The union density of private services workers is clearly lower than that of other categories of organizations. The average for Canada as a whole is 40%, while that for personal services workers is less than 11%. The situation is, to say the least, disturbing for unions, which have difficulty organizing these continually growing groups of workers. Community organizations do not appear to be exceptions to the major extrinsic rules of labour relations in the private services sector, with wages lower than the national average, an absence of job security, precariousness and low union density. Since the early 1980s, these organizations have multiplied, for example, in the areas of job rcadiness, employment entry, health and social services, and so on.

Even though these organizations have a social mission, provide services to members, function democratically, and are non-profit, they are nevertheless characterized by management-worker relations in the sociological and legal sense of the term, entalling an employment relationship that is the basis of unionization. There are appoximately 60,000 paid jobs in the community and association movement in Quebec, including organizations involved in the defense of social rights and popular education, health and social services, child care, housing and recreation. In the absence of research into the unionization of community workers, the theoretical basis of this paper draws on the analytic elements developed by Wheeler for explaining the propensity to unionize and, from there, to build our own conceptual framework.

Dissatisfaction is taken as the point of departure for the process leading to unionization. This dissatisfaction may result from working conditions considered to be inappropriate, a particular management style, or specific incidents that give rise to frustration. Feeling dissatisfied, frustrated, or even aggressive, the worker looks for a way to improve his or her situation. Several options are avallable. Economic theory predicts that the worker may decide to do nothing and put up with the situation, to leave the organization, or to put forward his or her point of view. If they decide to act, workers may do so individually, presenting their problem to management in the hope of solving it, or they may see the union as a tool that can be used to express dissatisfaction. The instrumental evaluation by the worker in large part depends on subjective elements that go well beyond the simple framework of objective utility. Thus, in order to fully appreciate workers' evaluations, their attitudes towards unions must be considered, that is, their beliefs and values regarding the movement itself (the affective dimension) as well as their individual perception of what they may gain or lose by unionizing (the instrumental dimension). Finally, a union must be willing to extend membership to the workers and offer them services to assist them in the process of unionizing. However, some unions have very restrictive policies in this regard.

The elements of this explanatory model were studied in field research conducted in three consecutive stages between May 1997 and May 1998. While the first two stages examined the propensity to unionize, the third examined the supply of union services. The study's first stage involved field work in six community services organizations in the Outaouais region. In each organization, the researchers interviewed the person in charge of administration and, depending on the size of the organization, between one and three workers. The information gathered in the first stage was then explored in more depth during the second stage of the field work through three discussion groups with community workers. Finally, in the third stage, the supply of union services was examined through six individual interviews with representatives of the two major labour federations in Quebec. Examination of the comments made by employees in interviews and group discussions regarding their needs at work reveals that employees are generally satisfied with the team-based management of work and the autonomy enjoyed in carrying out their tasks. On the whole, our analysis brought out four major sources of dissatisfaction: (1) relations with the board of directors, in which the workers have little involvement in decision-making; (2) extrinsic working conditions, particularly salaries and benefits; (3) the funding of organizations; and (4) the work itself. Internai mechanisms for raising and solving problems that give rise to dissatisfaction are rare, although in some organizations employees are consulted directly regarding wage determination. On this latter point, it also seems important to emphasize that many sources of dissatisfaction lie beyond the boundaries of the community organization. In fact, low wages, poor working conditions, understaffing and job instability are elements of dissatisfaction that are in very large part related to insecurity and lack of funding sources, which are external to the organization. This has a bearing on the evaluation of attitudes towards unions and the instrumental dimension.

The analysis of attitudes towards unions revealed that the beliefs and values of community workers are compatible with unionism. As a whole, both the managers, as well as the unionized and non-unionized workers interviewed, share the general values of the union movement and believe in its social relevance. However, they do not see unions as appropriate in the community sector, even though they believe that it is necessary to have an institution that plays a role similar to that played by a union. At first glance, it seems that the image held by workers of the effects of a union presence stand in the way of their willingness to unionize. In this study, the supply of union services, the third and final component of the framework used to analyze the unionization process, has two dimensions: the existence of union supply; and the particular form of union services offered. The interview data revealed that unions are very much aware of the specificity of community organizations as compared to private and public sector organizations. These characteristics serve to diminish the enthusiasm of unions, and cause others to reject outright the possibilities of organizing in this sector. None of the union officiais interviewed work for a union that is taking the lead in organizing this group of workers. The majority are content to wait for requests and to deal with such requests according to the criteria already mentioned. A minority simply refuse to organize in the community sector for reasons related to the number of workers concerned, the costs involved or the organizationai mission. Thus the union supply is relatively limited when it comes to unionizing community workers. Such workers must normally initiate the demand and approach the right union organizations, and the latter must perceive the group and the organization that employs them as viable. The specificity of community organizations can also affect the form of supply in that union representatives consider it essential to adapt union demands to the specific needs of the workers and community organizations. Certain demands that are proposed fit in well with the needs of these workers. However, none of the union officiais interviewed mentioned the issues of inflexibility that were referred to by so many of the workers. For their part, judging from the comments made during interviews and discussion groups, the workers do not appear to be aware that unions are willing to adapt their demand strategy to the particularities of the community sector. Rather, their opinions are influenced by their observation of labour relations in the much larger health and social services sector. Thus, the form of union supply and how it is perceived by community workers also have a negative influence on their willingness to unionize.

The results of the empirical measurement of the constructs that make up the explanatory model of the unionization process lead us to conclude that, as a whole and according to the situation as it was at the time of this study, the probability of unionizing community workers is low. Although an examination of the needs of this group of workers suggests that they would have a high propensity to unionize, this assumption is quickly undermined by the workers' perceptions of the utility of unionization. In fact, although these workers have a large number of unsatisfied needs at work, most of them do not believe that unionizing will help them to improve their situation. Several aspects of supply of union services also help reduce the potential of unionization. In addition to being limited due to various inherent constraints, it is only partially congruent with the needs of community workers. There is also a strong perception among these same workers that union services are not well suited to their situation.

The changes in the provision of health and social services in Quebec created by the trend towards ambulatory care could, however, have a positive effect on the propensity to unionize in the community sector. Indeed, during the past few years, many community organizations have experienced an appreciable increase in their funding base and in the number of workers they employ. Often this comes with control mechanisms regarding the mission of the organization and how its services are delivered. AH this is leading to significant change in these organizations. This may result in a degree of alienation among workers regarding the objectives of the organization and the services that it provides, thus partly countering the normative control referred to above. The increase in the number of employees also involves a degree of formalization of labour relations. However, as we have already seen, fear of the formalism brought about by unionization is one of the factors that contributes to resistance to unionization. Once established in the organization by management, this obstacle obviously diminishes accordingly.

In light of the analytical elements presented in this article, as well as the additional discussions with union representatives and workers in the sector, we believe that it is possible to develop a union strategy to increase the potential of unionizing community workers. At the outset, this strategy should target certain community organizations and be based on realistic demands adapted to their needs as well as include organization and demand structures that are suited to the reality of community organizations. Due to their more stable funding base, labour relations that are often more impersonal and greater union viabiliry, organizations with a larger number of employees should be targeted first. The organizing campaign initiated by the union itself could be carried out on the basis of a set of issues identified through contacts with employees who are representative of the environment. Once the issues are identified, credible demands can then be developed, demands that then become a springboard to unionization. The relatively recent breakthroughs by the CSN in the day-care sector and the structures of representation they have put in place also provide interesting possibilities for unionization in the community sector.