National day participation among immigrants in the Netherlands: the role of familiarity with commemorating and celebrating

ABSTRACT

This study explores to what extent varying levels of host national day participation among immigrants can be explained by previous participatory experiences related to their country of origin, in addition to socio-cultural factors related to the current country of residence. Utilising data from a large online immigrant panel, we concentrate on two prominent national days in the Netherlands: Remembrance Day and Liberation Day. On these days, Dutch society commemorates victims of the Second World War and celebrates freedom, respectively. Our results indicate that Dutch national day participation among immigrants is determined largely by previous familiarity with commemorating and celebrating through participation in holidays specific to immigrants’ country of origin. These findings highlight the need to place more emphasis on the role of previous participatory experiences when examining immigrants’ current patterns of participation in the host society.

KEYWORDS:

• National days
• participation
• immigrants
• country of origin
• The Netherlands

Introduction

Over the past few decades, increasing inflows of immigrants have led many European societies to introduce stricter integration requirements for naturalisation (Ersanilli and Koopmans 2011). Language requirements have been expanded and mandatory integration courses now often include a cultural section focusing on the most important norms and values of the host society. In addition, growing emphasis is being placed on national history (Duyvendak 2011; Miller and Ali 2014). Knowledge of one’s past and the associated rituals to remember is supposed to help a nation understand ‘who we are’ (Sapiro 2004, 10), and is therefore thought to be a vital aspect of successful political socialisation. One example is participation in the activities organised on designated dates on which a nation commemorates or celebrates a defining event in its history as a nation, known as ‘national days’ (or ‘holidays’; see Etzioni 2000). Recent research has shown that even though more frequent national day participation is not associated with stronger feelings of national belonging amongst all citizens, this association is certainly present amongst non-western immigrants (Coopmans, Lubbers, and Meuleman 2015).

At the same time, citizens from non-western origin were found to participate significantly less frequently in national days than the native population. This is not surprising, considering that large-scale immigration to most European countries did not start until after the Second World War (Messina 2007). Many of the national days referring to a ‘common’ past therefore only reflect the history of the native population. National days relating to a restricted audience do however run the risk of reinforcing societal segregation (see Collins 2004 on ritual insiders and outsiders). More research is therefore needed into the immigrants who decide to participate in national days, and the reasons why some immigrants participate more frequently than others.

Building upon insights from research on other forms of participation (e.g. voting, volunteering), this study aims to answer these questions by examining to what extent immigrants’ host national day participation can be explained by previous participatory experiences related to their country of origin, in addition to socio-cultural factors related to their current country of residence. We follow a recent line of research suggesting that to understand immigrants’ political participation, not only immigrants’ current situation should be taken into account, but also – and perhaps even more so – the ‘previous participatory context’, in other words, experiences related to immigrants’ country of origin (Voicu and Comşa 2014; Voicu and Rusu 2012).

Utilising data from a large online immigrant panel, this study concentrates on two prominent national days in the Netherlands, both dedicated to the commemoration of the Second World War: Remembrance Day and Liberation Day. The multi-ethnic context of the Netherlands, which has several large immigrant groups, is relevant for examining ethnic group differences. Moving beyond mere comparisons between natives and non-natives, we distinguish between immigrants from Turkey, Morocco, the former Netherlands Antilles, Suriname, Indonesia, and South Africa. As such, we are able to examine whether similar explanations for Dutch national day participation hold for different ethnic groups with varying migration histories.

To account for immigrants’ previous familiarity with national days, we begin by focusing on the different historical connections between the host country and the various countries of origin, and distinguish between countries with and without colonial ties with the Netherlands. Second, we look at immigrants’ previous war experiences. Wars figure prominently in national commemorations and celebrations worldwide, in particular the Second World War (Liu et al. 2005; McCrone and McPherson 2009). Hence, having a personal connection to this war – either directly or indirectly via family members – makes it more likely that an immigrant has already participated in activities similar to those organised on Dutch Remembrance Day and Liberation Day.

Finally, we look at immigrants’ participation in holidays specific to their country of origin. Although such days may be unrelated to the content of Remembrance Day and Liberation Day, participation in them implies socialisation with commemorations and celebrations in general. We argue that all three aspects result to some extent in familiarity with Dutch national days. Using unique individual-level information, this study is the first to examine empirically how different types of previous participatory experiences among immigrants relate to host national day participation.

Remembrance Day and Liberation Day in the Netherlands

On Dutch Remembrance Day, held every year on the 4th of May, Dutch society commemorates civilians and members of the armed forces of the Kingdom of the Netherlands who died in a war or on a peace-keeping mission since the outbreak of the Second World War. Although originally initiated in 1945 to commemorate the Dutch victims of the Second World War, this was changed in 1961 to include also more recent casualties of war (Vermolen 1995). The main event of the day is the two minutes of silence held at 8:00 p.m. Commemoration ceremonies are organised throughout the country, the largest one taking place in the capital. There is also plenty of opportunity to participate in a more private matter, as the main events are broadcast live on national radio and television. Traditionally, people that own a flag will have their flags flown at half-staff, honouring the victims of war.

On Dutch Liberation Day, which falls on the 5th of May, the nation celebrates its liberation from German Nazi occupation (1940–1945), and draws attention to current issues related to freedom (or its absence) worldwide (Vermolen 1995). On Liberation Day, citizens are invited to raise their flags and festivities are organised throughout the whole country. The day starts with an address on the fragility of freedom, functioning as a link between the May 4 commemorations and the May 5 festivities. In the afternoon, liberation festivals take place in the 12 Dutch provinces and 2 major cities. Many of the activities are broadcast live, so that people can follow the activities via television, radio, or online.

Theory

Previous familiarity with national days

Whilst much of the research on immigrants focuses on socialisation processes in the host country, i.e. the learning of norms, values, and skills necessary to function in society, experiences in the immigrant’s country of origin can also be considered relevant for immigrant participation (Quintelier 2009; Voicu and Şerban 2012). This may be particularly true for Dutch national day participation owing to the historical connections between the Netherlands and some countries of origin of the largest immigrant groups. Suriname, the former Netherlands Antilles, and Indonesia (the former Dutch East Indies) are all former Dutch colonies (Castles and Miller 2003). It has been argued that these countries have more in common with the Netherlands than more recent immigration countries such as Turkey or Morocco (Hagendoorn, Veenman, and Volleberg 2003), because of the – forced – introduction of Dutch institutions and educational curricula in these countries. Hence, immigrants originating from former colonies are more likely to be familiar with the Dutch culture, and even with specific Dutch national days, such as Remembrance Day and Liberation Day. We therefore expect that immigrants from former Dutch colonies will participate more frequently in Dutch national days than immigrants from other countries (Hypothesis 1a).

At the same time, also when comparing countries that do share a colonial past differences can be expected, one important reason being the timing of their independence from the Kingdom of the Netherlands. Indonesian independence was formally recognised by the Netherlands in 1949, four years after Indonesia’s declaration of independence. Suriname was part of the Kingdom of the Netherlands until 1975, and still retains close political, economic, and cultural relationships with its former coloniser. The former Netherlands Antilles were dissolved in 2010. Curaçao and St Maarten became independent countries within the Kingdom of the Netherlands, whereas Bonaire, St Eustatius, and Saba were granted a status similar to Dutch municipalities. Antillean immigrants therefore still have the Dutch nationality (SCP 2014). Hence, immigrants from more recent former Dutch colonies can be expected to participate more frequently in Dutch national days than immigrants from countries with less recent colonial ties (Hypothesis 1b).

Moreover, since we are looking specifically at national days that celebrate freedom and commemorate victims of war, we argue that immigrants who have experienced war in some manner (either directly or indirectly via family and friends) will be more motivated to participate in the organised commemorative and celebrative activities than those without any personal connection to war. This can be explained by what is often referred to as ‘mnemonic socialisation’ (Zerubavel 1996): having parents or grandparents who have experienced a war will result not only in more knowledge of the topic, but also in more familiarity with the ‘appropriate’ ways to commemorate the event, including participating in nationally organised commemorations and celebrations. This is especially true of those who have experienced the Second World War (either directly or indirectly), as this war is the main focus of the activities organised on Dutch Remembrance Day and Liberation Day (Vermolen 1995). We therefore expect that immigrants who have a personal connection to the Second World War will participate more frequently in Dutch national days (Hypothesis 2).

Finally, familiarity with other holidays, unrelated to the content of Remembrance Day and Liberation Day, may also play a role in explaining immigrant participation. Based upon the assumption that previous experiences in the country of origin can be transferred, adapted, and used once immigrants arrive in their country of destination, previous studies have argued that immigrants from countries that have a civic or political environment similar to the host country integrate more easily than immigrants from countries with a very different environment (Voicu and Comşa 2014; Voicu and Rusu 2012; White, Nevitte, Blais, Gidengil and Fournier 2008). This is explained by the acquisition of civic or political skills that are more compatible with those needed in the country of destination. A similar argument can be expected to hold for national day participation. Immigrants’ participation in holidays specific to their country of origin – either because they used to live there or because their parents taught them – implies socialisation with commemorating and celebrating in general, resulting in more familiarity with participation in commemorations and celebrations. As a result, these immigrants may be more inclined to also participate in national days organised by the host country, even though the actual content may very well differ. We thus expect that immigrants who participate more frequently in holidays specific to their country of origin will also participate in Dutch national days more often (Hypothesis 3).

Following our earlier argumentation on differences in national day participation based upon immigrants’ country of origin, also the association between holidays specific to immigrants’ country of origin and host national day participation can be expected to vary depending on the country of origin under examination. The former Netherlands Antilles, for instance, have probably the highest number of national days that resemble those in the Netherlands. In addition, they have numerous ‘flag days’, i.e. days that recognise the discovery of the different islands. At the same time, no national war commemorations or celebrations are found here, except those related to the abolition of slavery. This is similar for Indonesia and Suriname, where in addition to days of independence, a lot of religious days can be found. Morocco and Turkey do know national war commemorations. In Turkey, these are to a great extent focused on the exploits of Ataturk during and after the First World War. In Morocco, several national days are dedicated to the return of territory. An additional goal of the present study is therefore to explore whether the relationship as formulated in Hypothesis 3 is dependent upon country of origin.

Exposure to national days in the host country

Although concepts of linear processes of assimilation (Gordon 1964; Park and Burgess 1921) have been criticised as being too simple (Alba and Nee 1997), the general idea underpinning socialisation theory is that immigrants who migrated at an earlier age adapt more easily to their new environment than immigrants who migrated at a later age. In addition to the age at migration, also the time spent in the host society is considered a key factor for immigrant participation (White et al. 2008). We expect this to hold for national day participation as well. Immigrants who migrated to the Netherlands before the age of 12 (i.e. the age at which children in the Netherlands start secondary school) have had classes in Dutch history, and have therefore had the chance to learn about the Dutch past, including the Second World War. Moreover, through education they have had the opportunity to learn about the activities involved in Dutch Remembrance Day and Liberation Day. For example, many schools in the Netherlands have ‘adopted’ war monuments, and organise daytrips to attend commemoration ceremonies. We therefore expect that immigrants born in the Netherlands, or who migrated to the Netherlands before the age of twelve, will participate in Dutch national days more frequently than immigrants who migrated at a later age (Hypothesis 4a). Moreover, we expect the frequency of national day participation to increase with length of stay in the Netherlands (Hypothesis 4b).

Research has identified two other immigrant characteristics that are closely linked to immigrants’ participation in the host society. First, proficiency in the host country language is thought to be crucial to encourage familiarity with the host country culture (Huijnk, Verkuyten, and Coenders 2012). Knowledge of the host country language not only enables immigrants to acquire practical information on how and where to participate in organised events, but also allows them to understand why people participate. Abundant research supports this line of reasoning, indicating that host country language is positively associated with all kinds of civic and political activities, ranging from membership of voluntary associations to voting behaviour (Aleksynska 2011; Torney-Purta, Barber, and Wilkenfeld 2007; Voicu and Rusu 2012). We expect host country language use to also play a role in Dutch national day participation. Not only is all the practical information on the activities organised on these days provided mainly in Dutch, but it is also easier for immigrants to communicate with others about the actual content of these days if they know the Dutch language. We therefore expect that immigrants who more often use the Dutch language will participate more frequently in Dutch national days than immigrants who less often use the Dutch language (Hypothesis 5).

A second way to become more familiar with the host country’s culture is via one’s social network. Social capital, such as the size and strength of one’s social network, has long been considered one of the main ways to gather information (Granovetter 1973; Putnam 2000). Moreover, social networks can function as ‘recruitment networks’, through which people are invited to become a member of a civic association or participate in political activities (De Rooij 2012). Especially, social contacts with native citizens are thought to play an important role (Heath, Rothon, and Kilpi 2008). Having more native contacts in one’s network has indeed been found to be associated with more frequent participation in political activities (Quintelier 2009). Following this line of argumentation, contact with native Dutch citizens – who are more likely to be familiar with Dutch history and its commemoration – will most likely also increase immigrants’ familiarity with Dutch commemorations and celebrations. Hence, we expect that immigrants with a larger number of native Dutch contacts will participate more frequently in Dutch national days than immigrants with less native Dutch contacts (Hypothesis 6).

Method

Sample

This paper makes use of data taken from the LISS (Longitudinal Internet Studies for the Social sciences) migrant panel administered by CentERdata (Tilburg University, Netherlands). The migrant sample was drawn from the population register by Statistics Netherlands and was stratified by ethnic groups and weighted by household size. More information about the LISS panel sampling procedure can be found at www.lissdata.nl. Panel members who could not participate otherwise were provided with a computer and Internet connection. The sample included the four major non-western immigrant groups in the Netherlands, namely persons of Moroccan, Turkish, Surinamese, and Antillean origin. Additionally, persons of Indonesian and South African origin were included, as well as persons of not further specified western European, western non-European, and non-western origin, and a control group of persons of Dutch origin.

Data on the main variables of interest were collected in the ‘Freedom and Liberation Day in the Netherlands’ survey in April 2014. In total, 1705 household members were approached, of which 78% responded, resulting in a sample of 1325 respondents in 958 households. Given our interest in country of origin, respondents without information on their home country were deleted (N = 25). Of the 1300 respondents in total, 5.08% had a Turkish background, 5.77% a Moroccan background, 4.46% was from the former Netherlands Antilles, 4.77% from Suriname, 8.92% from Indonesia, 4.39% from South Africa, 26.23% had another western background (not further specified), 7.15% another non-western background, and 33.23% a native Dutch background. Whilst we do present comparisons with native Dutch respondents in our descriptive results, only those from a non-native Dutch background were retained for our explanatory analyses (N = 868). Due to the oversampling of respondents from a non-western background, non-western respondents were over-represented compared to the Dutch population, where only 12.06% is from non-western origin (Statistics Netherlands 2015)¹.

Measures

Dependent and independent variables

Host national day participation was measured with respect to two different national days. We included activities that the literature on political participation might consider ‘low cost’ activities, as well as ‘high cost’ activities (De Rooij 2012; Voicu and Comşa 2014). To measure participation in Remembrance Day, we asked respondents:

How often in the past five years, on the 4^th of May (Remembrance Day), did you: (a) fly a Dutch flag at half-staff; (b) observe two minutes of silence; (c) attend a memorial event; (d) follow Remembrance Day proceedings on television, radio or online?

To measure participation in Liberation Day, respondents were asked:

How often in the past five years, on the 5th of May (Liberation Day), did you: (a) hang out a Dutch flag; (b) visit a Liberation Festival; (c) follow Liberation Day proceedings on television, radio or online?

Response categories were: (1) ‘never’, (2) ‘once’, (3) ‘twice’, (4) ‘three times’, (5) ‘four times’, and (6) ‘every year’. After recoding the response categories to range from 0 to 5, two mean scores were created: one for Remembrance Day and one for Liberation Day.

Ethnic origin was measured using the country of birth of the respondent and his or her parents. When either the respondent or at least one parent was born abroad respondents were classified as having an immigrant background. This is a commonly used definition in the Netherlands, based on Statistics Netherlands (www.cbs.nl). A distinction was made between respondents from Turkish, Moroccan, Antillean (i.e. former Netherlands Antilles), Surinamese, Indonesian, South African, other western, and other non-western origin². This distinction was then used to examine potential differences in Dutch national day participation between respondents with and without a colonial past.

Personal connection to the Second World War was measured by asking respondents whether they had experienced a war or whether they knew people who had personally experienced a war. Respondents were then asked: ‘Which war did this person experience?’. Given the emphasis of Dutch Remembrance Day and Liberation Day on the Second World War (and its Dutch victims) (Vermolen 1995), only experiences related to the Second World War were taken into account. A dummy variable was created to distinguish between respondents with and without a personal connection to the Second World War. Both direct and indirect experiences within the family environment were considered a personal connection, whilst the experiences of non-relatives were not counted as such.

Participation in holidays specific to the country of origin was measured by asking respondents: ‘Do you ever celebrate national holidays of your country of birth / your parents’ country of birth?’ Respondents could answer ‘yes’ or ‘no’. In a follow-up question, respondents were asked which national holiday they participated in (if ‘yes’). Responses indicating a Dutch holiday were set to missing. A dummy variable was created to distinguish between respondents who did and did not participate in holidays specific to their country of origin.

Age at migration was calculated by subtracting the year of birth from the year the respondent first came to live in the Netherlands. A distinction was made between respondents who indicated that they had lived in the Netherlands all their lives, respondents who moved to the Netherlands before they were 12 years old, and respondents who moved to the Netherlands at the age of 12 or older. The latter group functioned as the reference category.

Length of stay was operationalised as a continuous variable, and calculated by subtracting the age at migration from the age during the time of the interview.

Dutch language use was measured as a combination of the items: ‘Do you speak Dutch with your partner?’, ‘Do you speak Dutch to your child(ren)?’, ‘Do you speak Dutch with your father?’, and ‘Do you speak Dutch with your mother?’ Answer categories comprised: (1) ‘no, never’; (2) ‘yes, sometimes’; (3) ‘yes, often’; and (4) ‘yes, always’. For respondents who did not have a partner, child, father, or mother, responses were set to missing. All valid answers were recoded to range from 0 to 3, and an average score was created based upon the answers available.

The number of native Dutch contacts in the respondents’ social network was based on the five most important persons in their lives (i.e. those with whom they discussed important matters) over the past six months. Respondents were given one point for every native Dutch contact that was mentioned, provided the contact was not one of their parents, siblings, children, or other family members. This sum score was then divided by the total number of non-family members mentioned by the respondent and multiplied by 100, resulting in the percentage of native Dutch contacts in his or her (close) social network.

Control variables

First, considering the close relationship between national day participation and feelings of national belonging found in earlier studies (Coopmans et al. 2015), we controlled for national belonging by asking respondents to what extent they agreed or disagreed with the following four items: ‘Being Dutch is an important part of who I am’; ‘I feel connected to other Dutch people’; ‘Whenever I talk about Dutch people I often say “We”’; and ‘I am proud to be Dutch’. Answer categories ranged from (1) ‘totally disagree’ to (5) ‘totally agree’. After recoding the answer categories to range from 0 to 4, a mean score was created by averaging the scores of the four items (Cronbach’s alpha = .86).

Second, immigrants’ motivation to participate in the host society has been found to also depend upon resources, potential time constraints (De Rooij 2012; Putnam 2000; Voicu and Şerban 2012), and the reaction of the host society (De Vroome, Verkuyten, and Martinovic 2014; Maxwell 2009). As we expect these constructs to be similarly important for participation in national days, we controlled for level of education, employment status, and perceived level of discrimination. Educational level consisted of seven categories: (0) ‘not yet started any education’; (1) ‘primary school’; (2) ‘intermediate secondary school’; (3) ‘higher secondary education’; (4) ‘intermediate vocational education’; (5) ‘higher vocational education’; and (6) ‘university’. Employment status consisted of three categories: ‘employed’, ‘unemployed’, and ‘other’ (including students, retired, and disabled people). Employed respondents acted as the reference category. Discrimination was measured by asking respondents: ‘In the past 12 months, have you been discriminated against, for instance because of your religious beliefs, sexual orientation, appearance, or age?’. Response categories were: (1) ‘no’; (2) ‘yes, sometimes’; and (3) ‘yes, often’. A dummy variable was created for respondents who reported being discriminated against either sometimes (21.10%) or often (3.40%).

Finally, following previous studies on immigrants’ participation (De Rooij 2012), also gender was taken into account, operationalised as a dummy variable for female. An overview of the descriptive statistics for the dependent, independent, and control variables amongst the immigrant sample can be found in Table 1. Appendix A contains the descriptive statistics per ethnic origin, including native Dutch (see Supplemental data section).

Table 1. Descriptive statistics (N = 868).

Variables	Min.	Max.	Mean/proportion	SD	Valid N
National day participation
Remembrance Day	0	5	2.26	1.23	855
Liberation Day	0	5	1.68	1.25	857
Ethnic origin
Turkish	0	1	0.08		868
Moroccan	0	1	0.09		868
Antillean	0	1	0.07		868
Surinamese	0	1	0.07		868
Indonesian	0	1	0.13		868
South African	0	1	0.06		868
Other western	0	1	0.39		868
Other non-western	0	1	0.11		868
Second World War	0	1	0.54		853
Holidays country of origin	0	1	0.39		738
Age at migration
>12 years old	0	1	0.34		867
<12 years old	0	1	0.17		867
Born in the Netherlands	0	1	0.49		867
Length of stay	4	88	37.34	18.22	866
Dutch language use	0	3	2.12	1.02	839
Native Dutch contacts (%)	0	100	42.77	46.03	868
National belonging	0	4	2.44	.82	520
Level of education	0	5	2.72	1.59	855
Employment status
Employed	0	1	0.50		868
Unemployed	0	1	0.07		868
Other	0	1	0.43		868
Discrimination	0	1	0.25		853
Gender (female)	0	1	0.55		868

Analytic strategy

Multivariate structural equation modelling with bootstrapping was applied using Stata, version 13 (StataCorp 2013). As we are dealing with clustered data (i.e. 868 individuals within 748 households), we took the non-independence of observations into account by computing standard errors using the generalised Huber/White/sandwich estimator, which allows for correlations between errors within clusters (Rogers 1993; Williams 2000). Moreover, using the full-information maximum likelihood (FIML) missing data estimation approach, we were able to include observations with missing values in our analyses. FIML uses all observed variables in the model to estimate the means and covariances of item nonresponse, and outperforms listwise deletion and simpler substitution methods (Cheung 2015; Enders and Bandalos 2001).

Results

Descriptive results

An analysis of variance (ANOVA) demonstrated significant ethnic group differences in both the level of participation in Remembrance Day (F(8, 1274) = 13.36, p < .001, adjusted R² = . 072) and Liberation Day (F(8, 1276) = 6.79, p < .001, adjusted R² = .036). Table 2 gives mean levels of national day participation per ethnic group. Native Dutch respondents participated most frequently in national commemorations and celebrations, closely followed by respondents from an Indonesian background, respondents with another western background, and respondents from a Surinamese background. Tukey–Kramer pairwise comparisons furthermore showed that these respondents participated significantly more often in Remembrance Day than respondents from a Turkish or Moroccan background. Respondents from a Moroccan background also participated significantly less frequently in Remembrance Day than respondents from a South African or other non-western background. Ethnic group differences in the participation rates for Liberation Day were less remarkable. Respondents from a Turkish and Moroccan background reported significantly lower participation rates than native Dutch respondents, as well as respondents from an Indonesian or other western background.

Table 2. Mean comparisons of national day participation between ethnic groups (N = 1300).

	May 4 (Remembrance Day)			May 5 (Liberation Day)
	Mean	SD	Group comparison	Mean	SD	Group comparison
1. Native Dutch	2.67	1.14	2, 3, 4, 9	2.06	1.34	2, 3
2. Turkish	1.66	1.22	1, 5, 6, 8	1.17	1.04	1, 6, 8
3. Moroccan	1.46	1.16	1, 5, 6, 7, 8, 9	1.20	1.10	1, 6, 8
4. Antillean	2.07	1.40	1	1.53	1.22	–
5. Surinamese	2.32	1.07	2, 3	1.84	1.22	–
6. Indonesian	2.54	1.20	2, 3	1.87	1.33	2, 3
7. South African	2.28	1.16	3	1.64	1.20	–
8. Other western	2.48	1.15	2, 3	1.84	1.22	2, 3
9. Other non-western	2.24	1.26	1, 3	1.65	1.34	–

Note: Group comparisons show those groups that significantly differ at the 5% level, based upon a Tukey–Kramer pairwise comparison test.

Explanatory results

Table 3 displays the results of the multivariate structural equation model for Remembrance Day and Liberation Day, in which we controlled for national belonging, level of education, employment status, perceived discrimination, and gender. As we were interested in predictors of national day participation amongst citizens from a non-native Dutch background, native Dutch respondents were excluded from these analyses. Since our descriptive analyses indicated respondents from an Indonesian background to score highest on national day participation, they acted as the reference category. In total, our model explained 34% of the variance in national day participation (R² (Remembrance Day) = .246; R² (Liberation Day) = .156). Additional stepwise analyses showed that 16% was explained by our measures of previous participatory experiences (R² (Remembrance Day) = .121; R² (Liberation Day) = .052), 17% by our socio-cultural indicators (R² (Remembrance Day) = .114; R² (Liberation Day) = .071), and 23% by our control variables (R² (Remembrance Day) = .155; R² (Liberation Day) = .107).

Table 3. Models for participation in Remembrance Day and Liberation Day (N = 868).

	Remembrance Day		Liberation Day
	B	SE	B	SE
Intercept	.339	.308	.308	.329
Ethnic origin (ref. Indonesian)
Turkish	−.091	.254	−.059	.231
Moroccan	−.341	.260	−.165	.249
Antillean	.117	.272	.092	.255
Surinam	.091	.266	.210	.284
South African	−.000	.289	−.030	.291
Other western	.109	.181	.124	.185
Other non-western	.211	.226	.164	.231
Second World War	.185†	.099	.109	.108
Holidays country of origin	.456***	.104	.280**	.105
Age at migration (ref. >12 years)
Migrated <12 years	−.237†	.132	−.324*	.144
Born in the Netherlands	−.295**	.100	−.445***	.119
Length of stay	.011***	.003	.009**	.003
Dutch language use	.163**	.063	.167**	.064
Native Dutch contacts	−.000	.001	−.002	.001
National belonging	.418***	.085	.376***	.092
Level of education	−.031	.030	−.066*	.032
Employment status (ref. employed)
Unemployed	−.120	.172	−.139	.181
Other	.061	.088	.088	.090
Discrimination	−.052	.104	−.046	.104
Gender (female)	.194*	.077	.060	.082

Note: Two-tailed p values are reported.

†p < .10.

*p < .05.

**p < .01.

***p < .001.

As can be seen in Table 3, the initial ethnic group differences in national day participation identified in our descriptive analyses were no longer present once controlled for the variables included in our explanatory analysis. Hypotheses 1a and 1b, in which we expected that immigrants from former Dutch colonies – in particular the more recent ones – would participate more frequently in national days than immigrants from other countries, could therefore not be confirmed. These results suggest that the model presented in Table 3 fully explains the lower levels of national day participation found among respondents with a Turkish and Moroccan background. Additional stepwise analyses revealed that whilst the initial lower participation rates of respondents with a Turkish background lost significance after the addition of our ‘experience with the Second World War’ variable, the initial significant difference for respondents with a Moroccan background disappeared after the inclusion of our ‘age at migration’ and ‘length of stay’ variables.

Contrary to what was expected, no significant differences in national day participation were found between respondents with and without a personal connection to the Second World War. Respondents who had experienced the Second World War – either directly or indirectly via family members – did not participate more often in Liberation Day than respondents who did not have any personal connection. For Remembrance Day, a marginally significant effect was found. No support was therefore found for Hypothesis 2, on the role of previous war experiences. Additional stepwise analyses indicated that an initial significant effect of personal experiences with the Second World War on participation in Remembrance Day disappeared after the addition of ‘participation in holidays specific to the country of origin’.

In line with our expectations, respondents who reported to participate in holidays specific to their country of origin participated in both Remembrance Day and Liberation Day significantly more often than respondents who did not participate in any such activities. This result confirms Hypothesis 3, on the importance of previous familiarity with commemorating and celebrating. No differences were found in the effect of participation in holidays specific to immigrants’ country of origin on host national day participation when comparing respondents from different ethnic origins. The results of this additional analysis can be found in Appendix B (see Supplemental data section).

As for our indicators of exposure to national days in the host country, it was found that immigrants born in the Netherlands participated less often in national days than immigrants who migrated after the age of 12. The difference between immigrants who migrated before the age of 12 (but were not born in the Netherlands) and those who migrated after the age of 12 proved only significant for Liberation Day. Similar results were found when including age at migration as a continuous variable in our model. These findings reject Hypothesis 4a, on the role of age at migration. Interestingly, additional analyses showed that the association found between age at migration and national day participation only became positive after the addition of our socio-cultural indicators. Moreover, in line with Hypothesis 4b, a significant positive association was found between length of stay in the Netherlands and national day participation.

Our results furthermore revealed a significant positive association between native language use and national day participation. In line with Hypothesis 5, respondents who, on average, spoke more Dutch at home were found to participate more frequently in both Remembrance Day and Liberation Day. In contrast, a higher number of native Dutch contacts in one’s social network was not associated with more frequent national day participation, refuting Hypothesis 6.

Finally, as for our control variables, a positive association was found between national belonging and national day participation. Furthermore, female respondents participated more frequently in Remembrance Day than male respondents, and respondents with a higher level of education participated more often in Liberation Day than respondents with a lower level of education.

Discussion

Using unique information on national day participation drawn from a large online immigrant panel, the current study examined to what extent national day participation among immigrants is associated with previous participatory experiences related to the country of origin, alongside socio-cultural aspects related to the current country of residence. We concentrated on two prominent national days in the Netherlands, Remembrance Day and Liberation Day. In doing so, we contribute to more general literature on immigrants’ participation by introducing what has been a little-researched topic until now, namely that of national day participation. Moreover, we build on a recent line of research that emphasises the importance of the country of origin’s participatory context when examining immigrants’ patterns of participation in their host society.

To examine familiarity with national days, we focused on three factors that we believe are potential indicators of previous participatory experiences related to immigrants’ country of origin. First of all, we distinguished between countries with and without a colonial past. Second, given that the national days under examination focus mainly on commemorating the Second World War, we distinguished between immigrants with and without a personal connection to this war. Third, we took participation in holidays specific to the immigrants’ country of origin into account. Contrary to our expectations, immigrants from former Dutch colonies were not found to participate in national days more frequently than other immigrant groups once we included other aspects in our explanatory models. Instead, previous familiarity with commemorating and celebrating largely explained the variation in Dutch national day participation among immigrants: immigrants who participated in holidays specific to their country of origin participated in the activities organised on Remembrance Day and Liberation Day more frequently than other immigrants. A personal connection with the actual content of the national days, in this case the Second World War, played a less big role.

This is an important finding, as it implies that participation in activities related to country of origin in no way undermines participation in activities organised by the host country. Instead, our results indicate that participating in country of origin specific holidays contributes to host country national day participation. This effect was found regardless of the country of origin in question. These results are in line with recent literature on immigrants’ political participation showing that a highly participatory previous context promotes involvement in political activities, even more so when immigrants are still involved in political activities in their country of origin (Voicu and Comşa 2014; Voicu and Rusu 2012). We believe further research should therefore focus more explicitly on the type of holiday that immigrants participate in, as well as their motivation for participating. Another factor that merits more attention concerns the different activities organised on host country national days, especially in view of the smaller effects found for Liberation Day than Remembrance Day. This may be because in the Netherlands, Liberation Day is only an official public holiday once every five years, but it may also be because it is easier to link national commemorations (as opposed to national celebrations) to one’s personal memories of an event – even in a different (national) context.

In line with previous studies (De Rooij 2012), our results furthermore showed that native language use plays a more important role in immigrants’ national day participation than economic aspects such as employment or education. Immigrants who more often use the Dutch language at home were found to participate more frequently in the activities organised on Dutch national days. Also the time spent in the Netherlands was positively related to national day participation. This is in line with classic assimilation theories (Gordon 1964; Park and Burgess 1921). National day participation thus seems to follow a pattern similar to other forms of participation, such as membership of civic associations or voting behaviour (Aleksynska 2011; Torney-Purta et al. 2007; Voicu and Rusu 2012).

Surprisingly, a higher number of native Dutch contacts did not improve immigrants’ host national day participation. A possible explanation concerns the content of this particular type of participation: as opposed to civic or political participation, participating in national celebrations or commemorations might be a more personal or private matter, and therefore less dependent upon one’s social environment. It could also be, however, that the measurement currently used, i.e. the five most important persons, is too narrow. We would advise future research to focus more on the role of weak ties, which are thought to play a more important role in information gathering (Granovetter 1973).

Since the design of the present study was cross-sectional, no strong causal claims can be made. In addition to the need for longitudinal research on national day participation among immigrants, this study has several other limitations. For one, we focus on a very specific type of participation, namely participation in national commemorations and celebrations related to the Second World War. Second, the focus of the Remembrance Day and Liberation Day activities changes each year. Stories of immigrants are more central to the overall theme in some years than in others, and this too may affect immigrants’ experiences of these days. These limitations make it difficult to generalise our findings to national days in other countries. Future research would therefore benefit from studying the determinants of a broader spectrum of national days across different countries.

In sum, our study shows that immigrants’ host country national day participation is determined largely by previous familiarity with commemorating and celebrating. An important predictor is participation in holidays specific to immigrants’ country of origin. Policies aimed at increasing host national day participation among immigrants could therefore profit from focusing more explicitly on the link between holidays in the host country and the country of origin. Moreover, these findings highlight the need to emphasise the role of previous participatory experiences when examining immigrants’ current patterns of participation in the host society.

Acknowledgements

The LISS migrant panel data were collected by CentERdata (Tilburg University, the Netherlands) through its MESS (Measurement and Experimentation in the Social Sciences) project, funded by the Netherlands Organization for Scientific Research.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Of the 12.06%, 2.35% Dutch citizens from non-western origin has a Turkish background, 2.25% a Moroccan background, 0.88% is from the former Netherlands Antilles, 2.06% is from Suriname, and 4.52% has another non-western background (not further specified) (Statistics Netherlands 2015).

2. Due to their socio-economic and cultural position, panel members from an Indonesian background – mainly people born in the former Dutch East Indies to (a) native Dutch parent(s) – were considered ‘western’ immigrants. Furthermore, it was assumed that most panel members from a South African background belong to the white, Afrikaans-speaking group.