as its suitability for use by those for whom it is intended; Crosby, P.B. (1979) defines quality as the accommodation to clients’ demands; Deming, W (1981) states that quality consists in the contribution to the satisfaction of clients’ needs; J. Horovitz (1992) understands quality as the level of excellence that the company has chosen to reach, in order to satisfy their key clients and, at the same time, the degree to which such quality is achieved.

What do we understand, then, by quality in social services? The American Society for Quality Control (ASQC), defines quality as the totality of functions characteristic of a product or service that determine its capacity to satisfy the needs of a particular group of users. This is, no doubt, a valid definition but surely does not satisfy everybody involved in social services, since it places special emphasis on the service user’s needs, which, in the case of social welfare, becomes problematic: for instance, professionals and users do not always agree on the definition of welfare or quality of life.

Adapting the Joint Commission on Accreditation of Hospitals (JCAH) definition, we understand quality in the sense of: achieving the best social welfare benefit with the minimum risk and the correct use of resources, in a satisfactory way for the user. Going on with our definition of quality, we could state that, in a wide sense, it would include:

In other terms we would say that the quality of the service’s programmes or provisions should involve:

We would suggest that there is another difficulty to add to what has gone before: it may be that an analytical definition of the quality concept is not possible, since it is closely linked to our subjective needs, and these are so changeable and diverse that it is necessary to come to a practical agreement on what we call quality and on the domain of application. In other words, and in order to understand one another, we must arbitrarily determine what is quality in the interventions we are looking at; that is, we must formulate operative definitions. This is of incalculable importance, since it allows us to delimit, according to our needs, the quality concept in each situation, for each problem (A. González Dagnino, 1994).

Some topics appear in our social services context that we need to determine and take into account; for example, those related to:

Working from the argument that public administration duties are not directed towards profit or gain, bad quality in management and attention to citizens can be justified. This idea is false. Every quality concept can, and must, be applied to public administration. Andrés Senille’s (1993) work illustrates the main reasons for introducing services with quality criteria:

1. To ensure the continuity and future of the team.

2. To reduce the debt and be economical.

3. To improve internal client service.

4. To improve external client service.

5. To integrate the client into the organisation.

6. To improve the image.

7. To be more competitive.

8. To improve management.

9. To obtain results in terms of objectives and economy.

10. To promote the leadership style.

When the topic of quality in social services is debated, the great problem arising concerns the way to measure it. In private service companies the acceptance or non-acceptance of the product is a quite reasonable system of controlling quality, and its indices and measures have been developing over many years. However, in public administration this measuring tradition does not exist, nor are the standards on which it should be based very clear (Senac Azanza, 1995). In Spain, the Observatorio de Calidad de los Servicios Públicos (Public Services Quality Observatory) constitutes the first attempt to make measurements. In spite of many difficulties, it has eventually permitted a numerical expression of improvements in quality in certain public services.

If it is true that quality analysis was primarily oriented towards tangible products, it is also true that, in the last decade, there has been a considerable bibliographical contribution to the quality of services in general, of which (in the absence of the social services analyses), we have reviewed those in the health domain, by authors such as Vuori (1988), Donabedian (1984), Heather Palmer (1990), etc.; and in educational services - J.D. Wilson (1992) and Gento Palacios (1996), for example. We did not neglect quality strategies coming from companies, but reasons of space prevent us developing more arguments on the topic.

In the research line we are developing for the social services domain, in particular for local administration services, we started from A. Donabedian’s conception on the methodology of evaluation of medical attention, and R. Vuori’s on quality control in sanitary services, in order to deal with the problem of quality in social services; we later introduced, for some specific aspects on assistance quality, the approach developed by Dr. Heather Palmer of Harvard University, since it deals with the development of the quality concept based conceptually on the establishment of a quality guarantee system. This approach, in turn, followed that of the Washington Institute of Medicine, which defined a system of quality guarantee as one that makes medical assistance more effective, improving health levels and people’s satisfaction levels, with the resources that individuals and society have agreed to devote to health. This position, which focuses on medical assistance, implies a primarily healing conception of quality, expressed in the analysis of different dimensions in the quality concept which easily adapt to the social services domain:

Many authors use these concepts of quality dimensions, established by the Harvard group, adding or modifying aspects which do not essentially change it. For instance, some have complemented the analysis of the quality process with procedures to detect and prioritise problems of quality, and to evaluate interventions that may derive (all based on general planning techniques). This evaluative process approach allows the focusing of quality analysis on specific aspects, or even on a particular individual. For example, the result of a given intervention and/or treatment, the acceptability of a public-attention timetable, the accessibility of a given centre, or the competence of a given worker compared to others. This results in considerable improvements in attention to users. In summary, the specific, partial and focused character of this conception of quality makes it especially apt for use in the area of social intervention, in social services centres, in individuals with particular and special needs, and for specific situations of social emergency.

Obviously, in the absence of methodological propositions oriented towards social services, we had to select new procedures, and we thus considered a series of key questions: Exactly how does the user evaluate the quality of a service? ¿Does he/she make a global evaluation, or rather judge specific facets of the service in order to reach a global decision? If the latter is the case, what are the different facets or dimensions used to evaluate the service? Do these dimensions vary depending on the different services? If user expectations play a fundamental role in the evaluation of a given service quality, what are the aspects that form and influence those expectations? (Medina, 1993b).

In order to measure quality of services, it is necessary to establish certain requirements which refer to characteristics that are detectable by the user, measurable and, hence, able to be assessed and controlled. Below are some examples:

Delays, waiting time, number of phone calls, visits and interviews for solving or dealing with problems, precision or correctness in performance of tasks.

Aesthetic aspects of the organisation/institution, environmental aspects, hygiene, security, organisational ambience, friendliness, comfort/convenience, interpersonal relation level.

Response capacity to demands, response to the unexpected, people involved in the service provision process, length of a given service, waiting-list/delay rate, complaints and claims system.

User-institutional worker communication, level of information adequate for circumstances, service reliability, mutual satisfaction obtained, competence of persons intervening in the process.

Some people prefer to talk about two types of quality indicators: objective indicators, fixed by the organisation according to quality management, and subjective indicators, referring to quality perceived by the user/client. Organisations draw up service designs including quality parameters, but, on the other side, the client has expectations, needs, beliefs, and so on, about how the service should be. The institutional objective should be towards a matching of the two types of indicator. Organisations must put themselves in the client/user’s place, see through his/her eyes; otherwise, they will not be working within the framework of a search for quality. There are two types of access to subjective indicator: research on users and claims and complaints.

According to Berwick (1991), to improve the quality of attention it is necessary to have specifications of the process that are clear, scientifically proven and periodically reviewed, indicating the most effective guidelines and steps in a given situation.

In social services, there is no practical guide that allows the attainment of stable quality levels and guarantees that the obtention of quality services by the user will not be merely a matter of luck.

We are aware (Medina, 1994) that when professionals and social services institutions begin to discuss quality specifications, the road to consensus will be a long and a hard one. However, it is becoming more and more urgent to obtain definitions of quality for social services in relation to specific topics, since measures cannot be designed in the absence of criteria, and no improvement can be targeted without measures. Put another way, without specifications, attention procedures for users will continue to be extremely variable, and populations will continue to receive a quality of attention dependent on luck.

The consideration of user satisfaction is becoming more and more important in assessing quality of attention. It is also important to be aware that the instruments used for measuring it have not yet been definitively standardised, due to different conceptions of quality, the diversity of environments in which they have been applied (primarily in English-speaking countries), and the different methodological approaches used in their application.

Since the first investigations, mainly those of Hulka (1975), and more recently Ware (1983), in the United States, numerous studies in the field of satisfaction analysis have been carried out. Spain has been no exception, especially in the field of health, both at the level of hospitals and of primary attention, though different methodologies have been applied.

User satisfaction with regard to attention received should be conceived as a quality control measure; thus, its study and evaluation is important for the management and working practices of the social services, with the aim of achieving a better adaptation to user needs and desires.

Authors such as Howwell (1976), Fleming (1981), Ware (1983), Suñol (1987), Vuori (1988) and Fitzpatrick (1991) have underlined the importance and pertinence of including user satisfaction with services as part of and complementary to other quality control activities.

Most user satisfaction measures refer to his/her opinion on the attentional process received. The method underlying most of the procedures measuring user satisfaction involve theorising about the way this complex phenomenon may divided into concepts, then designing surveys to evaluate responses to these concepts; this, in fact, imposes a pre-conceived scheme on the interviewees. We found only one approach, that of Ware and Snyder (1985), that used, in the health domain, a method for measuring satisfaction in which it is intended to discover how interviewed users structure their own perceptions on assistance received. Ware (1983), on the basis of documents, confirms the need for a balance between affirmative and negative statements on the different concepts within satisfaction questionnaires. Such an equilibrium greatly helps to counterbalance the potential bias produced by acquiescent responses - that is, the tendency to answer affirmatively to all statements, regardless of their content. It is often said that one of the characteristics of social services users is his/her ignorance. This is particularly true in the case of people with low educational level, such as the majority of home help (Servicio de Ayuda a Domicilio) users (Medina Tornero et al., 1995).

However, other studies have shown that users have more and more sophisticated quality criteria with respect to assistance received. These criteria may emphasise aspects which professionals judge to be less important or irrelevant for quality assistance. This is quite understandable, since professionals and consumers have different implicit and explicit goals which are not necessarily symmetrical or reciprocal (Buck, 1984). Even so, patient’s/user’s satisfaction is an essential qualitative criterion in professional activity. Moreover, there is evidence that criteria applied by users often match those of professionals.

Service quality, from the user perception point of view, may be defined as the amount of discrepancy or difference existing between users’ expectations or wishes and their perceptions. Zeitham, Pasuraman and Berry (1993) suggested this definition in their interesting research on clients’ opinions about a set of companies and services. There is, however, a growing awareness of the importance of client’s satisfaction as a determinant element of good quality assistance.

The commonest method of collecting patients’ opinions is the survey, but complaints presented by them can also be studied. As T. Peters (1987) pointed out, any user’s complaint is a sign of a certain deficiency, and usually provides a good opportunity to improve the criticised service. Structuring a system which collects, codifies, and studies claims, complaints, and critical opinions about actions, demonstrates the maturity of an organisation willing to work for the establishment of quality.

Continuing this argument, we could say that two types of approach would exist. On the one hand, those giving rise to what we call logical quality (Donabedian, 1966), or the efficiency with which scientific-professional knowledge is used within the intervention domain, be it a social service, a residence for the elderly, or whatever; and on the other, what we define as felt quality, or quality as it is perceived by the user.

This distinction becomes especially interesting if we take into account that what users understand by quality -and therefore expect- in social services does not always match what professionals consider as quality service.

A different approach to the quality concept is that of those authors who try to define it by analysing its contents. For example, A. Donabedian (1966) considers that, when talking about quality of assistance, we should mention:

H. Vuori (1982), meanwhile, states that quality of assistance should provide services with the following characteristics: accessibility (availability of resources for the user, wherever she/he needs them), appropriateness (adjusted to user needs and expectations), effectiveness (optimal results achieved), efficiency (effective at a reasonable cost), co-ordinated between the different levels and providing satisfaction for those producing and receiving them.

According to Juran (1988), every organisation must knit together three basic activities to reach increasing quality levels. These basic activities, known as Juran’s Trilogy: planning, measurement (control) and improvement of quality, are intimately interconnected, and cannot be understood separately. The measurement system (information) is the pivot sustaining the planning on the one hand, and the improvement on the other; it has no sense in itself, but it is essential to have measures for planning and improving services that we provide to users.

Planning of services without data on user needs and expectations, characteristics of available resources or effectiveness and costs of processes, for example, is unthinkable. In fact, the design of the measurement system, conceptually, forms part of the quality planning stage. That is, there should be no strategic and operative planning objective, no social intervention programme, no proposal for change, that does not include data about the starting point (current quality level) and results it could be expected to reach (quality level attainable), and that does not systematically measure (monitor) the key points in the new process and the results obtained from it.

To do this we need permanent information about structure characteristics (needs and expectations of external -users- and internal clients and state of material resources), about functioning of key processes and about results obtained.

Knowing user needs and expectations allows us to focus planning on the user, not on resources, as is usually the case. Knowing, concurrently, that key processes are working allows us to detect (before bad results come in) whether they are deviating from the expected quality functioning level. Results measurement will allow us to know, in a global sense, if we are reaching the desired quality level.

Measurement is an essential need if we want our services management to be based on the facts, on demonstrable realities, rather than based on intuition that is influenced by our own prejudices and interests. Measurement is a way of combating a deeply-rooted cultural feature in organisations: unsubstantiated opinion (making firm statements without supporting data).

Nor is it appropriate for data (not always valid and reliable) to be used as a weapon among different layers of the organisation, used for apportioning blame. As Berwick (1989) says: it must be recognised that most of the data used to design significant quality measures, in an institution or a professional or social services company, is based on, or collected by, the workers themselves. If they perceive these measures as alien, improper or manipulated, they will lose reliability. The social services system well knows what happens when quality control is made with a "foreign" inspection philosophy. That is, when it is taken out of the context of the planning and improvement activities to which it must necessarily be associated. This defective approach to quality measures may obviously lead to great resistance from workers - resistance which may be quite problematic. On the other hand, it should be underlined that the design of quality measures is a difficult task. While not impossible, it is not as easy as it may seem, if we consider the amount of attention given to the design and validation of indicators, when it is assumed that we should be guided by planning and service improvement. The question is not to have many measures whose validity and reliability are dubious, but to have valid and reliable key-measures at one’s disposal.

A final aspect to be considered is the interpretation of indicators. Even if we assume they are valid and reliable, each one has a certain sensitivity and specificity. They may detect false failures of quality (false positives); they may also fail to detect real quality failures (false negatives). In order to interpret results, we must know the range of sensitivity and specificity values. Another factor to be taken into account is the influence of chance in measures usually taken from sampled cases.

In conclusion, measuring quality levels is essential. However, measuring is not easy: a valid and reliable system must be developed, and one must also learn to interpret measures and to use the system in a positive way in order to guide services planning and improvement towards user needs and expectations.

Why measure quality in the social services? Denia O’Leary (1993), president of the Joint Commission on Accreditation of Healthcare Organizations expressed it quite clearly, and we have adapted it to our professional field:

"We measure assistance activity for two basic reasons. First, we measure in order to make judgements and decisions.

Second, we measure in order to make a base for future improvement. The permanent improvement of assistance activity is a basic expectation in professionals and organisations. Every improvement opportunity has its initial base in effective measurement. Effective measurement requires some knowledge of what is relevant and important to measure, measurement instruments availability, an ability to apply the tools, and the will to correctly measure the right things".

There are several ways to measure quality in services:

Every kind of indicator -except strategic ones- intends to measure quality of services, either from a finalist (results), managerial (process) or initial resources (structure) perspective, following classical methodology. Any good management control system must provide information on quality of services given by the local council at all levels: quality in the structure, in the process and in the results. In this case it must be very clearly acknowledged that the key to success in management of quality lies in citizens’ satisfaction: this is the final "great result" in municipal actions. Quality controls in local administration must measure the state of quality attributes present in municipal service provision. According to López Camps and Gadea Carrera (1992), these attributes are:

Reliability. A good service is provided every time.

Response capacity. Things are done on time and quickly.

Competence. Municipal workers must be competent.

Accessibility. Services are convenient and easy to obtain.

Courtesy. Citizens are treated with kindness and attention.

Communication. Users are informed in clear language.

Credibility. There is knowledge of and interest in citizens’ problems.

Security. Citizens’ safety in the use of services is guaranteed.

Comprehension. Citizens must perceive that the council understands their problems.

Physical support. The environment where the service is provided must be a quality one.

Once more, these authors are inspired by the work of Donabedian in developing their arguments; thus, many of these attributes refer mainly to structure (physical support, competence...), others to the process (reliability, comprehension...), and the majority to results (user satisfaction). Quality indicators selected (structural, procedural, or based on results) should collect the greatest amount of information on each of these attributes.

In a classic -although recent- text by Nutting, Burkhalter, Carney and Gallagher (1991), a series of methods for evaluating quality of assistance in primary health attention is presented. No less interesting than the content on methods is the preface to the Spanish edition, written by Pedro Saturno, a veteran expert in quality analysis who, under the title Métodos de evaluación de la calidad asistencial en atención primaria: mitos, trucos y trampas (Evaluation methods for quality of assistance in primary attention: myths, tricks, and traps), gives us the opportunity to avoid being dazzled by methodology per se, when it is not accompanied by the corresponding logical and operative support. I have decided to include here some of his considerations, which are of great interest, especially for those new to the subject of quality evaluation.

Saturno suggests that evaluating quality involves, in general, measuring what exists, comparing it with what we should find, and looking for the way to shorten the distance between reality and desire (or obligation). The problem is that quality is a multi-faceted entity, susceptible to various approaches and definitions -sometimes cryptic, sometimes biased, sometimes, the best ones, more eclectic and conceptually clear- none of which, however, guarantee evaluative specificity.

Saturno presents a set of myths and their corresponding traps; he begins the epistemological route with the myth of global quality, the listing trap, and considers that the most common attitude is for interested people to search passionately for a list or for "the list" of criteria, indicators or things, thereby solving the problem of measuring quality. It must be noted -and here is the trap- that there are so many and varied activities that may be conceptualised as assistance quality, that it is illogical and unfeasible to try to enclose the matter within any fixed category.

Another myth is results evaluation, and is accompanied by the apparent validity trap. He suggests there is a myth of evaluation of results based on data, and points out that this myth can only be sustained through a lack of reflection on the validity of measurements, which accepts as valid, without discussion, any results criterion, on the basis that, in the end, assistance is for producing results. He demonstrates -with no lack of criteria- that it is an artefact to separate process and results by giving an a priori validity to all types of data.

The third myth, called the brilliant method, with its corresponding trap, unfeasibility, is a call for humility. Not everything proven from an experimental perspective is proven with a universal character. Saturno thinks that, unfortunately, to consider only the theoretical arguments may lead us into the trap of trying to use or recommend an unfeasible or barely practical method.

Professor Saturno concludes his arguments with a phrase which may pass into the history of methodology: Against myths and traps: the practical validation trick. His view is that any assistance quality evaluation method will be valid from a practical point of view only to the extent that is feasible and leads to changes in the way of doing things and in the results of assistance. Only in practice will the virtues and faults of the chosen approach become obvious. Following this view, it is necessary to adopt and adapt the methodology we think most useful for us, and to base our judgement of its utility on the results obtained through practical application.

One of the most frequently used methods for evaluating quality is that known as Audit, with implicit and explicit criteria, which has been widely developed in the health domain. Currently, we are trying to develop an adaptation of it, for application to social services centres. Below, we briefly present some of the arguments and the variables being used. From this perspective, it is understood that the quality of a service must be evaluated in terms of three different dimensions:

Taking into account all of the above, we can state that the quality of a service can and must be objectively measured, and can be designed before the act of assistance. It should be borne in mind that, in order to be effective, a model involving the consideration of these data should fulfil a series of guarantees:

Impact on target population.

Impact on user’s levels of accessibility to services.

Impact on launching costs of the plan.

Possible indirect effects on other kinds of services.

An adaptation has been made to test the effectiveness of the quality control theory perspective, inspired by Vuori (1988), on the evaluation of a social services centre, using the following variables:

The greatest problem found is that corresponding to the lack of criteria and standards for making quality evaluation measures: the official/administrative absence of what is understood, or what is wanted to be understood, by "good assistance" or "quality level", and of implicit and/or explicit criteria and normative standards or criteria that would serve as official referents. Hence, sets of standards have had to be constructed based on experience in quite specific services domains, and corresponding to opinions about users’ satisfaction level.

On consideration of the design of quality evaluation methods in the social services, we have also taken into account -though as yet not completely developed- an adaptation of R.H. Palmer’s proposal (1990) for the evaluation of primary assistance. In short, it would have the following dimensions, which we should suitably put into context for organisations providing social services:

Identification and sampling of cases:

Addressing the subject of quality, either from the evaluation perspective -as in this work- or in terms of the establishment of a system of management, of continuous improvement, of control or of total quality, opens up a universe of possibilities, many of them new in the social services domain. For reasons of space here we have been forced to leave out some very interesting aspects related to quality, such as: accreditation criteria and more detailed standards for social services; quality manuals and procedure guides; human resources and their central importance for the operative and logistical development of quality; organisational structure -work ambience, culture, leadership, interpersonal relationships, motivation, the role of political management, and so on. We should not sign off without a reflection on the evaluation of the costs incurred by a quality system, and a final comment on the need to train employees in the use of the techniques and tools that contribute to the development and implantation of quality in social services organisations.