The motivations of volunteers, that is, the reasons for their decision to devote part of their time and energy to helping people unknown to them and for relatively prolonged periods, appear in the psychosocial literature as the most important factor in explaining differences between volunteers and non-volunteers, and between volunteers that continue and volunteers that abandon their activities (Penner and Finkelstein, 1998; Omoto and Snyder, 1995; Black and Di Nitto, 1994; Clary and Snyder, 1991).

However, so-called "burnout syndrome" among volunteers has begun to be studied only quite recently. López-Cabanas and Chacón (1997) found that between 1989 and 1996 there appeared only five works in the principal psychological and medical databases that studied volunteer burnout directly. Three of them compared the variables that facilitate burnout in professionals and volunteers (Capner and Caltabiano, 1993; Glass et al., 1992; Paradis and Usui, 1989), while Lafer (1991) studied burnout among volunteers working in hospitals, and Cyr and Dowrick (1991) studied the syndrome in crisisline volunteers.

Burnout has traditionally been studied in relation to health professionals leaving their jobs, and can be defined as "a response to chronic emotional stress whose principal features are physical and psychological (emotional) exhaustion, a cold and depersonalised attitude in relations with others (patients) and feelings of inadequacy in the face of the tasks to be carried out (reduced sense of personal realisation)" (Maslach and Jackson, 1982).

Some aspects of the work of health professionals facilitate the appearance of burnout syndrome. Among them those worthy of mention are the following (Pines and Aronson, 1989):

Volunteers that work face-to-face with AIDS and cancer patients, like professionals working in the same field, are subject to great emotional demands, given the seriousness of these illnesses; in the case of AIDS there is the additional factor of social rejection of the patients. These considerations justify the study of burnout in relation to permanence or abandonment in this type of voluntary work.

In Spain the issue of volunteer permanence has only recently begun to be studied in an empirical way. The Social Psychology Department at the Complutense University in Madrid opened up, some two years ago, a line of research aimed at determining which psychosocial factors are related to greater or lesser length of service of volunteers in organisations.

Our first objective was to validate Omoto and Snyder’s (1993; 1995) Volunteerism Process Model. This model identifies three stages in the volunteerism process: antecedents, volunteerism experience and consequences of volunteerism.

The first stage includes the variables that can be considered as antecedents of volunteer behaviour, among which are the motivations that lead people to seek and become involved in voluntary work and that result in their remaining involved over a period of time.

Omoto and Snyder, following Functional Theory –initially postulated for attitudes ( Smith, Bruner and White, 1956; Katz, 1960) and later applied to the motivations of volunteers (Clary and Snyder, 1991)– include five different motives that may explain volunteer behaviour: values, personal development knowledge, personal development, esteem enhancement and community concern.

According to this theory, in volunteerism there coexist a wide variety of motivations that fulfil different functions for individuals. Different people may become involved in the same voluntary activity for different reasons, which may fulfil a range of psychological, personal and social functions; these reasons, in turn, may change for the person over the course of his/her involvement in the voluntary activity.

When voluntary work is based on an altruistic interest in helping others, on humanitarian values, on a desire to make a contribution to society, or even when it is inspired by religious motives, the functional perspective would indicate that this behaviour fulfils a function of the expression of values for the individual.

People can also decide to become involved in voluntary work in order to adapt to the normative influence that may be exercised by friends, family, etc. In these cases, helping behaviours are controlled by costs and benefits, and volunteerism becomes a means of fitting in with the reference group. Moreover, some people may see in voluntary work a way of widening their circle of social relationships. In all of these cases volunteerism fulfils an adaptive function of a social or utilitarian nature.

Voluntary work may also serve, for some people –or for all volunteers to some extent– as a form of self-protection from their own fears and anxieties.

Finally, Clary and Snyder (1991) point out that for some people and in certain circumstances, voluntary work may provide an opportunity to acquire knowledge. Through the exercise of voluntary activities one can learn from the people with whom one comes into contact, and this can offer new perspectives through having to face certain human problems and situations, thus satisfying intellectual curiosity with regard to increasing knowledge of the reality in which we live. Furthermore, however, volunteerism can be the key to gaining higher levels of specific knowledge, skills and experience (first aid techniques, skills for working with children or the disabled, etc.), which may prove useful in the future. Voluntary work provides the opportunity to use skills that would otherwise be difficult to practise.

The empirical data appear to support the Functional Theory of Motivation, since in the vast majority of studies the volunteers expressed a wide variety of motives.

The different reasons for becoming involved in voluntary work and for remaining in it can be classified in two technical categories: motivations centred on others (other-oriented) and those centred on oneself (self-oriented) (López-Cabanas and Chacón, 1997), or, according to Gidron’s (1978) terminology, altruistic and egoistic.

The other variables included by Omoto and Snyder (1995) in this first stage of the volunteerism process are: "disposition to help", that is, the personal dispositions of volunteers that favour their carrying out helping activities, and the "social support" volunteers have in their everyday life generally, and specifically for carrying out volunteer behaviour.

In the second stage are the variables that promote or restrain the volunteerism process, such as "satisfaction" with the activities carried out and "integration in the organisation". This stage may also include variables not studied by Omoto and Snyder (1995), but which may also be related to length of service, such as burnout. According to this scheme, burnout would be an unexpected result of the experience of voluntary work that would have a negative affect on permanence, either directly or through a negative relationship with satisfaction.

Finally, the third stage includes the variable "length of service" as the main consequence of the volunteerism process.

Applying a methodology of structural equations, Omoto and Snyder (1995) confirmed the existence of a direct and positive relationship between initial motivation to be a volunteer and total length of service.

Moreover, by means of correlational methodology, significant relationships were detected between length of service and three of the Motivation Questionnaire subscales, namely, the three we can refer to as self-oriented (Knowledge, Personality Development and Increase in Self-esteem).

Some authors have suggested that the motivations influencing the decision to become a volunteer are different from those that influence the decision to continue (Gidron, 1984; Oda, 1991; Winniford, Carpenter and Stanley, 1995). Gillespie and Kind (1985) found that volunteers’ motivations change with age, gender and marital status. In this line, it would be interesting to find out whether the volunteers that have been in an organisation longest present different motivations from those who have been in it for least time, and whether age and other sociodemographic variables are also sources of differentiation among the motivations of Spanish volunteers.

Following the Volunteerism Process Model, we began a series of empirical studies on the motivations of Spanish volunteers, as we considered that the recognition of the existence of a wide variety of motivations for carrying out a given behaviour does not preclude an interest in discovering the common patterns of variation that may exist with regard to volunteer motivations.

In the first study the adapted measurement instruments were piloted; in the second, we calculated the reliability and factorial structure of the Volunteer Motivation Questionnaire (Omoto and Snyder, 1995); in the third, we studied the relationship between motivations and variations in the permanence of volunteers in organisations; finally, in the fourth study presented here we investigated the degree of burnout of volunteers and made an initial approach to inferring its relationship to abandonment.

The first study (Chacón, Vecina et. al., 1997), was a pilot study that used 34 volunteers from the AIDS field to test the reliability of and to adapt the instruments measuring the concepts of the Volunteerism Process Model, and especially the AIDS Volunteer Motivation Questionnaire (AVMQ).

The differences between the US sample, made up mostly of homosexuals, and the Spanish one, comprising mainly heterosexuals, obliged us to introduce various modifications in the process of adaptation of the AVMQ, such as a more general translation of the subscale Community Concern. While the original questionnaire referred specifically to the homosexual community, the adapted questionnaire covered HIV-positive people in general.

With regard to the relationship between Motivations and Expectation of continuing in the organisation, which was taken as a predictor of length of service, no significant correlation was found. In contrast, a positive relationship was found between the process variables Integration in the organisation and Satisfaction, and Expectations of continuing in the organisation.

We consider that the fact that no relationship was found between Motivations and Expectations of continuing may be due to the considerable differences between Omoto and Snyder’s (1995) study and our own. A particularly important difference is that in our study there was no follow-up to check actual permanence –there were volunteers who had already been with the organisation for a long time, and it was assumed that their motivation was very high (this result is clear from an inspection of the scores), and their expectations of continuing also. This may also reflect a ceiling effect in the variable Motivations: volunteers’ score being so homogeneously high in this variable, the variability decreases, and with it the correlations.

The next study concentrated entirely on the AVMQ. The objective was to analyse the factorial structure of the AIDS Volunteer Motivation Questionnaire, designed by Omoto and Snyder (1995), and to study its reliability in a sufficiently large sample (Chacón and Vecina, 1999).

A sample of 130 volunteers from eight different organisations completed the AVMQ. With the resulting data, we carried out an exploratory factorial analysis, finding that the factorial structure of the AVMQ, obtained in a Spanish population, provided five factors. The results, in general, confirmed the factorial structure of the original questionnaire and endorsed the Functional Theory of Motivation on which the work of Omoto and Snyder (1995) is based: there are different and varied reasons for carrying out the same behaviour, in this case volunteer behaviour. Table 1 shows the items making up each of the five factors, as well as the reliability of the subscales and the percentage of explained variance. (table 1)

The subscales of Values, Community Concern and Knowledge appear clearly in the Spanish sample as the reflection of three different types of motivation, as proposed by the Functional Theory of Motivation (Clary and Snyder, 1995), and as Omoto and Snyder (1995) confirmed in their sample of American volunteers.

We consider the main differences between the two factorial structures to be due to the different sexual orientation predominant in each sample. These differences especially affect the original Personal Development subscale, since the sense of self-defence against the anxiety produced by AIDS in the group of homosexuals is absent in the heterosexual Spanish sample. The five items of this subscale are dispersed among the rest of the subscales. The two items referring to the need to test and assess oneself (87 and 72) now saturate in the Estheem Enhancement scale, and all of these items make up a factor we have also called Personal Development, but giving it a more unified and general meaning referring to all aspects related to oneself. This new factor would represent the function of the defence of the self or self-protection with regard to one’s own fears and insecurity, both general and specific (in the face of AIDS, for example). The two items referring to the need to form relationships with others (66 and 80) have become isolated, constituting the final factor, called Social Relationships, and the item referring to the need to acquire experience in emotionally difficult matters (68) now saturates in the factor Knowledge, since experience would be one of the possible consequences of learning and knowledge about an issue.

As far as internal consistency is concerned, the high reliability of the Total Questionnaire total and its subscales permits us to affirm that it is of a sufficient level in the AVMQ adapted to the Spanish population to measure the motivations: Personal Development, Concern for the Community, Knowledge, Values and Social Relationships of volunteers working in the AIDS field, though the last-named factor would need to be confirmed in future research that included a greater number of items.

In a more recent work (Vecina and Chacón, 1999) we have attempted to make advances in the study of the motivations of Spanish volunteers, trying to ascertain the influence of sociodemographic variables (age, sex, educational level, work situation, type of voluntary organisation, length of service in the organisation and acquaintance with any AIDS sufferer prior to decision to become a volunteer) on these motivations, measured with the AVMQ. A second objective of this work was to discover whether there existed a specific motivational profile that could be associated with length of service in a voluntary organisation, since in this way we could identify the motivational profile of those volunteers that spent longest in an organisation.

This study was carried out with a sample of 112 volunteers from nine different organisations. With the data obtained we carried out, on the one hand, variance analysis to check for possible motivational differences between the different groups formed according to sociodemographic variables. We also calculated the power of the differences found. Moreover, we carried out a cluster analysis with the volunteers’ scores on the AVMQ scales and with length of service in the organisation (Group 1 was made up of volunteers that had been in the organisation for less than 3 months, those of Group 2 had been working for between 4 and 7 months, those of Group 3 between 8 and 18 months, and those of Group 4 more than 19 months).

From the results obtained it can be concluded that, in general, volunteers consider themselves motivated especially by Values, that is, by a humanitarian concern to help others. Moreover, for those volunteers that have spent longest in an organisation, other-oriented motivations are significantly more important than for those with the shortest periods of service, which indicates that these motivations are, in principle, those most closely related to permanence in volunteerism.

Nevertheless, a closer analysis of the cluster analysis data reveals that volunteers who have spent longest in an organisation present a motivational profile characterised by giving greater importance to self-oriented motivations than volunteers with shorter periods of service, and by giving to both other-oriented and self-oriented motivations much more importance than the rest of volunteers that have spent less time in the organisation. Thus, it can be supposed that self-oriented motivations also play an important role in the permanence of volunteers (see Graph 1)

In any case, it is not clear as to whether the level of importance given by the longest-service group to Values motivation and all motivations in general –especially self-oriented ones– is equally high from the moment of deciding to become a volunteer or whether it increases with the passage of time or due to organisational and personal characteristics. In order to respond to this question it is necessary to carry out longitudinal studies that take into account the real evolution of motivations over time.

In the study described, the measures of the variables were taken at a single point in time, so that we can only venture a tentative hypothesis about the evolution of motivations over time. It can be supposed that people who decide to become volunteers would be motivated by both other-oriented and self-oriented motivations. At this initial moment, we can assume that the level of expectations is very high and the level of knowledge of the reality of the work very low, so that, after the first few months of contact with reality, volunteers present a considerable reduction in their motivation level. This would lead to a high percentage of drop-outs, due fundamentally to the non-satisfaction of initial motivations. Subsequent to these first few months, and following the hypothesis of Piliavin and Callero’s (1991) model, the volunteers that continue would begin to incorporate into their personal identity the role of volunteer, resulting in an increase in other-oriented motivations and a decrease in self-oriented ones. This profile, which we have called self-sacrificing, cannot be maintained in the long term, so that at a fourth stage, after eighteen months, volunteers would incorporate self-oriented motivations into other-oriented ones.

Nevertheless, as we commented earlier, the evolution of volunteers’ motivations described is only a working hypothesis, which should be confirmed in future studies.

Finally, we can conclude that, according to the data provided by this study, volunteers who were highly concerned about others, but without neglecting themselves, were those that had spent longest in an organisation.

The basic objective of this study was to check whether volunteers carrying out social work activities with AIDS or cancer victims were affected by burnout syndrome or not. We considered this type of volunteer to be exposed to situations of high emotional demand –similar to that of health professionals, in whom the syndrome has typically been studied.

In order to fulfil this objective we applied an adaptation of the "Maslach Burnout Questionnaire" (Maslach and Jackson, 1982), composed of 22 items that measure the frequency with which certain feelings are experienced (1 means the feeling is never experienced, while 5 means it is experienced every day). Items are grouped in three subscales: Emotional exhaustion (9 items), Depersonalization (5 items) and Personal accomplismment (8 items). The reliability index ( of the total scale is 0.64, and those of the subscales are 0.8, 0.5 and 0.7, respectively.

The questionnaires were applied in 8 different organisations to volunteers that had been working for more than 6 months. The continuation of this study would involve carrying out follow-ups in which repeated measures of burnout would be taken for those who abandon volunteerism.

A total of 69 volunteers completed the questionnaire. Mean length of service was 35 months (two years and eleven months), there was a predominance of women (54%) over men, and mean age was 34 years, ranging from 19 to 71. Also, 54% of volunteers knew no-one with AIDS or cancer before deciding to become a volunteer, and 60% knew no-one involved in voluntary work before making their decision. 19% worked in more than one volunteer organisation.

The principal result, obtained by means of data analysis, indicates that the volunteers of the sample analysed presented the typical symptoms of burnout "a few times a year". An analysis of each of the subscales showed that volunteers presented signs of emotional fatigue and low personal gratification with a frequency of "a few times a year", whilst in almost no case were symptoms of "depersonalization" presented.

In order to discover whether there were significant differences in the frequency with which burnout symptoms were experienced, in accordance with the sociodemographic variables measured, one-factor variance analyses and Student’s t test were carried out. Among the most notable results are the following:

The direction of these differences indicates that the volunteers working in more than one organisation more frequently experienced burnout symptoms than those that worked in just one, and that this was especially the case for personal Accomplismment.

It can also be seen that volunteers working in religious organisations presented lower levels of burnout and higher personal Accomplismment than those in lay organisations.

The fact that burnout symptoms are experienced with low frequency may lead us to think that volunteers are not affected by this problem. Nevertheless, certain aspects should be clarified that call into question this initial conclusion.

Firstly, there is an important difference between being a volunteer and working in an organisation as a professional. While the former is free to leave as soon as s/he feels uncomfortable or dissatisfied, professionals are obliged to remain in their job for much longer periods, in which burnout symptoms intensify. This leads us to suppose that it would be difficult to find working volunteers suffering from intense burnout since, obviously, they would abandon the organisation before reaching such extremes.

Moreover, it should be borne in mind that all the volunteers in our sample were working at the time the measures were taken, and that the follow-up studies that would allow us to compare those who stayed and those who gave up have yet to be carried out. Thus, the probability that the volunteers in our sample suffered from burnout is very low.

It should be considered that the questionnaire used for measuring burnout, which is an adaptation of that originally designed for professionals, is not sufficiently sensitive to take account of significant burnout indicators in volunteers, so that it is necessary to modify it.

The longitudinal study we are carrying out, and whose first phase is the research presented here, will allow us to confirm whether volunteers drop out due to burnout, so that the maximum burnout level is reached before dropout, or immediately after it. We hypothesise that burnout will be maximum at this point, though we predict that it will rarely reach the levels found in studies using samples of professionals.

The results of these initial studies permit us to conclude, first, that the instrument for measuring the motivations of volunteers in the field of AIDS is a sound one: apart from being reliable, it reproduces the five-factor structure of the original questionnaire. This allows measurement of the motivations: "Values", "Community Concern", "Knowledge", "Personal Development" and "Social Relationships".

Secondly, and regardless of their length of service, volunteers would appear to identify as the most important motivation for them that of "Values" (other-oriented motivation). Moreover, if we relate length of service with motivations, it can be seen that other-oriented motivations are significantly more important for those volunteers with longest service in an organisation. However, analysis of the motivational profiles associated with length of service allows us to confirm that volunteers that have spent longest in an organisation (more than 18 months) lend more importance to self-oriented motivations (knowledge, personal development and social relationships) than those with least time served, so that we can conclude that both other-oriented and self-oriented motivations, at their highest levels, are related to permanence in volunteers.

Thirdly, the sample of volunteers studied would appear to present low levels of burnout. Nevertheless, this fact should not lead to the conclusion that volunteers are not affected by burnout since, in the first place, the volunteers of the sample were in work at the time the measures were taken, so that it is improbable that they were suffering from burnout, and in second place, the follow-up studies that would reveal which subjects remained and which ones dropped out, and the burnout level prior to dropout, have not yet been carried out. More extensive longitudinal studies are necessary, both in order to monitor the evolution of motivations over time and to confirm whether or not burnout influences volunteer dropout.