Factors associated with Spanish older people’s membership in political organizations: the role of active aging activities

Abstract

This study explores older people’s membership in political organizations by using data from the Survey on older people 2010, carried out by Spain’s National Institute for older people and social services. The objectives were to describe the extent of this kind of participation among Spaniards aged 65 and over, and to analyze the factors that are associated with it. Results show that only slightly less than 7 % of the sample belonged to a political organization. To analyze the factors related to this membership, a set of models of multivariate analyses were run, including socioeconomic resources and participation in other types of active aging activity (participation in leisure, learning, and productive activities). Educational level, leisure activities, learning activities, and only volunteering in the case of productive activities were found to be associated with membership in political organizations. Results provide partial support for the socioeconomic resources model and suggest that engagement in leisure activities, learning activities, and volunteering might have an enhancing effect on membership in political organizations.

Introduction

The new generations of elders are changing what we know about old age. They are more likely to be better educated and have better health, greater financial resources, and a longer life expectancy than their predecessors. In addition, many of them are keen to contribute to family and community, showing that old age can also be a time to engage or remain engaged in meaningful roles. Understanding and promoting the involvement of older people in productive activities have been placed on the agenda of researchers and policymakers in recent decades (Hinterlong and Williamson 2006). In this respect, the World Health Organization (WHO 2002) has coined the term “active aging” to highlight not only the ability to keep physically active in later life but also the possibility of staying involved in social, economic, cultural, spiritual and civic affairs.

Despite the growing influence of the concept of active aging in social policy and research, however, there is still little consensus regarding its definition (Boudiny and Mortelmans 2011; Perales et al. 2014). Studies of the concept tend to cover just one of its dimensions, applying either an economic-oriented approach focused on extending working lives, or a health-oriented approach focused on physical activity (Boudiny 2013). Among the many problems that such one-dimensional definitions imply (discussed in detail elsewhere, see Holstein and Minkler 2003; Ranzijn 2010), perhaps the most important are that they fail “to acknowledge the heterogeneous nature of older people” (Foster and Walker 2013, p. 4) and that they neglect other activities beyond work and physical activity that are meaningful and contribute to older people’s well-being, such as leisure activities, learning activities, or civic engagement (Boudiny 2013). With regard to the latter of these two points, one kind of activity that has not been analyzed in depth within the context of active aging is political participation.

The present study, which seeks to fill this gap, is aimed at exploring older people’s membership in political organizations by using data from a Spanish representative sample. Our objectives are to describe the extent of this kind of political participation in Spain and to identify the factors that are associated with such membership. Specifically, we will test which role active aging activities might have on membership in political organizations.

Political participation and aging

Political participation in older age has usually been studied within the wider framework of civic engagement, a concept for which, to date, there seem to be no single, widely agreed-upon meaning (Adler and Goggin 2005). Although some scholars have restricted the definition of civic engagement to specific actions such as community service or political involvement, others (for instance Putnam 2000) have used the term quite broadly to refer to the wide range of activities that create social capital. As a consequence, in recent years there has been a growing awareness of the importance of accurately distinguishing among the different activities classified under the concept of civic engagement (Berger 2009).

Two distinct spheres within this concept have been identified (Adler and Goggin 2005; Berger 2009; McBride et al. 2006): social participation and political participation. The former refers to actions that primarily connect individuals to others, while the latter involves actions that are mainly oriented toward influencing political outcomes. Thus, political participation can be defined as “… the individual, non-professional, and voluntary participation in activities that aim, directly or indirectly, at influencing political outcomes, changing the institutional premises for politics or affecting the selection of personnel or their choices” (Nygard and Jakobsson 2013a, p. 67).

Despite the importance of political participation in sustaining and improving representative democracies, especially in the current context of aging populations, research on this topic is still at an early stage (Nygard and Jakobsson 2013a). At least three aspects are in need of further development. First, most previous research on political participation has focused on a single activity, namely voting (Binstock 2000; Strate et al. 1989), or on a combination of a few types of political activity, such as voting, attending meetings, and contacting politicians (Nygard and Jakobsson 2013a, b; Verba et al. 1995). So far, however, there has been little discussion of other political activities, such as older people’s membership in political organizations. This type of activity is particularly interesting since it is likely to be more stable across time than other types of political involvement (Verba et al. 1995).

Second, most research on older people’s political participation comes from English-speaking countries. However, the experience of old age may differ according to country and may have different consequences for political participation (Goerres 2009). The case of Spain is particularly interesting in this respect. Spanish elders lived most of their childhood and youth under a dictatorship which banned any kind of political expression. It was a time of severe limitations of liberties and rights, followed by a smooth transition to democracy which allowed the re-emergence and legalization of political organizations. Comparatively, elders are still the cohort with the lowest levels of political engagement in Spain. Although Spaniards aged 65 and older have voting turnout rates that are similar to those of younger generations (Centro de Investigaciones Sociológicas 2011), they are less likely to take part in a demonstration or a strike, to sign a petition (IMSERSO 2008), or to be members of a political association (Morales 2003). Moreover, in the last General Election in November 2011, persons aged 65 and older constituted 23.2 % of the population entitled to vote (Instituto Nacional de Estadistica 2011). This means that, if mobilized, they could represent a significant lobby for influencing politics and current social policies.

Third, although some studies have analyzed the relationships and mutual influences between different forms of political participation, the possible trade-off between political participation and other forms of active aging activity, such as leisure, learning, or productive activities has not been examined. The present study aims to fill this gap.

Membership in political organizations and active aging activities

When it comes to the question as to why older people participate in politics, most previous research has focused on the individual resources that foster political participation (Nygard and Jakobsson 2013a). For example, in their study of political participation, Verba and Nie (1972) showed that participation, in a broad sense, was related to socioeconomic status. People with a higher level of resources such as income, education, or occupational status were more likely to engage in political activities. In the same vein, the classic resources model of civic voluntarism (Verba et al. 1995) states that political participation could be explained as a function of mobilizing factors, such as higher education, greater civic skills or more extensive social networks. In the present study, we test the relevance of education and income in the case of an institutionally channeled political activity, specifically, older people’s membership in political organizations.

Education has a strong effect on a wide range of political activities. People with a higher educational level are more likely to vote (Wolfinger and Rosenstone 1980), to attend meetings and volunteer for a candidate (Burr et al. 2002), and to contact a politician, appeal against a decision, write a letter to the press, or boycott a product or company (Nygard and Jakobsson 2013b). Melo and Stockemer (2014) argue that the ability to think critically and understand complex political scenarios, something that is fostered by the level of educational attainment, underlies the higher likelihood of participating in politics.

Income, too, is related to political participation. Wealthier people are more likely to engage in many forms of political activity. The previous literature has found that they are more likely to donate money (Burr et al. 2002), be affiliated with a political organization, contact public officials, make campaign contributions, or do campaign work (Verba et al. 1995). Having adequate financial resources facilitates the development of civic skills and social capital, which in turn may increase the likelihood of political participation (Burns et al. 2001; Melo and Stockemer 2014).

In addition to the role of socioeconomic resources, other factors may be associated with membership in political organizations as well. As explained earlier, belonging to a political organization could require a more stable level of commitment than other kinds of political activity. Thus, people who participate in this form of political activity may not only need a high level of civic skills (Verba et al. 1995), but also require a considerable amount of time (Burr et al. 2002). With respect to the allocation of time and effort, other activities could either compete with or complement political participation. For instance, being involved in other active aging activities could divert the time and effort that are needed to be engaged in political organizations. On the contrary, participating in other activities could potentially provide social supports and institutional ties, as well as generating interest that could facilitate joining political organizations.

Although there is thus far no study examining this issue in relation to older people’s political participation, there are some related inquiries that focus on volunteering. For example, it has been found that employed adults tend to have higher rates of volunteering than the unemployed or retirees (Choi 2003), and that volunteering also seems to be fostered by caregiving (Burr et al. 2005). These findings suggest that such activities may complement each other rather than compete with one another. Therefore, it is relevant to explore whether or not the fact of engaging in other activities is related to older people’s membership in political organizations—and, if so, what seems to be the effect (enhancing or detrimental) on membership in political organizations.

In this respect, the issue of type of activity might be especially relevant, although it has rarely been studied (Burr et al. 2007). Older people who maintain an “active” lifestyle usually engage in a broad range of activities. However, not all the activities labeled as “active” have the same characteristics. Older people’s participation includes a number of activities which might require differing investments of resources and might be pursued in response to different motivations. Such dimensions could be a key to learning whether being involved in certain activity increases or undermines the likelihood of belonging to political organizations.

In this vein, Manell (1993) argued that activities can be classified according to the level of effort required to undertake them, that is, as high-or low-investment activities. In their analysis of social participation, Bukov et al. (2002) proposed, among other dimensions, that social activities may be classified as self-oriented activities (when the individual receives resources from the social environment) or socially oriented activities (when the individual contributes resources to the social environment). Taking these dimensions into account, it is possible to identify three different types of active aging activities: (1) self-oriented activities involving low investment, (2) self-oriented activities involving high investment, and (3) socially oriented activities which, by nature, tend to involve a high investment. Leisure activities are a good example of the first type. In this case, personal enjoyment is the main motivation, and free time is the only requirement to undertake them. The second kind of activity is also self-oriented, but involves a higher investment of effort. This is the case with learning activities, such as participating in educational programs for older adults. Finally, the third type of activity is oriented to other people and generally requires a high investment of effort. This is the case with productive activities, which can be limited to the micro-level of family relationships (such as taking care of relatives) or extended to the social level (such as volunteering). Leisure, learning, and productive activities have been considered as part of the active aging model in previous studies (e.g., Bass and Caro 2001; Boulton-Lewis 2006; Boudiny and Mortelmans 2011). In the current study, we explore the relevance of undertaking leisure, learning, and productive activities for older people’s membership in political organizations.

Objectives and hypotheses

The objectives of this study were to describe the extent of older people’s membership in political organizations and to analyze the factors associated with such involvement. First, we tested a model based on socioeconomic resources, including income and education. Next, we added to that model the co-occurrence of different types of active aging activity (participation in leisure, learning, and productive activities) in order to find out whether the inclusion of these variables increased the predictive value of the model.

Two hypotheses arose from the consideration of the theoretical framework:

• Higher levels of socioeconomic resources (education and income) will be associated with a higher likelihood of membership in political organizations.

• If other active aging activities have a detrimental effect on membership of political organization, they will tend to diminish the likelihood of being a member of a political organization. This negative association will be stronger in the case of high investment activities (e.g., learning or productive activities) than of low-investment activities (e.g., leisure). On the other hand, if active aging activities are mutually reinforcing, participation in leisure, learning or productive activities would be expected to be associated with a higher likelihood of being a member of political organizations.

Materials and methods

Participants

The data for this study were obtained from the Survey on older people 2010, which was carried out by Spain’s National Institute for older people and social services. This survey was designed to study the living conditions of older persons and was based on a nationally representative Spanish sample of 2535 respondents. Participants were selected by simple random sampling using the telephone directory. The sample was also weighted by sex, age (four age groups: 65–69, 70–74, 75–79, 80 and over), autonomous community (17 Spanish regions), and location size (up to 5000 inhabitants, 5001–10,000 inhabitants, 10,001–20,000 inhabitants, 20,001–100,000 inhabitants, and 100,001 inhabitants or more) to ensure that the distribution of these variables in the final sample resembled their distribution in the Spanish population. Interviews were conducted by phone.

Measures

Belonging to a political organization was the dependent variable and we used indicators of socioeconomic resources and indicators of three types of active aging activity (participation in leisure, learning, and productive activities) as predictive independent variables.

Outcome variable

The dichotomous outcome variable was belonging to a political organization, defined as actual membership in a political organization, such as political parties or older people’s associations with political aims. Participating in such political organizations was coded as 1, not participating as 0.

Socioeconomic resources

Regarding socioeconomic resource variables, we used education and income. Education was grouped into three categories (incomplete primary education or less, primary education, secondary education or higher), and income into three categories (≤300, 301–900, ≥901 euros per month). Income levels were established according to the distribution of frequencies, grouping categories which were closer to the distribution tertiles.

Engagement in different types of active aging activity

Participation in leisure activities was measured by means of three activity variables: attending shows, doing sports, and attending a recreational center for older adults at least once over the past week. Participation in learning activities was measured by one activity variable: attending courses at least once over the past week. Participation in productive activities was measured through three variables: taking care of a dependent person over the past 2 months, taking care regularly of grandchildren while their parents work, and participating regularly in any kind of volunteer activity. For every activity, participation was coded as 1, no participation as 0.

Control variables

As in other studies on older people’s political participation (e.g., Nygard and Jakobsson 2013a, 2013b; Kam et al. 1999) gender and age were considered as control variables. Both were entered as dichotomous dummy variables. Age was grouped into two categories: 65–74, 75, and over.

Analytic strategy

First, bivariate analyses were conducted to test the association between the dependent variable, membership in a political organization, and the respective independent variables. Second, a successive series of multivariate analyses were run. In a first step, we tested a model based on socioeconomic resource variables, and then—in three steps—we added the indicators of engagement in different types of active aging activity (leisure, learning, and productive activities) in order to learn whether the inclusion of these variables increased the predictive value of the model.

Results

Only 6.9 % (n = 176) of the sample belonged to a political organization. As noted, we carried out a series of bivariate analyses first to explore the relationship between this dependent variable and all the independent variables.

There were significantly more men than women among the older people who were members of a political organization (χ ² (1) = 6.61; p < 0.001). Members were also more likely to be better educated (χ ² (2) = 39.58; p < 0.001), and to have a higher income (χ ² (2) = 58.95; p < 0.001) (Table 1).

Table 1.

Comparative socioeconomic profile of members versus non-members of political organizations (in percentages)

Variable	Members (n = 176)	Non-members (n = 2359)	Total (n = 2535)
Gender***
Male	63.1	50.5	50.5
Female	36.9	49.5	49.5
Age**
65–74	59.1	49.0	49.7
75 and over	40.9	51.0	50.3
Education level***
Incomplete primary or less	56.4	76.4	75.0
Primary education	21.5	14.7	15.2
Secondary or more	22.1	8.9	9.8
Income***
≤300 euros	11.9	16.8	16.5
300–900 euros	49.1	67.9	66.5
>900 euros	39.0	15.3	17.0

p-values are based on the Chi square statistic. * p < 0.05; ** p < 0.01; *** p < 0.001

As for engagement in other types of activity, older people who belonged to political organizations were more likely to participate in leisure activities. Specifically, they were more likely to have attended shows (χ ² (1) = 30.64; p < 0.001), to have done sports (χ ² (1) = 9.01; p < 0.01), or to have attended a recreational center for older adults (χ ² (1) = 38.79; p < 0.001). Members of political organizations were also more likely to participate in learning activities, defined as attending courses (χ ² (1) = 17.21; p < 0.001). In addition, they were more likely to participate in two kinds of productive activities: caring for grandchildren (χ ² (1) = 4.07; p < 0.05), and volunteering (χ ² (1) = 4.63; p < 0.05) (Table 2).

Table 2.

Comparative profile of members versus non-members of political organizations (in percentages), based on the co-occurrence of different types of active-aging activity

Variable	Members (n = 176)	Non-members (n = 2359)	Total (n = 2535)
Leisure activities
Attending shows***
Yes	21.0	8.4	9.3
No	79.0	91.6	90.7
Doing sports**
Yes	39.8	29.0	29.8
No	60.2	71.0	70.2
Attending a recreational center for older adults***
Yes	60.3	36.6	38.2
No	39.7	63.4	61.8
Learning activities
Attending courses***
Yes	15.3	6.8	7.4
No	84.7	93.2	92.6
Productive activities
Caring for dependents
Yes	9.9	9.6	9.6
No	90.1	90.4	90.4
Caring for grandchildren*
Yes	36.4	29.2	29.7
No	63.6	70.8	70.3
Volunteering*
Yes	12.5	7.9	8.2
No	87.5	92.1	91.8

p-values are based on the Chi square statistic. * p < 0.05; ** p < 0.01; *** p < 0.001

In order to explore the relationships between these variables, a series of logistic regression analyses were run (Table 3). When only socioeconomic resource variables were entered into the regression, results showed two significant main associations. The odds of belonging to a political organization were twice as high for people with primary studies (odds ratio = 2.00, p < 0.01) and more than twice as high for people with secondary studies (odds ratio = 2.40, p < 0.01). No effects were found for income when education was considered as well.

Table 3.

Variables associated with membership of political organizations: hierarchical logistic regression

Variable	Model 1	Model 2	Model 3	Model 4
	Odds ratio [95 % CI]	Odds ratio [95 % CI]	Odds ratio [95 % CI]	Odds ratio [95 % CI]
Gender (Female = 0)
Male	1.07 [0.72–1.61]	0.87 [0.57–1.33]	0.95 [0.62–1.45]	1.02 [0.66–1.58]
Age (65–74 = 0)
75 and over	0.70 [0.49–1.01]	0.83 [0.55–1.23]	0.83 [0.55–1.24]	0.90 [0.59–1.37]
Education level (< primary = 0)
Primary education	2.00** [1.28–3.12]	1.93** [1.23–3.02]	1.88** [1.20–2.94]	1.87** [1.19–2.94]
Secondary or more	2.40** [1.42–4.06]	2.18* [1.27–3.74]	2.09** [1.21–3.59]	2.01* [1.16–3.48]
Income (≤300 euros = 0)
301–900 euros	0.90 [0.51–1.60]	0.86 [0.48–1.55]	0.86 [0.48–1.55]	0.84 [0.46–1.51]
>900	1.71 [0.87–3.35]	1.74 [0.88–3.45]	1.82 [0.91–3.62]	1.79 [0.89–3.57]
Attending shows (No = 0)
Yes		1.84* [1.15–2.96]	1.70* [1.05–2.76]	1.75* [1.08–2.84]
Doing sports (No = 0)
Yes		1.06 [0.71–1.60]	1.00 [0.67–1.50]	0.98 [0.65–1.47]
Attending a recreational center for older adults (No = 0)
Yes		2.16*** [1.49–3.12]	2.03*** [1.40–2.95]	2.05*** [1.41–2.97]
Attending courses (No = 0)
Yes			2.23** [1.37–3.64]	2.04** [1.24–3.36]
Caring for dependents (No = 0)
Yes				0.99 [0.54–1.82]
Caring for grandchildren (No = 0)
Yes				1.16 [0.78–1.72]
Volunteering (No = 0)
Yes				1.80* [1.04–3.11]
Model sum. Chi Square (df, p value)	49.94 (6, <0.000)	75.42 (9, <0.000)	84.63 (10, <0.000)	89.11 (13, <0.000)
Log likelihood	981.66	956.18	946.97	942.49
Nagelkerke	0.061	0.091	0.102	0.107

Model 1 Socioeconomic resources variables, Model 2 Socioeconomic resources variables + Leisure activities variables, Model 3 Socioeconomic resources variables + Leisure activities variables + Learning activities variables, Model 4 Socioeconomic resources variables + Leisure activities variables + Learning activities variables + Productive activities variables

* p < 0.5; ** p < 0.01; *** p < 0.001

When leisure activities were included in the model (model 2), the above effects were maintained and two other main relationships were found. The odds of belonging to a political organization were more than twice as high for people who had attended a recreational center for older adults (odds ratio = 2.16, p < 0.001) than for those who had not. Furthermore, people who had attended shows had an 84 % higher odds of being members than people who had not attended (odds ratio = 1.84, p < 0.05). The goodness of fit of the model, as indicated by Nagelkerke’s pseudo R ², increased from 0.061 to 0.091.

The inclusion of learning activities into the regression (model 3) added another significant main association. The odds of belonging to a political organization was more than twice as high for people who had attended courses (odds ratio = 2.23, p < 0.01) than for those who had not. All the previous effects remained. The goodness of fit of model 3 was better than the one provided by model 2 (Nagelkerke’s pseudo R ² increased from 0.091 to 0.102).

Finally, when productive activities were included in the regression (model 4), only volunteering showed a significant association with this kind of political participation. Thus, the odds of being a member of a political organization were almost twice as high for people who also volunteered (odds ratio = 1.80, p < 0.05). Caring for dependents and caring for grandchildren did not have significant effect on the probability of belonging to a political organization. Although model 4 was the one that provided the best fit, Nagelkerke’s pseudo R ² was just slightly higher than in model 3 (0.107 vs 0.102).

Since membership in a political organization could be considered a rare event (with a rate as low as 7 % in our sample), we tested the robustness of the findings by following the suggestions described in King and Zeng (2001) and using King’s software for treating rare events in logistic regression, implemented in Stata 12. This procedure yielded exactly the same pattern of significant (and non-significant) statistical relationships that was described above.

Discussion

The present study was designed to explore older people’s membership in political organizations. Specifically, we were interested in addressing the following objectives. The first objective sought to determine the extent of Spanish older people’s membership in political organizations. The second objective of the study was to ascertain the relevance of two models, one based on socioeconomic resources and the other adding the co-occurrence of different types of active aging activity, to predict this type of political behavior.

Regarding the first objective, a key result of this study is that only slightly less than 7 % of Spanish people over 65 years old belonged to political organizations. This percentage contrasts with the around 15 % of members reported by Morales (2003) among Spaniards from the dictatorship generation. However, it must be taken into account that differences in question wording and the inclusion of different political groups in the lists presented to participants make it difficult to compare results from different surveys (Morales 2002). Although our study does not explain why participation was so low, generational cohort effects could play an important role in this issue. As explained earlier, the current generation of Spanish older people belongs to the post-Spanish Civil War generation. They grew up in a dictatorship that severely curtailed liberties and rights and banned any kind of political activity that was inconsistent with the official regime. Previous research has suggested that people who grew up in times of limited political participation might perceive less scope for political participation in later life (Goerres 2009), and this could be the case of Spanish elders. Unfortunately, the nature of the data analyzed in this study do not allow us to draw conclusions regarding this matter.

In relation to our second objective, ascertaining the relevance of two models to predict political participation, the following conclusions can be drawn. First, the results of this study provide partial support for the socioeconomic resources model (Verba and Nie 1972). In accordance to our first hypothesis, educational level had a strong effect on older people’s membership in political organizations: the higher their educational attainment, the greater their likelihood of being involved in a political organization.

However, in contrast to what we hypothesized, income was not independently related to membership in political organizations. Although this variable was significantly associated with this kind of political participation at the bivariate level, it was not found to be significant in subsequent multivariate analyses. In other words, the connection between income and membership in political organizations may have been spurious and caused by the influence of education, which is discussed above. So, our data show that being involved in political organizations depends more on educational background than on income per se.

Another noteworthy result of our study comes from the second prediction tested. Our results show that participating in active aging activities (leisure, learning, and some productive activities) actually increased the likelihood of being a member of a political organization. That is, our data tend to support a mutually reinforcing model of active aging activities, rather than a competition model.

Regarding leisure activities, the increased odds could be attributed to the effect of social relationships on political participation. In this vein, richer interpersonal networks may be crucial to political recruitment and these networks are often enhanced through participation in social institutions. Verba et al. (1995) stated that social institutions play an important role in political engagement, serving as a site for exposure to political cues and as a locus for recruitment to political activity.

In relation to learning activities, our results suggest that older people who are involved in political organizations are not only better educated, but also more likely to undertake learning activities. Education enhances the development of civic skills, which in turn facilitate participation (Verba et al. 1995). Thus, this participation in learning activities may, in turn, encourage greater involvement in political organizations.

Such pattern of results provide support for the resources model of civic voluntarism (Verba et al. 1995), which states that political participation is a function of mobilizing factors, such as educational opportunities, extensive social networks or institutional ties. Moreover, our results suggest that although social participation may not be political in itself, it can be a resource that fosters political participation (Berger 2009).

Finally, our results also show that, although active aging activities tend to reinforce participation in political organizations, not every activity contributes to the same degree. Specifically, among the productive activities, only volunteering increased the likelihood of being a member of political organizations. The fact that this effect did not hold in the case of care-related activities may suggest that productive activities might have different meanings at the macro-social level than at the micro-level of family relationships. In this vein, we must mention that traditional family roles are still strongly present in the current generation of Spanish older people. Care-related tasks, such as looking after grandchildren while parents are at work, are frequent and intensive activities among Spanish elders (Hank and Buber 2009). Taking our results into account, almost half of all participants undertake one of the two care-related activities considered (caring for dependents and caring for grandchildren). Thus, it could be that these activities are not associated with other kinds of activities principally because they are widely prevalent among the Spanish population and might be perceived as “natural” and compulsory roles among Spanish elders.

The current study has a number of limitations which need to be taken into account when interpreting the results, and which should be addressed in further research. Firstly, the Survey on Older People, which is a cross-sectional survey, does not allow us to differentiate the potential role of any cohort effect in the results. Longitudinal studies are needed to chart the development of political participation in the last decades of life and to distinguish between cohort and age-related effects. Secondly, the survey does not provide data regarding other kinds of institutionalized political activities (e.g., voting) and non-institutionalized political activities (e.g., signing a petition or boycotting). Consequently, our conclusions are limited solely to the specific type of political involvement studied, namely, membership in political organizations.

Thirdly, although membership in a political organization may require a deeper and more stable level of commitment than other types of political activities, the survey data available did not include measures of the intensity and frequency of this political participation or of the time devoted to it. Therefore, we cannot draw conclusions regarding the level of commitment of members of political organizations. Finally, given the cross-sectional nature of the survey, the possibility of reverse causality cannot be over-ruled. That is, it might be the case that membership in political organizations lead to participation in the other activities considered, rather than vice versa. Our suggestion of effects stem from theory. Their unequivocal empirical verification, however, requires additional research.

Despite these limitations, our research nevertheless contributes to the literature on older people’s political participation by offering results that were obtained from a nationally representative sample. This is an important contribution taking into account that research on older people’s political participation is still at an early stage. Furthermore, the present inquiry explores a kind of political activity that has been largely underexplored in previous literature. Finally, our analysis provides new perspectives on the factors that are associated with membership in political organizations, and as such, proposes potential targets for policy and programs, if the aim is to increase older people’s political participation.