CONVENTION DELEGATES IN SPAIN: STATE OF THE ART Astrid Barrio Universitat de València Astrid.barrio@uv.es Juan Rodríguez Universitat de València Jrteruel@uv.es Oscar Barberà Universitat de València O.barbera@uv.es Montserrat Baras Universitat Autònoma de Barcelona Montserrat.baras@uab.es Paper presented to the Workshop “The current state of party member research”, Copenhagen, February 3rd - 4th, 2011 Abstract: The aim of this paper is to present an assessment of the state of the art in the study of party members in Spain. In particular, the paper is focused on the study of convention delegates, by far the most studied topic of this literature in Spain. To that purpose it first describes how party convention works and how they have evolved over time. Then a brief evolution of the different team projects and questionnaires is presented. This finally allows us to critically review the major theoretical, methodological and practical problems raised in this research. Keywords: Party members, convention delegates, political parties, Spain Introduction1 The work on party members in Spain began in the 1970s as part of historical research on political parties. Over time, the study of party members has gained autonomy and relevancy, nevertheless important gaps remain. Generically, there are three lines of research in the study of members of political parties in Spain. One line of research attempted to quantify party membership and it established series of reliable data on trends in the number of members. Most of these studies are based on figures provided by the parties themselves, while others, more recently, opted to estimate the overall membership in Spain using data such as that from the European Social Survey. The findings from these studies suggest that Spain seems to be a deviant case in the general trend in many Western countries of falling membership. Although party membership figures in Spain have traditionally been low, they have not stopped rising since the transition to democracy, a situation probably closely linked to the late formation of parties in this country. A second set of studies has focused on examining the involvement of members within the parties. For this, party statutes and regulations have been mainly analyzed. These studies have attempted to make comparisons between parties and the role of members over time. This has highlighted how the organizational changes made by the parties (centralization, presidentialization, etc.) have affected the functions and activities that members develop inside organizations. The third line of research has been based mainly on surveys to the members of the parties. In this section, the study of members has been relegated to second place, since virtually no surveys have been conducted in Spain to party members. Some researchers have tried to overcome this problem through the use of population surveys. However, for theoretical, methodological as well as practical reasons the vast majority of researchers have focused attention on convention delegates. In some cases, these studies have extended to include the observation of other internal organs of the party. In general, studies on convention delegates have been dominated by case studies and this has hindered, until now, the accumulation of knowledge and above all, the development of a general theory about the changes that these individuals have been experiencing over the years and the impact they have had on the internal life of parties. The aim of this paper is to present an overview of the study of party convention delegates with particular attention to the research work carried out by the authors of this paper. The first section describes the object of study and the attention that it has received from the Spanish political science. The second section shows the current status of research carried out by our team. Finally, a critical review will be conducted 1 This work is part of project SEJ-2006-15076-C03-02 (2006-2009) and MICINN CSO2009-14381C03-02 (2009-12) funded by the Spanish government and project SGR 2009-1290 (2009 -13) by the Catalan regional government. In addition, its researchers are part of the research group on Elites and Political Parties, funded by the Institut de Ciències Polítques i Socials (ICPS) in Barcelona. of the major theoretical, methodological and practical problems raised in this research. 1. Studies of members of party conventions in democratic Spain. 1.1 The evolution of party conventions in Spain It is not a coincidence that the preferred field of study concerning members of political parties in Spain is the party conventions. For theoretical, methodological and practical reasons these have become the best objects to analyze the human basis of political parties. From a theoretical point of view and following the path of Cayrol (1974) it is often argued that the conventions are instances particularly relevant for those who are interested in the sociology of activism. The party conventions in Spain are the highlight of the political party life. This is where the party line is set, operating mechanisms are reviewed and a direction is decided, consequently these conclaves usually gather the most active party members identified by political science as party activists. Moreover, from a methodological perspective and in practical terms it is relatively easy to conduct a survey to a group of people concentrated in a physical space for a short period of time. This makes them a contingent of manageable individuals, so that the completion of the survey by means of collective writing does not pose a great challenge for researchers. As discussed later in this paper, these reasons have been the subject of serious discussion over the years but this has not prevented members of conventions to continue to be analyzed mainly through this mechanism. Overall, conventions in Spain tend to work through the mechanism of delegates. This means that not all party members attend them but only a portion of them. They act as representatives of the various constituencies that elected them and their number varies depending on the number of party members. Thus, numerous constituencies have more influence in the convention. Party conventions are becoming more numerous, a fact largely attributable to the will of the parties to publicly display their strength but as we point out later the disproportion between party members and delegates tends to grow in big parties. Although constituencies and the methods by which delegates are elected have changed substantially depending on the party and time, these procedures appear to have given greater opportunities for member participation (Méndez Morales, Ramiro, 2004: 193). However, it should be noted that in Spain there are still parties that hold or have held conventions in the way of assemblies, i.e. they are open to the entire membership (provided membership fee has been paid). However, the trend is for this system to be eliminated and currently only the Esquerra Republicana de Catalunya (ERC) maintains it. The fact that delegates are elected and not appointed renders these members of peculiar relevance. In general, to be chosen candidates must lead an active party life, which allows them to be known and to acquire some merit in the internal areas of the organization. Very rarely someone who only pays the party fee will be appointed, however someone who actively participates in party life, whether this person holds an internal position or not, shall have a good chance of being chosen. There are some elements that can shape the selection process and that differentiate parties over time. Firstly, the selection processes have evolved within the parties, which must be taken into account when making comparisons between delegates over time. Political parties have tended to gradually regulate and centralize the process of appointing delegates (Méndez, Morales, Ramiro, 2004: 189 et seq. 2), which may have affected the profile of applicants likely to be chosen as delegates. Secondly, it should be noted that delegates are not the only participants in conventions. It is common to all parties that in addition to the elected delegates there is a variable percentage of co-opted delegates or ex officio members attending by virtue of their position (they are members of the highest executive bodies and party representatives, public officials, etc.). However, most parties have tended to set statutory limits on the presence of non-elected delegates. These limits have changed over time although several parties like the PP and CDC have, for some time now, established quotas close to 20% of delegates on the basis of their positions. Ultimately, we should add that many of the Spanish parties have tended to make possible representation of delegates by other ways than constituency selection. Undoubtedly, the bulk of these representatives come from its auxiliary organizations, especially the youth, although some parties also include the election of delegates through sectorial bodies (teachers, elderly, and so no). On the other hand, it should be noted that there are two elements that have tended to restrict the opportunities of membership participation in party conventions. The first element is the progressive reduction of the frequency of conventions. Over the years, holding regular conventions has tended to spread and as a general rule most parties currently hold conventions every four years. Usually they coincide with the end of the electoral cycle although some of them are held before election cycles in order to ensure the stability of party leaders in the face of potential bad results. Thus, the holding of annual or biannual conventions that characterised the lives of Spanish political parties during the political transition and the early eighties changed in the nineties when conventions are held every three or, more commonly, every four years. However, all parties statutes anticipate special conclaves in exceptional circumstances -internal crisis, electoral defeats, resignations-. The second element has to do with the growing tendency to increase the number of delegates representing each party member (Figure 1). This trend is especially visible in both major national parties, the PSOE and the PP. This is due to the fact that the steady increase in registered members since the transition has not been accompanied by a substantial increase in the number of delegates at conventions. In the case of other parties such as the PCE / IU or CDC the trend has oscillated; in the case of PCE / IU it is due to a crisis in the number of registered membership in the eighties and in the case of the CDC due to an increase in the number of delegates as membership grew. 2 Papers by Méndez, Ramiro Morales (2004), Ramiro (2005), Rodriguez, Barbera, Barrio and Baras (2011) provide enough information on the procedures for electing delegates in each of the three national parties and their changes over time. Figure 1. Ratio of party members per convention delegate in some Spanish political parties from 1979 to 2009 Source:For the convention delegates: author’s own compilation from data submitted by the parties and the press. For the party members Méndez, Morales, Ramiro (2004),and Linz, Montero, Ruiz Jiménez (2005) The data on the horizontal axis refers to the conventions held by each party since 1979. They do not correspond with the number of conventions of each organization. 1.2 The evolution of convention delegates’ studies. Studies on convention delegates in Spain started almost in parallel to the process of transition to democracy. These pioneering studies were strongly influenced by French and Italian political sociology, especially through the works of Duverger, Cayrol, Ysmal or Alberoni. In 1979, Tezanos surveyed delegates from 28th convention of the PSOE (Tezanos, 1980), while Joan Botella, Jordi Capo and Joan Marcet conducted a comparative survey of delegates from various Catalan parties (Botella et alt., 1979). It was the first approach to party activists based on the line proposed by Cayrol (1974). Their results enabled comparisons to be drawn between the convention delegates and other sectors of the Spanish political elite of the transition, such as parliamentarians or members of the party leadership (Pitarch et alt., 1980; Tezanos, 1983). Similarly, Tezanos performed pioneering work in 1980 in Spain by interviewing a representative sample of members of the PSOE (Tezanos, 1981). A few years later, under the leadership of Julian Santamaría, some of these authors formed a Spanish team which joined the European Political Party Middle Level Elite (EPPMLE) project, led by Cayrol and Reif (Reif, Cayrol and Niedermayer, 1980; Cayrol and Reif, 1983). This team conducted surveys to three Spanish parties (PSOE, AP, CDC) between 1980 and 1982. The first results of this research were published in the mid-eighties (Capo and Botella, 1985) and were subsequently used in monographs on political parties (Marcet, 1984, Colomé, 1988, Lopez Nieto, 1996 and 1997). Throughout the eighties, the study of convention delegates was only continued in Catalonia, where Gabriel Colomé undertook a considerable number of surveys in the Catalan socialist and communist conventions (Colomé, 1988, 1992a, 1992b, 1996, 1998). These studies noted the evolution of convention delegates of these parties. In contrast, the lack of a comparative perspective prevented the development of a more systematic analysis of delegates as a whole. This comparative approach did not take place until the mid-nineties, when the interest in comparing the sociology of the Catalan party delegates resurged. In 1996 for the first time a team conducted a survey of delegates from various political parties with parliamentary representation in Catalonia (Sánchez, 1999). This work was based on a questionnaire similar to that used in previous studies and it maintained the approach taken by the EPPMLE, who considered the universe of individuals surveyed as intermediate party elites and not as activists. Since 2000, surveys of convention delegates in Catalonia received new impetus from the research group led by Montserrat Baras, under whose direction several waves of surveys were carried out. Baras’ team developed a new and much more extensive questionnaire and in 2000 successfully forwarded the survey to all Catalan parties with parliamentary representation (Baras, 2004) and to their youth organizations (Barberà, Barrio and Rodriguez, 2002a). For the first time, data was collected from all Catalan parties represented in parliament and it was possible to complete the sociological and ideological map of the parties’ delegates and their youth organizations. In 2004, a similar questionnaire with slight modifications was used in the conventions of the same parties with the goal of collecting a series of chronological data. This time the success of data collection was unequalled. In 2006, Montserrat Baras’ team redefined the project. Firstly, the territorial scope of analysis was extended and all Spanish political parties became the object of study. Secondly, within the framework of a coordinated research project funded by the Spanish government, they adapted some questions in order to make it comparable with the survey of the Observatorio de Politica Autonomica (Autonomic Policy Centre), which, under the direction of Francisco Llera, conducts regular surveys of the population in various autonomic regions of Spain. Finally, new questions were added in line with the questionnaires used in the UK to interview party members (Seyd and Whiteley, 1992, 2004, Whiteley, Seyd and Richardson, 1994). Table 1. Surveys of convention delegates conducted in Spain (1979-2010) Party Year 1979 1980-82 1984 1987-88 1989-90 1993 1996-7 2000 2004 2006-10 AP PP √ √ √ PP Cat. √ √ PSOE √ √ √ - PSOE And. √ PSC PCE √ √ √ √ √ √ √ √ √ √ - PSUC ICV √ √ √ √ √ √ √ CDC ERC UDC C’s CpC UPN CHA CC PRC √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Acronyms: AP/PP: Alianza Popular/Partido Popular; PP Cat: PP en Cataluña; PSOE: Partido Socialista Obrero Español; PSOE And: PSOE en Andalucía; PCE: Partido Comunista de España; PSC: Partit dels Socialistes de Catalunya; CDC: Convergència Democràtica de Catalunya; PSUC: Partit Socialista Unificat de Catalunya/ Iniciativa per Catalunya; ERC: Esquerra Republicana de Catalunya; UDC: Unió Democràtica de Catalunya; C’s: Ciutadans/Ciudadanos; CpC: Ciutadans pel Canvi; UPN: Unión del Pueblo Navarro; CHA: Chunta Aragonesista; CC: Coalición Canaria; PRC: Partido Regionalista Cántabro. Once the questionnaire was updated, the group's priority was to expand the number of parties surveyed with particular emphasis on regionalist parties of parliamentary and governmental significance (Table 1). Also regional conventions of the major national parties were analyzed. Overall, between 2000 and 2010, Baras’ team has performed more than thirty surveys of party conventions and obtained more than 5,000 responses on a universe of study of around 20,000 individuals. This has enabled advances in the diachronic analysis of the delegates as well as in crosssectional comparisons of their sociology, attitudes and behaviors (Argelaguet, 2008, Baras et al., 2011). In parallel to the sociological approach other studies have also taken convention delegates as a research subject in recent years. However, these studies have not focused on the sociological analysis of delegates but on their role in the process of selecting leaders and on the organizational effects generated within the party (Ramiro and Morales, 2004; Mendez, Morales, Ramiro, 2004; Ramiro, 2005, Rodríguez et al., 2011). 2. The content of the surveys: questions and indicators The fundamental question that has guided this research was to get to know party delegates from a demographical point of view as well as from an ideological point of view. To answer this question, surveys have included various types of questions that have evolved over time. They can be grouped in four main categories. The first category includes socio-demographic questions (sex, age, place of birth, residence, education level, employment status, occupation, religion, etc.). According to the perspective adopted for these studies these questions made it possible to attune the intermediate elite’s condition attributed to party delegates. A second group includes questions on associations (membership in other political and social organizations) and political career such as previous affiliation to other parties, length of membership, participation in previous conventions or possession of internal party positions which involve institutional representation and appointment in the administration. This has made it possible to outline a specific profile of party delegates of extensive experience in politics with a high degree of association and party internal responsibilities. This profile has raised questions about who they are and what these individuals represent. Thirdly, the various surveys have included questions concerning the ideological identity in relation to the individual’s positioning in the left-right axis and their position regarding national identity. Subsequently, questions were added about the position of the party and other formations. From this, it has been possible to draw a map of ideological distribution of delegates and to test some hypotheses concerning the internal ideological differences between the various party members (Baras, Barbera and Barrio, 2008), in the line of the work carried out by John D. May (1973), and similar to other studies done in Spain (López Nieto 1996; 1997; Mendez and Santamaría, 2001). Questions relating to the valuation of the parties and their leaders were included along with the questions of ideological position. This made it possible to raise explorative questions about the parties’ political alliances and intraparty conflicts (Baras et al., 2010a.) However, indicators of ideological and political identity (axis position, evaluation of leaders and parties) have highlighted difficulties and restrictions in the interpretation and use of quantitative comparative analysis. Since 2008 the questionnaire has undergone a major expansion in order to promote comparison and deepen the theoretical scope. New questions were added about the ideological preferences and delegates’ involvement in the party. Along with existing questions, new questions were added concerning political socialization and political information acquisition, perception of own role and that of the party leaders, their linkage with other social organizations, and also questions concerning opinions regarding the internal functioning of the party. Similarly, questions were added on various topics on the political agenda (territorial organization of the state, individual values, political attitudes), which identify the delegates ideological preferences. Basically, the new survey questions are designed to observe the behavior and participation patterns of party delegates. Regardless of how the delegates are from a demographic perspective and their ideology, we attempt to ascertain the nature of their behavior and the reasons behind it. The new questionnaire certainly allows new perspectives in the analysis of party delegates. 3 Theoretical, methodological and practical problems in the study of convention delegates in Spain. 3.1. Theoretical problems: the changing conceptualization of the delegates Defining the nature of the object of study is one of the major challenges facing the study of party delegates since it involves the always difficult task of defining and identifying the various stakeholders that make up political parties (Heidar, 2006). What are we looking at when we observe these individuals? The scholars’ viewpoint has differed depending on the criteria followed; on whether it is the position held within the party or the behavior that defines the degree of involvement. In the seventies, part of the French political sociology tended to argue that the study of delegates was a good way to approach the study of party activists (Cayrol, 1974; Cayrol and Ysmal, 1982). However, in the eighties, the authors of EPPMLE defined them as intermediate party elites following positional criteria (Reif, Cayrol and Niedermayer, 1980; Niedermayer, 1986)3.Despite the success of this denomination, one of the main criticisms made of the project was precisely the positional approach used to conceptualize delegates (Pierre, 1986a and 1986b). Following in part Duverger’s theory of concentric circles and therefore taking the structure of mass parties as a reference, studies concerning delegates in Spain have tended to adopt Cayrol’s perspective about who party delegates are and what they stand for. Nevertheless, the intermediate elite status was not ignored. As noted earlier, this tradition distinguishes the different layers of the parties’ human basis according to their degree of participation or involvement. Consequently, delegates 3 Niedermayer set up a typology that differentiated between "ordinary party members, local party activists, local party elites, middle-level party elites and party leadership" (Niedermayer, 1986: 253) . are considered the active members of political parties and the Duvergerian concept of party activism was used (Baras, 2004). Graph 2. Presence of delegates in party internal positions (%). Source: Annex I However, this consideration has for some time not been satisfactory, not only because of the changes experienced by political parties in recent years but also due to the findings derived from the results of research. Convention delegates are no longer -and possibly have not been since the restoration of democracy in Spain-, simple active members who supposedly devoted their time to the party on a voluntary basis, and who are driven by collective identity or solidarity. In contrast, data shows that many of them have held internal positions in the party (Figure 2), though this data is fragmented and needs to be analyzed carefully. Their high and growing level of involvement in the organizational hierarchy has long questioned the conventional assumption that these individuals are mere party activists or even according to the old definition of Cayrol and Ysmal, which, as noted, was taken as the basis for the research of delegates in Spain. Graph 3. Presence of delegates in public office and administration (%). Source: Annex I The progressive involvement of delegates in the party's organizational structure has evolved in parallel with a growing trend towards political professionalization. The pattern of progressive participation of delegates in public office or positions within the administration seems quite widespread in all parties but as indicated later this data should be analyzed with caution because of its fragmentation (Figure 3). The data seems to coincide and it is confirmed by the results of other research on parties in Spain (Oñate, 2008). The gradual professionalization of delegates suggests changes in the structure of incentives by which they participate in politics. Although they may also be motivated by incentives such as collective identity or solidarity, the growing professionalization highlights the importance of selective incentives at the time of collective action. Figure 4. Presence of delegates with union membership Source: Annex I It is also possible to appreciate these changes when analyzing the membership of delegates of leftist parties (PSOE, PSUC-ICV and PSC) to the various unions (Figure 4). The progressive decline in membership from the eighties must surely be understood in the context of the relationship changes experienced during this period between the leftist parties and unions. However, it is also possible that they may be related to changes in the structure of incentives and especially in the change of priorities in the incentives that guide collective action. 3.2. Methodological problems: representativeness, reliability and comparison Research on delegates conducted over the years raises some methodological problems related to the representativeness of the data collected, the reliability of the method itself and the limitations of comparing between individuals and parties. One of the most significant methodological problems of studying delegates through surveys has to do with representation. This problem is particularly relevant when the number of questionnaires collected is low. The answer rate to our questionnaires has tended to be a percentage equal to or above 20% with some exceptions (Figure 5). One way to control the potential bias that may occur in the population surveyed is to contrast the data collected with the official information on delegates that the parties make available to the media. This can be done by means of certain demographic variables (gender, age, territorial origin). However, it should be noted that this exercise of control of the bias is not always possible because sometimes parties avoid sharing this type of information with the media or researchers4. 4 The example of UDC which in 2008 managed to omit the number of delegates to the media illustrates this point. Graph 5. Percentage of responses over the total of delegates. Source: Annex I Secondly and linked to the issue of representativeness, there is a problem of reliability resulting from the method used for distributing the survey to delegates5. In the Spanish case, the collection of information is done through questionnaires to be completed by the delegates themselves at the convention. In the case of the delegates, this data collection technique does not suffer from the typical problems of bias resulting from barriers related to education because the vast majority of them are highly educated (Niedermeyer, 1986: 255). In contrast, there are other problems of bias that are more difficult to solve. The most difficult issue is undoubtedly the difficulty in obtaining information from the upper echelons of the party elite. This is due in part to their fear of being identified but, above all, due to their lack of time to fill in the questionnaire. Finally, the main methodological obstacle faced is the degree of comparison that can be performed based on the data from our research and that conducted previously. Restrictions on comparative analysis of delegates may arise for three main reasons: the changes that occur in the object (regulatory changes in the functioning of the convention or the selection of delegates); changes to the tool used for data collection (changes in the questionnaire) or because of the difficulty in handling additional data or in accessing databases from other investigations. From the outset and as noted in section 1.1, we point out the limitations in the comparisons that result from regulatory changes that have occurred in the organisation of conventions. This is a problem with relevant theoretical implications because changes of the object reveal profound changes in the functioning of political parties and the internal distribution of power. Beyond the changes in the evolution of the object, there are some intrinsic problems related to the method used. On the one hand, we must mention the differences in the 5 The questionnaire is included in the conference documentation and delegates must fill out and submit it to the researchers in a given place. They are responsible for reminding convention delegates to complete and collect it. content of the questionnaires, which sometimes may have been designed from different perspectives and for different purposes and, therefore, with different questions over time. Early research on convention delegates in Spain was mainly intended to define the delegates’ sociological portrait and to explore their basic ideological orientation. As already noted, it was not until very recently that questionnaires began to incorporate questions on behaviour and political views. On the other hand, problems may occur due to the discrepancies in the measurement used for the indicators, even for questions that have been maintained over time. One of the most common discrepancies lies in the use of different scales for the identification on the left-right scale (Niedermayer, 1986: 257). In the Spanish case, while some research used a scale of 1 to 10 (Botella and Capo, 1985, Mendez and Santamaria, 2001, Lopez Nieto, 1996 and 1997), others have opted for scales from 1 to 7 (Sanchez, 1999 , Baras, 2004). The third problem in comparative analysis is related to the technical handling ofdata. One of the difficulties of diachronic comparison arises when research has been conducted by different teams at different times. In some cases access to databases is difficult. In other cases, the problem is more serious because the databases and questionnaires were lost; this is the case for surveys conducted in the late seventies and much of the eighties. In these cases, we only have published data with limited opportunity for handling and comparison with other databases. This leads to important limitations to the diachronic analysis of delegates and parties, as highlighted in the next section. However, the difficulty in comparing data results also from the handling of data from different parties when the comparison is done on an individual level at a given point in time and collected by the same team of pollsters. In fact, one of the main incentives for conducting surveys of different parties with the same questionnaire is the possibility of synchronous observations not only at the aggregate level by parties but also individually. In these cases, the universe of analysis would not consist of delegates from each party but of all delegates in the party system or at least of those surveyed parties. To date, there has been little analysis at an individual-level using data derived from surveys of delegates (Barberà, Barrio and Rodriguez, 2002b; Baras et al., 2010b). Undoubtedly, research of Spanish party delegates at an individual level is a major challenge for the future. This would explore the behaviour, opinions and attitudes of delegates similar to the analysis conducted for party members (Seyd and Whiteley, 1992 and 2004). The main obstacle to the analysis at an individual-level arises from the strategy for data collection. As discussed above, the researchers’ main concern until now has been the representativeness of each of the surveys for eachparty and not the development of a pooled sample for all parties. As discussed in the next section, this is due to some extent to the lack of guarantees that researchers have had that sampling could be carried out successfully in all selected parties. The result of this process is that different numbers of surveys are collected for each party depending on the number of delegates and the success at the time of distributing the questionnaire. Secondly, this raises problems at the time of statistical aggregation of polls. The following question regarding the criterion of aggregation arises when creating a data matrix that adds the individual matrices of each party: indiscriminately aggregation that gives more weight to parties where a higher number of surveys were collected? Weighted aggregation depending on attendance at each convention? Weighted aggregation giving more weight to parties with greater affiliation or electoral support? In the first and second cases we would be looking at the population attending party conventions. In the third case, our focus would be on the population of activists of the party system. The first two approaches are simpler, the latter being preferable to the first approach as it avoids the bias resulting from the different degree of response. If we take this second approach, we should choose from different weighting criteria: number of votes obtained by each party in the most immediate election or number of party members. In the first case, the problem is the existence of various types of elections in multi-level systems, the second difficulty is that the information is not always available or it is not always reliable enough. Moreover, the problem of weighting is linked to the broader problem of how to conceptualize delegates and to establish a reference group that they represent theoretically. As discussed earlier, the answers to the academic debate have changed over time. Perhaps for this reason the most conservative response is that which can be most useful, surveys are representative of delegates attending the convention, therefore statistical inference should be limited to this group. In the Spanish case there is one more reason to justify choosing this option. Party conventions are still the place where the leader is elected, this means that delegates are their selectorate. Unfortunately, what may be a solution for the Spanish case is a serious obstacle in comparative terms as these processes have been transferred to members in many other parties through the primaries. 3.3. Practical problems Third but not least, there is a set of practical problems that arise during the investigation. The main one is accessing the parties. The first contact is usually established by sending a letter to the head of the party organization and in parallel relatively informal telephone contacts are made. The response from organizations varies; some are open and display frank cooperation while others show reluctance and mistrust towards the researcher. Parties tend to be suspicious and overcoming their reluctance is one of the most complex tasks researchers face. Sometimes we have not been allowed to distribute the survey. There was a situation whereby the distribution of the survey was aborted by the organization after having received approval and once the process was started. We conclude that access to parties is not always easy and success often depends more on informal channels, i.e. personal contacts than on formal and institutionalized channels. In some cases, the informal channel may be less effective when a personal contact being inadequate – due to internal likes and dislikes- this may hinder the completion of the survey. Once the organization’s authorization and support for the distribution of the survey is in place, we must overcome the distrust of many delegates who are not happy for the academy to analyse them. Much energy is needed to convince them of the benefits of the project and of the fact that there is no hidden agenda. The outcome is not always positive, especially when the reason for refusal to complete the survey is simply laziness or lack of time by the delegate. Furthermore, as already indicated, this type of survey does not imply significant bias because the conventions’ delegates have high levels of education. However, it is common for them to complain about the length of the survey, some of the questions and the guarantees of anonymity as someone with a deep knowledge of the party could actually identify some of the stakeholders interviewed. In short, these are the usual problems faced by researchers of political parties throughout history. However, the reluctance usually fades and organizations work amicably thereafter once a survey is distributed and the parties know the research method and understand that the survey is used for strictly academic purposes. Conclusions Although research on the evolution of party membership and the role of party members has been conducted in Spain, the truth is that convention delegates constitute the main subject in most of the research carried out. This type of research has a long academic tradition and it reached its peak in the mid-eighties with the EPPMLE project. Despite the theoretical, methodological or practical problems of this research, it is, undoubtedly, a useful tool to research a topic that otherwise could be hardly addressed. It is for this reason that these types of studies are still conducted in countries such as France, Italy, Belgium and Germany Despite its obvious usefulness, the main limitation of this type of research resides in the insufficient use of diachronic and synchronous comparative studies both nationally and in other countries. EPPMLE investigations marked a milestone in this direction which unfortunately seems not to have been sustained over time. The research team led by Montserrat Baras made efforts to draw comparisons at a national level where research is conducted on one party or the comparisons are between regional parties. As already noted, it is still a long way to achieving this and one of the potential difficulties lies in the recovery and systematization of information from past research. The next step in this project is to conduct research on party members in the medium and long term. This type of research has been successful in some countries like Norway, the United Kingdom and Belgium and it is undoubtedly one of the major gaps in this field in Spain. To avoid limitations in the research on delegates it would be desirable for this project not to be limited to a national scale but to include Europe from the outset. Annex I. Summary of data form different surveys of party conventions (1979-2008). AP 1981 PP 1993 PP 2008 PSOE 1979 PSOE 1981 T PSOE 1997 CDC 1979 CDC 1981 CDC 1996 CDC 2008 PSUC 1979 PSUC 1988 ICV 1996 ICV 2008 PSC 1979 PSC 1987 PSC 1997 PSC 2008 19,4 18,0 33,1 7,8 4,8 24,5 13,4 4,7 20 37,2 13,1 16,4 29 35 14,2 11,6 25 27,5 More than 65 years old 9,3 9,1 8,6 2,4 2,1 13 1,7 4,6 ‐ 11,2 5,8 6,5 19,4 4,9 1,1 9,1 14,1 4,9 University studies 56,2 ‐ 68,4 48,7 58,7 69,1 71,2 55,4 56 55,2 46,2 46,24 49 70,5 72,2 61,1 53 71,8 Non active 13,6 18,2 19,3 9,8 4,5 ‐ 6,6 5,2 12 17,2 17,3 13,9 9 18,3 15,9 11,9 9 16,2 Post within the party 66,5 76,0 77,6 ‐ 79,3 ‐ ‐ ‐ 69 71,2 ‐ 65,2 66,4 72 ‐ 66,2 70 66,2 Local ‐ ‐ 51,5 ‐ 38,7 ‐ ‐ ‐ 35 56,2 ‐ 51,1 ‐ 54,9 ‐ 40,8 ‐ 54,9 Public post 11,0 48,0 62 ‐ 57,3 70,1 ‐ ‐ 37 52,2 ‐ 19,1 28 30,8 ‐ ‐ 35,8 63,4 Local ‐ 28,7 50,4 ‐ ‐ ‐ 24,3 ‐ ‐ 18,8 18 ‐ 49,8 ‐ 23,2 ‐ ‐ 18,6 ‐ 31,2 ‐ ‐ ‐ 29,9 ‐ ‐ ‐ 26,8 Women ‐ ‐ 38,9 Fathers members of a party ‐ ‐ 32 Ex‐ member of another party ‐ ‐ 13,1 19,2 26,8 ‐ 6,9 ‐ ‐ 12,7 30,5 26,4 61,4 45,4 52,7 39,6 27 26,8 5 or more years as member ‐ ‐ 75,8 29,4 50,3 83 4,7 ‐ 74,2 82,2 74,8 ‐ 80,8 79,3 ‐ 73,6 82,1 83,2 Attended previous conventions ‐ ‐ 63,2 ‐ 79,7 80,4(b) 27,3 ‐ ‐ 57,4 4,2(b) ‐ 59,1 72 34,1 55 57,7 64,8 Member of a trade union ‐ ‐ 10,3 ‐ ‐ 50 9,9 5,4 10,4 11,8 77,7 80,8 66,4 54,6 67,8 72,6 44 40,1 27,7(a) 24,3 (a) Convention attendants 1400 ‐ 2643 897 799 945 900 1054 ‐ 2027 1217 ‐ ‐ 485 590 ‐ ‐ 1129 Surveys collected 269 ‐ 513 296 641 444 402 403 ‐ 521 884 528 524 328 437 223 260 142 % 19,2 ‐ 19,4 33,0 80,5 47,0 44,6 38,2 ‐ 27,2 72,6 ‐ ‐ 67,6 74,1 ‐ ‐ 11,6 Source: For AP/PP the data come from Capo and Botella (1985), López Nieto (1996 y 1997) and the 2008 Baras team’ survey. For the PSOE data come from Tezanos, (1980 y 1983), Botella (1985), Méndez and Santamaría (2001). For CDC sources are Botella, et. alt (1979), Pitarch et. alt (1980), Marcet (1984), Sánchez (1999), and the 2008 Baras team’ survey. For the PSC data come from Botella et. alt (1979), Pitarch, et. alt. (1980), Marcet (1984), Colomé (1988, 1992b,1998), Sánchez (1999), and the 2008 Baras team’ survey. Fort the PSUC/ICV data are from Botella et. alt (1979), Pitarch, et. alt. (1980), Marcet (1984), Colomé (1992a), Sánchez (1999), and the 2008 Baras team’ survey. a. Father member of the same party; b. Attended the last party conventions. – no data available. References Argelaguet, J. (2008): “Elementos de cambio y de continuidad en la militancia de ERC (1993-2004)” Papers-Revista de Sociologia, 92: 75-96. Baras, M. (ed) (2004): Els militants dels partits politics a Catalunya. Perfils socials i percepcions polítiques. Barcelona: Institut de Ciències Polítiques i Socials. Baras, M., J. Argelaguet and P. Correa (2011) “Radiographie des militants des partis politiques en Espagne“, Pole Sud, 33 :65-82. Baras, M, Barberá, O. and Barrio, A. (2008): “Más allá de la ley de May: disparidades curvilíneas y conflicto intrapartidista. El caso de Cataluña”, Working Paper, 267, Barcelona, Institut de Ciències Polítiques i Socials. Baras, M., O. Barberà, A. Barrio y J. Rodríguez (2010a): “Estructuras de opinión en los partidos políticos y competencia multidimensional: el caso de Cataluña (2004)”, Revista Española de Ciencia Política, 22: 49-70. Baras, M., O. Barberà, J. Rodríguez and A. Barrio (2010b): “Representation within political parties:Spanish party delegates’ opinions on intra-party democracy”, paper presented at the workshop “Party members and Activits: State of the art and comparative perspectives”, XVIInd ISA World Congress of Sociology, Gothenborg, 11-17 july 2010. Barberà, O., A. Barrio and J. Rodríguez (coord) (2002a): Els militants de les organitzacions polítiques juvenils a Catalunya. Barcelona: Fundació Jaume Bofill/Diputació de Barcelona. Barberà, O., A. Barrio and J. Rodríguez (2002b): “Els militants de les joventuts polítiques i els joves catalans” en O. Barberà, A. Barrio and J. Rodríguez, coord., Els militants de les organitzacions polítiques juvenils a Catalunya. Barcelona: Fundació Jaume Bofill/Diputació de Barcelona Botella, J., J. Capo and J. Marcet (1979): “Aproximación a la sociología de los partidos políticos catalanes”, Revista de Estudios Políticos, 10: 143-206. Capo, J. and J. Botella (1985): “Las elites intermedias de los partidos políticos españoles: resultados preliminares”, AAVV Problemas actuales del Estado social y democrático de derecho [Actas del IV Congreso Español de Ciencia Política y de la Administración]. Alicante: Universidad de Alicante. Cayrol, R. (1974): “Les militants du Parti socialiste, contribution à une sociologie”, Projet, 88: 929-940. Cayrol, R. and C. Ysmal (1982): “Les militants du PS originalité et diversités”, Projet, 165: 572-586. Cayrol, R. and K. Reif (1983): “Conscience Individuelle Et Discipline Partisane: La culture politique des militants de l'Europe de l'Ouest”, International Political Science Review, 4: 36-47. Colomé, G. (1988): El Partit dels Socialistes de Catalunya. Estructura, funcionament i electorat. Barcelona: Edicions 62. Colomé, G. (1992a): “La composicio sociològica dels delegats del VIII Congrés del PSUC a l’Assemblea Fundacional d’IC”, Papers – Revista de Sociologia, 39: 43-58. Colomé, G. (1992b): “Los delegados del Partit dels Socialistes de Catalunya (19821990), Revista del Centro de Estudios Constitucionales, 12: 25-44. Colomé, G. (1996):“The Leftist parties in Catalonia”, Working Paper 88/1996. Barcelona: Institut de Ciències Polítiques i Socials. Colomé, G. (1998): Els delegats del VIII Congrés del PSC”, Papers de la Fundació 106/199X. Barcelona: Fundació Rafael Campalans. Duverger, M. (1954): Political Parties: their organization and activities in the modern world. London: Methuen. Heidar, K. (2006): “Party membership and participation” en Katz, R. and W. Crotty, eds., Handbook of Party Politics, London: Sage. Linz, J. J. and J.R. Montero; A.M. Ruiz Jiménez (2005): “Elecciones y política”, Carreras, A. and X. Tafunell (coord): Estadísticas históricas de España: Siglo XIX-XX. Madrid: Fundación BBVA. López Nieto, L. (1996): “Los mecanismos del aparato: organización y funcionamiento de AP”, Ponencia presentada en el II Congreso de la Asociación Española de Ciencia Política y de la Administración. Santiago de Compostela 18/20 abril. Lopez Nieto, L. (1997): “Il lungo cammino della destra spagnola. L’ascesa elettorale di Alianza Popular/Partido Popular”, Quaderni dell’Osservatorio Electorale 37/1997. Marcet, J. (1984): Convergència Democràtica de Catalunya. El partido i el moviment. Barcelona Edicions 62. May, J.D. (1973): “Opinion structure of political parties; the special law of curvilinear disparity”, Political Studies, 21/2: 135-151. Méndez, M.,L. Morales and L. Ramiro (2004) “Los afiliados y su papel en los partidos políticos españoles”, Zona Abierta, 108-109: 153-207 Méndez, M. and J. Santamaría (2001): “La ley de la disparidad ideológica curvilínea de los partidos políticos: el caso del PSOE”, Revista Española de Ciencia Política, 4: 35-69. Niedermayer, O (1986): “Methodological and practical problems of comparative party elites research: the EPPMLE project”, European Journal of Political Research, 14: 253-259. Oñate, P. (2008): “Los partidos políticos en España”, Jiménez de Parga, M and F. Vallespín (eds): España Siglo XXI – La política. Madrid: Biblioteca Nueva. Pierre, J. (1986a): “Attitudes and behaviour of party activists. A critical examination of recent research on party activists and 'Middle-Level Elites'”, European Journal of Political Research 14:4, 465–479 Pierre, J. (1986b): “Three men run aground?”, European Journal of Political Research, 14: 691-693. Pitarch, I. J. Botella, J. Capo and J. Marcet (1980): Partits i parlamentaris a la Catalunya d’avui (1977-1979). Barcelona: Edicions 62. Ramiro, L. (2005): 'Programmatic Adaptation and Organizational Centralization in the AP-PP', South European Society and Politics, 10:2, 207- 22. Ramiro, L. and L. Morales (2004): “Latecomers but 'early-adapters'. The Adaptation and Response of Spanish Parties to Social Changes”, Lawson, K. and T. Poguntke (eds): How Political Parties respond. ECPR/Routlegde. Reif, K., O. Niedermayer and H. Schmitt (1986): “Quantitative = survey = attitudes = issue stands = ideology?” A rejoinder to Jon Pierre's 'Attitudes and behaviour of party activists' European Journal of Political Research 14:5-6, 685–690 Reif, K., R. Cayrol and O. Niedermayer (1980): “National Political Parties’ Middle Elites and European Integration”, European Journal of Political Science, 8: 91112. Rodríguez, J., O. Barberà, A. Barrio and M. Baras (2011): “¿Se han hecho más democráticos los partidos en España? La evolución en las reglas de elección del líder (1977-2008)”, Revista Mexicana de Ciencias Políticas y Sociología, 208. Rohrschneider, R. (1994): “How iron is the iron law of oligarchy? Robert Michels and national party delegates in eleven West European democracies”, European Journal of Political Research, 25: 207-238. Sánchez, J. (ed) (1999): Estudis de les elits dels partits polítics de Catalunya. Barcelona: Institut de Ciències Polítiques i Socials. Seyd, P. and P. Whiteley (1992): Labour’s grass roots: the politics of party membership. Oxford: Clarendon Press. Seyd, P. and P. Whiteley (2004): “British party members: an overview”, Party Politics, 10: 355-366. Tezanos, J. F. (1980): “Radiografia de dos Congresos: una aportación al estudio sociológico de los cuadros políticos del socialismo español”. Sistema, 35: 7999 Tezanos, J. F. (1981): “Estructura y dinámica de la afiliación socialista en España”, Revista de Estudios Políticos, 23: 117-152. Tezanos, J. F. (1983): Sociología del socialismo español. Madrid: Técnos. Whiteley, P., P. Seyd, and J.J. Richardson (1994): True blues: The politics of Conservative Party membership. Oxford: Oxford University Press.