The mobilizing effects of political media consumption among external voters

This study examines political media consumption among non-resident citizens, and whether following politics in traditional and social media in their country of residence and origin has a mobilizing effect on voting in origin-country elections. The topic of our study is inspired by the trend towards increased enfranchisement of external citizens, improved methods for participation from abroad, and the transformation of the media landscape with enhanced possibilities for external voters to follow politics in their country of origin. Based on a survey directed towards a stratified random sample of Finnish external citizens in 15 countries, we find that politically oriented media consumption in the country of origin substantially increases the likelihood of participating in origin-country elections and that this effect holds for traditional media channels as well as for social media. A corresponding mobilizing effect is, however, not found for following politics in the country of residence. This demonstrates the value of relevant information for political participation.


Introduction
This study examines how political media consumption, i.e. the use of different media for the acquisition of political information, affects the propensity among non-resident citizens to participate in origin-country elections. Voting from abroad has existed since the early twentieth century, but for a long time the right to vote was restricted to military and diplomatic personnel stationed abroad. However, as voting rights have expanded and access to electoral participation from abroad has been improved (e.g. through postal voting), the issue has become increasingly relevant (Lafleur 2013;Arrighi and Bauböck 2017;Peltoniemi 2018). Over the same period, the media landscape and communication technology have undergone a radical transformation, improving the possibilities for external voters to follow politics in their country of origin. With the proliferation of satellite television, and later Internet-based communication (e.g. smartphones, social media, streaming services), external voters have gained access to permanent and instantaneous sources of information of origin-country news (Christiansen 2004;Lafleur 2013). From media, citizens get information about topical political issues and about party and candidate stances on those issues; information that is important for casting an informed vote. While there is abundant research demonstrating that media consumption is key to political mobilization in domestic contexts (see Strömbäck 2017), corresponding research focusing on external voters is more limited in scope (Lafleur and Sánchez-Domínguez 2015;Gherghina and Tseng 2016).
Hence, we want to examine how different forms of political media consumption affect the propensity to vote in origin-country elections. More specifically, we study if consumption of political news among external citizens has a mobilizing effect in line with what previous research has found for domestic politics and to what extent it matters if you consume political news from the country of origin and the country of residence. Furthermore, we explore the extent to which the potential mobilizing effect of the following politics is homogenous across different media formats (traditional or social media) used by the respondents.
We rely on a survey of Finnish external citizens (n = 1974) based on a stratified random sample of 4,800 Swedish-speaking Finnish non-resident citizens in 15 countries. The sampling method is worth highlighting, since surveys of external citizens generally are limited to convenience samples and a small number of countries (Ahmadov and Sasse 2016). Moreover, due to its open-list proportional electoral system, in which voters need to identify a single candidate to cast their vote for (from a large selection of contestants), Finnish parliamentary elections present a challenging context for external voters to navigate. Hence Finnish external citizens can be considered a most-likely case for finding an effect of politically oriented media consumption.
The findings from our study show that following politics in the country of origin substantially increases the likelihood of participating in origin-country elections, an effect found for both traditional media channels and for social media. A corresponding pattern is not found for following politics in their country of residence. This indicates that political mobilization is not driven by the attention paid to politics in general, but rather that the mobilizing effect of political media consumption is specifically related to the source and content of the media in question. The findings appear very robust and hold under control for a variety of factors known to influence participation in elections.

Information, media consumption, and voting in elections
Information is key to political mobilization and participation in elections. Information helps voters to discover politically relevant issues and problems and to identify their personal political preferences and interests. Access to information allows citizens to match their preferences with those of a political party or a candidate and make well-grounded decisions about which political alternative to vote for (Carmines and Stimson 1980). Hence, it is not surprising that more knowledgeable people tend to participate in elections more actively than those who are less knowledgeable (Delli Carpini and Keeter 1996).
Information on political issues is generally gathered from the news media, and research has convincingly established a positive link between media usage and various forms of political participation. Such a positive link is found for turnout in elections (Prior 2005), as well as for other forms of political participation (Liu et al. 2013). Media usage, and, in particular, politically oriented news consumption, seems to function as a positive spiral, with increased usage stimulating interest (Boulianne 2011;Strömbäck and Shehata 2010), political efficacy (Aarts and Semetko 2003;Hansen and Pedersen 2014), as well as political trust (Strömbäck et al. 2016), which, in turn, furthers political participation in general, including turnout in elections.
Today, through digital and traditional news media, voters have access to more options in terms of the media environment (Shehata and Strömbäck et al. 2021), but also to more information than ever before. There is an increased number of channels for news consumption, and these channels are easily available to use through various platforms. This has been described as a change from a low-choice to a high-choice media environment (Strömbäck 2017). A greater supply does, however, not necessarily imply higher consumption of political news. Studies point rather towards a greater differentiation across groups of citizens and a growing share of people who hardly follow the news at all (Edgerly et al. 2018). The increasingly fragmented media environment also suggests that politically uninterested people are less likely to be incidentally exposed to news, which points towards even stronger differences in political engagement and activity among high and low consumers of political news. The positive mobilizing effects of media usage found for traditional media channels have, however, also been identified for digital social media (Boulianne 2015;de Zuniga et al. 2012;Dimitrova et al. 2014).
Particularly relevant from the perspective of this study, is that the contemporary high-choice media environment (Strömbäck 2017) is highly transnational. With the changing media landscape, it follows that it has become easier for citizens to keep themselves informed about political affairs in countries in which they do not live. Indeed, one of the most important elements differentiating the present migration domain from that of other periods in history, is the vastly improved possibilities to keep in contact with the country of origin (Ros 2010). Not too long ago, the proliferation of satellite television and Digital Broadcasting Satellite (DBS) technology was considered revolutionary in the way it allowed emigrants to engage in transnational media consumption, by giving access to much of the same information that their co-nationals, still living in the country of origin, have access to (Christiansen 2004). Today in an age of widespread Internet penetration and mobile technology, emigrants have immediate access not only to their country of origin's television and radio stations but also to the newspapers and other media. Moreover, through social media, video conferencing, and Internet SMS services, non-resident citizens have the possibility to interact directly with residents from the country of origin, almost as if they would live there (Komito 2011;Metykova 2010).
That said, the growing differences in political media consumption identified within many countries (Edgerly et al. 2018) can be expected to be even more enhanced among non-resident citizens. Interest in following the country of origin's media is, hence, likely to vary significantly between groups of non-resident citizens (Lafleur 2013). They are also likely to be interested in transnational media consumption for different reasons. Some will consume media to stay informed about the social and political developments of their country of origin, and others will be more interested in keeping up with their favourite sports team (Metykova 2010).

External electoral participation
While there is a growing body of research examining the electoral behaviour of external citizens (e.g. Lafleur and Chelius 2011;Lafleur and Sánchez-Domínguez 2015;Ahmadov and Sasse 2016;Peltoniemi 2018;Ciornei and Østergaard-Nielsen 2020;Nemcôk and Peltoniemi 2021), it is still limited in scope, at least compared with research on the electoral behaviour of resident citizens.
Based on previous research, we know that turnout in elections tends to be quite low among external citizens (e.g. Lafleur and Chelius 2011;Peltoniemi 2018;Ciornei and Østergaard-Nielsen 2020). Distinct patterns that can explain variation in turnout are, nevertheless, more difficult to discern. There are, e.g. multiple studies showing that the propensity to vote decreases as the duration of the stay abroad grows longer (e.g. Ahmadov and Sasse 2016;Peltoniemi 2016;Chaudhary 2018;Finn 2020). A quite intuitive result, suggests that the connection with the political life of the country of origin weakens over time. However, there are also studies with contrasting findings. Making use of a survey of almost 900 Romanian emigrants in France, Germany, Italy, and Spain recruited via e.g. Facebook groups, Romanian associations, and organizations abroad, Gherghina and Tseng (2016) find that length of stay in the country of residence is positively related to turnout in elections held both in the country of residence and of origin. Moreover, drawing on a survey of Columbian expatriates in five cities in the USA and in Europe by Escobar et al. (2015) find no relationship between years spent abroad and turnout.
That research on the electoral behaviour of external citizens generates conflicting findings is not particularly surprising. Information about the non-resident population of a country is often limited, and the respondents are more difficult to reach than resident citizens. Hence, collecting reliable and representative data on external voters is more difficult than for resident voters. Moreover, the composition and experiences of emigrant populations vary greatly between countries. Some countries have a long experience of emigration and established diasporas, whereas large-scale emigration is a more recent phenomenon for other countries. Furthermore, on top of the usual institutional variations (e.g. electoral systems of the countries of origin), the right to vote and accessibility of external voting varies dramatically between external citizens from different countries (Lafleur 2013;Ellis et al. 2007). The theoretical literature, also, suggests that the relationship between country of origin and country of residence engagement can take on different forms, and that it is not necessarily a zero-sum relationship, i.e. increased engagement in one country leading to decreased engagement in the other (Tsuda 2012). Hence, drawing on previous research regarding external voting behaviour is arguably quite complicated.
With these challenges in mind, we take a closer look at the available research on political participation among external citizens, where our main explanatory variable of interest, political media consumption, has been included. Drawing on a survey conducted among 300 Bolivian emigrants in five cities in different countries, Lafleur and Sánchez-Domínguez (2015, p. 22) find that "In spite of the abundance of electronic media and the availability of home country television channels abroad, most external voters seem poorly informed about the election in which they participate." Consequently, they find no evidence of political media consumption on vote choice among emigrant Bolivians. In a study of Romanian emigrants in Western Europe, Gherghina and Tseng (2016) investigate the impact of media exposure on turnout in both the origin and residence country elections by using a measure assessing the use of the different types of media in the country of residence. They find a small but significant effect of their combined media measure on turnout in origin-country elections. While not technically examining the behaviour of non-resident citizens, Careja's and Emmenegger's (2012) study of return migrants in Central and Eastern European countries also suggests that media usage has a significant and consistent effect on political participation among migrants.
There are also some studies that examine the relationship between different types of media consumption and other forms of political behaviour. Chauvet et al. (2016) find in their study of Malian migrants in France that exposure to French media is correlated with a stronger interest in French political affairs, together with more scepticism towards Malian institutions and lower interest in Malian politics. In another study focusing on Romanian emigrants, Grad et al. (2020) show that those who read origin-country newspapers and follow origin-country TV stations are significantly more likely to try to influence the vote choice of their families in the country of origin, compared to their co-nationals who do not follow origin-country media.

Hypotheses
Based on the, admittedly somewhat limited, available evidence on the relationship between political media consumption on electoral participation for external citizens we have formulated four hypotheses. For our first hypothesis we mainly rely on the research on political media consumption and turnout among domestic voters. Abundant research on domestic voters has found political media consumption to be positively related to turnout (Prior 2005;Liu et al. 2013), and the available research on external voters appears to support this pattern (e.g. Gherghina and Tseng 2016;Grad et al. 2020). Hence, our first expectation is that, H1. There is a positive link between political media consumption and turnout among external voters.
What differentiates external from domestic voters is the transnational character of their media consumption, which is likely to include both country of residence and country of origin sources. However, previous research on external voters does not separate between the effects of origin-and residence-country political media consumption on electoral behaviour, nor do they examine the effects of different types of political media consumption on voting. Nevertheless, since information is considered to be an important prerequisite for political participation (e.g. Delli Carpini and Keeter 1996) and we are interested in turnout in country of origin elections, we assume that information provided by media in this context is particularly relevant. Media consumption in the country of residence is far less likely to provide voters with information of relevance for elections in the country of origin, and is therefore not expected to enhance political participation in home country elections. Hence, we posit the following two assumptions: H2. Origin-country political media consumption is positively related to turnout in origin-country elections. H3. Residence-country political media consumption is not positively related to turnout in origin-country elections.
Lastly, despite the importance that new communication technologies have played in connecting external voters with their country of origin and making information about the origin country more easily available, we expect the mobilizing effect to be largely the same across different media formats. For external voters, political media consumption, especially of information from the country of origin, we expect that availability of information is what matters the most. In other words, obtaining relevant information is more important than the format used to consume political news.
H4. The format of the media whereby external voters get information about politics does not matter for turnout.

Participating in Finnish elections from abroad
Institutional factors play a very important role for turnout when voting from abroad. Until recently, Finland relied on in-person voting as the method for voting from abroad. A vote was cast in a designated polling place (e.g. embassy) in advance (early voting). For the latest parliamentary election in 2019 there was a total of 219 advance polling stations in 86 countries available to Finnish external citizens (Ministry of Justice 2021). However, prior to the parliamentary elections in 2019, postal voting was also introduced as a method for voting from abroad in Finnish elections, thus significantly improving the possibilities to vote in Finnish national elections and levelling the field of electoral participation for Finnish non-resident citizens (Nemcôk and Peltoniemi 2021).
Despite the practice of voting becoming easier for external Finnish voters, the Finnish electoral system presents its own challenges to external voters. In fact, we think that the Finnish electoral system represents a context in which politically oriented news consumption is very likely to have a mobilizing effect on external voters. Finland has an open-list proportional electoral system with mandatory preferential voting. Votes are cast for individual candidates nominated by parties in 13 constituencies, and it is not possible to cast a vote for a collective party list. The total number of preference-votes cast for candidates nominated by a specific party determines how many seats that party wins. The electoral system has a very simple structure, with each voter casting a vote for a single candidate by writing the number of that candidate on the ballot. Despite its simplicity, the system offers many challenges to voters, related to information processing. Voters need to identify a single candidate out of a large selection 1 (Marcinkiewicz and Stegmaier 2015).
The task of identifying a single candidate out of a large number of nominees is likely to be even more challenging for voters who are living abroad since it involves a great deal of information processing. Since it is not possible to cast a vote for a collective party list, voters cannot simply rely on habitual behaviour and vote for a party they identify with. Rather they need updated information on the candidates running for their preferred party. Moreover, some of the relatively easily available cues that domestic voters apply to identify a suitable candidate, such as roots in the local community (Shugart et al. 2005), might be irrelevant for voters abroad. This leads us to conclude that Finland is a most-likely case for finding a mobilizing effect of political news consumption (Table 1).

Data and methods
The study relies on a survey of non-resident Finnish citizens conducted in the aftermath of the 2019 parliamentary elections in Finland. The invitation to participate in an online survey was distributed by mail to a disproportionate stratified random It should be noted that our survey was conducted among a sub-group of Finnish non-resident citizens, namely Swedish-speaking Finns. Swedish-speaking Finns, or Finland-Swedes, represent a linguistic minority in Finland, speaking one of the country's two official languages, Swedish. From an institutional perspective, being a Swedish-speaking Finn living abroad is no different than being Finnish-speaking. As an officially bilingual country, Finland provides its citizens with all necessary official information on voting practices and regulations in both Finnish and Swedish (Language act 2003). Moreover, the socio-demographic and socio-economic composition of the non-resident Swedish-speaking Finns are similar to non-resident Finns in general (Harjula and Himmelroos 2020). The biggest difference in political behaviour between Swedish and Finnish-speaking Finns is that a majority of the Swedish-speakers vote for an ethnic minority party, The SFP (Bengtsson and Grönlund 2005). The migration pattern of Swedish-speaking Finns, also, deviates somewhat from that of the Finnish-speaking Finns. The Swedish-speaking Finns are substantially more likely to emigrate than the Finnish-speaking Finns (Kepsu 2016), and although Sweden is the most common destination for both Swedish-speaking and Finnish-speaking Finns, Swedish speakers are more likely to migrate to Sweden (Statistics Finland 2019).

Dependent variable
Our dependent variable measures participation in the 2019 Finnish parliamentary election. The variable is derived from the survey question: "Many citizens living abroad do not vote in Finnish elections for one or other reason. Did you vote in the Finnish parliamentary election in April?" and the response options were "Yes" (1), "No" (0), "Don't know" (0). See Appendix B for more details and descriptive statistics for all included variables.

Independent variables and covariates
Our independent variables derive from a set of items on political media consumption asked both from the perspective of the country of origin and the perspective of the country of residence. The first set of items are based on the following question: "How often do you follow Finnish politics…", with four specifications (a) "… on television", (b) "… on the radio" (c), "… in the daily papers" and, (d) "… on social media (e.g. Facebook, Twitter, Snapchat, Instagram, etc.)". The answer options were "Every day," "Several times a week," "Once or twice a week," "Not so much," "Not at all," and "Don't know." For the residence-country perspective, the question was phrased as follows: "How often do you follow the politics of (name of the country of residence)…", with the same specifications and answer options.
We performed factor analysis using oblique rotation (promax) to single out patterns in the political news consumption that external citizens engage in (Table 1). The oblique rotation is applied to take into account that these practices are likely to be correlated. Using an eigenvalue cut-off of 1.0, we attained three factors that explain a total cumulative variance of 67.7%. The two first factors cover forms of political news consumption from three different traditional sources, TV, Radio, and Newspapers, distinctly divided into country of origin and of residence. This shows a distinct pattern in the extent to which external citizens follow politics in their country of origin and of residence, and we will use the factor scores derived from the analysis to analyse the extent to which these two types of political media consumption impact the likelihood of turning out to vote in origin-country elections. The third factor, in turn, covers following politics via social media and includes politics in both the country of origin and of residence. Since we theorize that it is important to distinguish between politics in these two contexts in order to identify the potential effect on voting in origin-country elections, we will, in the explanatory analysis, not include the third factor, but rather apply separate measures (using the single survey items) to analyse the mobilizing effect of following politics on social media for both the country of origin and of residence.
In our regression analyses, we also add multiple covariates to control for the robustness of the effects of our main variables of interest. Considering the well-established finding that voting is related to a person's social status, the models include basic sociodemographic variables such as age, gender, and education (e.g. Verba and Nie 1972;Wolfinger and Rosenstone 1980). Findings from a study of Swiss expatriate voters suggest that external voters rely more on socio-structural voting than domestic voters do (Goldberg and Lanz 2019). Lafleur and Sánchez-Domínguez (2015, p. 21) also find that different socio-economic and demographic characteristics of external voters are powerful predictors of emigrant behaviour and their political choices.
We further want to control for the influence of variables related to the experience of non-residency and ties to the country of origin (for more details on this relation see also Peltoniemi, Nemcok and Wass 2023in this special issue). As we discussed above, previous literature points towards findings that the time spent abroad affects the political behaviour of external voters by reducing the propensity to vote (Ahmadov and Sasse 2016;Peltoniemi 2016;Finn 2020). We also include a few variables that have been found relevant in other studies examining the political behaviour of external citizens: resident country language skills (e.g. Budria et al. 2017) and double citizenship (e.g. Himmelroos and Peltoniemi 2021). We further include a control for friends in the country of origin and for Finnish language skills since this might impact the extent to which our respondents of Swedish-speaking origin are able to follow the election campaign, where many debates and events are carried out in Finnish.
The last block of variables in our analysis is related to politics and political mobilization. The general level of political interest is likely to impact both the extent to which people follow politics via media and if they turn out to vote (Blais et al. 2002) and is, therefore, likely to dampen the effect of political media consumption on voting. Being contacted during the campaign or not is likely to have a separate mobilizing effect (Parry et al. 2008). And since voting has been shown to be a habitual practice (Dassonneville 2017), we also control for participation in national and local elections in residence-country elections. 3 Our binary regression analyses with robust clustered standard errors at the country level 4 comprise of three different models, with the first model including solely the four variables measuring different types of political media consumption, the second adding the sociodemographic and external citizenship-related variables, and the third and last model, further adding the politically related variables.

Results
Our multivariate analysis presented in Table 2 delivers a clear-cut result. In agreement with previous research on domestic voters, we find that media usage, and in our case, following politics in media, increases the propensity to vote (H1) also for nonresident citizens. There is, however, no universal effect across the different measures of political media consumption under scrutiny. Rather, we find that the type and focus of the media usage play a significant role when it comes to mobilization of non-resident citizens. In line with hypotheses H2 and H3, the mobilizing effect is limited to political media consumption directed towards the country of origin (H2) while following politics in the country of residence is not positively related to voting in origin-country elections (H3). The overall pattern for the variables for origincountry political media consumption is that the size of the effects decreases substantially as we extend the models with more covariates, but they remain significant.
In relation to hypothesis 4, stating that the format of the media whereby external voters get their political information does not matter for the propensity to vote, we find that it is confirmed if we limit our attention to origin-country media. It is, however, worth noting the irregularities in the effects for residence-country media consumption. While following residence-country politics via social media has a distinctly negative effect on voting in origin-country elections, the effect of following politics in traditional media varies across the presented models. It is hence related to which covariates that we include in the analysis. While the effect is non-significant in models 1 and 3, it is positive in model 2.
The magnitude of the effects is visualized in Fig. 1a-d, where predictions of voting at different levels of political media consumption are presented. 5 These predictions are calculated based on the third and full model with all other covariates set at their mean values. For traditional political media consumption, which displays the strongest effect on voting in all of the models in Table 2, we find that the predicted value of voting is 0.26 for a respondent scoring the lowest value on the traditional political media consumption, compared to 0.68 for a respondent at the highest end of Table 2 Estimation of the effects of political media consumption in country of origin and of residence on turnout in origin-country elections (logistic regression) ***p < 0.001 **p < 0.01 *p < 0.05, ^ reference category "Moved over 20 years ago". All variables are standardized (0-1). Robust clustered standard errors are applied at the country level M1 Political media consumption M2 Political media consumption + sociodemographic M3 Political media consumption + sociodemographic + political  Fig. 1a. Among those who do not follow political news in the traditional origin-country media, only one-fourth reports having voted, compared to more than two-thirds of those with the highest level of political news consumption. The corresponding effect of following origin-country politics via social media (1b) is slightly weaker, with a predicted value for voting of 0.29 for someone who never follows origin-country politics via this channel, compared to 0.55 for someone who follows origin-country politics via social media on a daily basis. Following politics on social media in the country of residence, in turn, has a demobilizing effect, with the predicted value of turning out in origin-country elections being 0.24 for someone who follows residence-country politics on social media every day, compared to 0.46 for a respondent who never does so (Fig. 1d). All in all, the mobilizing effect of following origin-country politics in both traditional and social media is clearly non-trivial. The purpose of this article is to study the effect of political media consumption on voting. However, we find it relevant to comment upon a few other variables included in our models, especially considering the rich data we have at our disposal. In line with previous research, our analyses show that the propensity to vote decreases with the length of the stay abroad (e.g. Ahmadov and Sasse 2016;Peltoniemi 2016;Chaudhary 2018;Finn 2020). It is also worth noting that dual citizenship has a negative effect on voting in origin-country elections in the final model, which would seem to suggest that greater integration in the country of residence may have a negative impact on transnational turnout. Skills in Finnish, which is the language in which most of the political debate is taking place in the origin-country under scrutiny here, do, in turn, have a mobilizing effect, as does having friends in Finland. In line with other research on turnout in elections, we find that older and highly educated citizens, politically interested citizens, and citizens that are reached by the election campaign, are more inclined to vote. These patterns hence appear to be of a general nature and to be applicable also to non-resident citizens.

Concluding discussion
In this study, we examine the extent to which different types of media usage affect electoral participation in the country of origin among non-resident citizens. In line with previous research on media consumption and political participation, we expected a positive effect from political media consumption on participation in elections.
Our outlined expectations of a positive link between consumption of political news in media and participating in elections from abroad were indeed confirmed. A non-trivial effect was identified for both traditional media channels and for social media, and these effects remained even when controlled for socio-demographic and other political factors. Our third hypothesis suggesting that following politics in the country of residence is not positively related to turnout in origin-country elections, was also supported. Interestingly, we even identified a relatively strong, negative relationship for following residence-country politics through social media, suggesting that the more someone follows residence-country politics in social media, the less likely they are to turn out in origin-country elections. We assume that using social media to follow residence-country politics indicates greater integration in the host country, which, in turn, may reduce the interest to take part in origin-country politics. This finding also relates to our fourth hypothesis suggesting that the format of the media whereby external voters obtain information does not have an impact on their propensity to vote. While the fourth and last hypothesis is largely supported by our findings regarding origin-country political media consumption, the finding that residence-country social media consumption has a negative relationship with the propensity to vote, contradicts this hypothesis.
In sum, we find substantial relationships between political media consumption and the propensity to vote among external voters, pointing towards a distinct mobilizing effect of following political affairs in the media. The findings have lined up well with both of our expectations and the relatively limited previous research on the subject. The main contribution of this study is that it distinguishes between political media consumption in the country of origin and of residence and, thereby clarifies the relationship between political media consumption and turnout among external voters. Previous research on these matters is limited, and the extent of the connection between media consumption and turnout that has been examined, means that the findings have been somewhat contradictory (Careja and Emmenegger 2012;Lafleur and Sánchez-Domínguez 2015;Gherghina and Tseng 2016).
Generalizability is always going to be a challenge when examining a specific case. To what extent are our findings regarding political media consumption and the propensity to vote among external Swedish-speaking Finnish citizens relevant to other groups of external citizens? Due to the information-intensive nature of the Finnish electoral system, our case can be considered a most-likely case for identifying a strong link between political media consumption and voting. It is hence plausible that the effect of closely following politics in the country of origin is slightly less important for voting in other contexts where external voters can cast a vote for a party list and, to a larger extent, are able to base their voting decision on a previously established party identification (see also Vintila et al. 2023 in this special issue).
It is further important to acknowledge that we are examining a sub-group of the Finnish non-resident population, those speaking the second national language (Swedish) as their native tongue. However, there is little reason to believe that these findings would not be representative of non-resident Finns in general, or other emigrants from a western country for that matter. The fact that our findings are very well in line with what has been found in studies of domestic media consumption and political participation (see Strömbäck 2017) indicates that this link is not particularly context-sensitive, but rather that it is a general pattern that travels across country borders and even electoral systems. Hence, we think it is safe to assume that the findings are not limited to the population we conducted this study on.
The findings from this study highlight the importance of information for political participation. Without updated knowledge of the relevant issues and the structure of political conflict in a society, participating in elections most likely seem less relevant, and deciding how to vote can be difficult. Based on our findings, the expanded possibilities for keeping oneself updated on political affairs across national borders that come with the online availability of different media channels, should provide good grounds for active engagement among non-resident citizens. However, more research is needed on the extent to which external voters consume media from their country of origin to stay informed about politics. If political media consumption is limited among non-resident citizens, the impact it has on turnout will remain limited as well.

Appendix
Appendix A: Share of Swedish speaking Finnish non-resident citizens in sampled countries and share of sample  Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.