Bringing it all back home? Social media practices by Swedish municipalities Academic research paper on "Media and communications"

Article

Bringing it all back home? Social media practices by Swedish municipalities

European Journal of Communication 28(6) 681-695 © The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0267323113502277

ejc.sagepub.com

(DSAGE

Anders Olof Larsson

University of Oslo, Norway

Abstract

Since the mid-1990s, the Internet has often been pointed to as having the potential for reinvigorating democratic processes. While such overly optimistic claims have largely been disproven by empirical research, a similar rhetoric is now commonly heard in conjunction with the rise of so-called social media like Twitter or Facebook. This study assesses social media practices by politicians on the local level. Featuring a quantitative structural analysis of social media practices by all 290 Swedish municipalities, the study gauges the spread of these types of emerging online platforms in local governments and uses statistical analyses to explain the observed variations. Results indicate that the uptake of social media by municipality governments is slow and in some cases decreasing, perhaps due to the legal and operational risks involved. Overall, the study mainly supports the normalization hypothesis, concluding that offline patterns of municipality characteristics are largely translatable to the online also in the era of 'Web 2.0'.

Keywords

Content analysis, Internet, methodology, political communication, technology

Introduction

The notion of a crisis for democracy is a very common one in the political communication literature. Arguably a broad theme, the predicament is often discussed in terms of voter alienation (Haug, 2008) or declining party membership and lack of participation in civil society (Kalnes, 2009), all in all leading to a 'widespread disillusionment with democratic processes' (Giddens, 2002: 71-72) among citizens in mature democracies. These difficulties could also be seen as a repercussion of the changes to political campaigning that have been identified during the last few decades. An ongoing process of professionalization

Corresponding author:

Anders Olof Larsson, Department of Media and Communication, University of Oslo, Postboks 1093, Blindern 0317, Oslo, Norway.

Email: a.o.larsson@media.uio.no; andersoloflarsson@gmail.com

and an increasing focus on negative campaigning is believed to increase the gap between citizens and those running for office (e.g. Lilleker and Malagon, 2010; Schweitzer, 2011).

As often in times of alleged systemic crises, the role of technology is brought to the fore, sometimes at the hands of pundits, politicians as well as academics (for an overview, see Chadwick, 2006; Dahlberg, 2011; Hilbert, 2009; Negroponte, 1995; Norris, 2000). While we might be quick to conclude that such hopes are a product of the digital era, thoughts regarding the positive democratic impact of new communication technologies have long been held (e.g. Barnett, 1997) - with regard to, for example, mainstays like television (e.g. Tichenor et al., 1970). While the broadcasting of televised political content was seen as holding potential for 'a more informed society and establishing new communication channels' (Polat, 2005: 436), such somewhat optimistic accounts have been contested by empirical research. Meanwhile, as 'periods of technological transition always invite speculation about radical systemic changes' (Kalnes, 2009: 251) the Internet has emerged as the new medium believed to carry with it 'important changes ... in the processes of political information' (Jouet et al., 2011: 362).

The role of the Internet as employed for political purposes has been scrutinized in a number of different contexts, using different approaches. While research themes like 'e-democracy' (e.g. Chadwick, 2003) have sought to define the potential merits of the new medium, most empirical accounts from the 1990s onward have found the role of the Internet in electoral processes to be at best limited. The relatively recent introduction of a 'Web 2.0' paradigm of online publication and consumption (e.g. O'Reilly, 2005), and of social media services like Twitter, Facebook, YouTube and Flickr, geared towards audience contributions, has again given rise to hopes regarding the political potential of the Internet (e.g. Castells, 2007). Based in large part on the suggested 'social networking success of the 2008 US presidential campaign' (Lilleker and Malagon, 2010: 40), rhetoric regarding the crucial role of the Internet in political contexts is being heard once again.

The present article sets out to study the spread of social media use in what appears to be a context largely missing scholarly insights. What is presented here is a structural, exploratory analysis of social media practices by Sweden's 290 municipalities. Moving beyond a descriptive approach, the study employs content analysis, analysis of secondary data and statistical methods to gauge what factors appear to have influence over social media utilization on the local government level.

The study makes a contribution to our knowledge about the role of the Internet in local governments in three ways. First, as the majority of studies on the topic at hand have been performed in a US or UK context (e.g. Gibson and Rommele, 2005; Lilleker et al., 2011; Schweitzer, 2008), insights from other countries are needed. Second, few studies have focused on municipal governments, favouring instead scrutiny of the national level (e.g. Borge et al., 2009; Winswold, 2007). Finally, studies have primarily been performed during election campaigns - an established delimitation, but nonetheless limited in that we learn very little from such studies of the day-to-day uses of these services at the hands of government.

There are a number of different grounds for examining the case of Sweden. As an established, modern democracy, boasting high turnout when it comes to voting in elections at all levels of government, as well as when it comes to Internet penetration and use

(e.g. Facht and Hellingwerf, 2011), the country could be seen as a 'hotbed' of sorts for issues pertaining to the use of novel communication technologies for political purposes. Moreover, on the level of local government, Sweden offers considerable variety with decidedly urban as well as rural societies, all in all making up 290 municipalities to be assessed with the approach described earlier.

Political actors online: Two common perspectives

Since the mid-1990s, we can trace several tendencies throughout the available body of work on online political communication. Two such trends are as follows.

First, while empirical research has often indicated limited changes towards a democratization of political communication processes, the dichotomized perspective on these issues suggested by Resnick (1998) seems to be a common starting point over a decade later. Arguing for the normalization of cyberspace, Resnick proposed that rather than significantly transforming patterns of political engagement, the Internet would come to be more like the offline, thus extending the 'inequalities of the offline world' (Polat, 2005: 451). This was seen as the counterpoint to what is sometimes called the equalization hypothesis, suggesting that the Internet would serve to diminish the differences between minor and major political actors (see also Bentivegna, 2006; Gibson et al., 2008; Lilleker et al., 2011).

Second, while issues such as those under discussion here can be assessed on macro, meso or micro levels of analysis, most studies on how political actors make use of the Internet have been performed on what could be understood as the meso level - focusing on uses of these services during election campaigns in one specific country (e.g. Schweitzer, 2005, 2008, 2011) - or on the macro level - gauging such practices through comparative studies in multiple countries (e.g. Gibson et al., 2008; Hara and Jo, 2007; Jaeger et al., 2010). In the context studied here, the micro level is understood as focusing on local communities - such as municipalities. While a multitude of studies have evaluated online political communication on the meso and macro level, there is a dearth of research looking into these matters in the municipal context.

This study, then, attempts to broaden the scope of research into online political communication by assessing the dichotomous perspective described previously as applied to municipal social media use. As research has focused on individual politicians or political parties, the emphasis on municipalities can at least in part be seen as in alignment with previous analysis of these matters. While what is studied here is not the social media use by politicians, the object of study - municipalities - can nonetheless be seen as political organizations - given that they are at least partly governed by politicians with varying ideological persuasions, giving rise to different priorities within the community.

The case of Sweden

While 'the Internet "happened" in the 1990s' (Loveland and Popescu, 2011: 2), the uptake of the new medium was comparably more rapid in the Nordic region and in the country under scrutiny here - Sweden. Arguably one of the early adopters of the Internet on a country-wide scale, Swedish governments, labour unions and other interest groups

Table 1. Categorization of municipalities.

Municipality category Description N (%) of municipalities

Metropolitan municipalities Population > 200,000 3 (1)

Suburban municipalities 50% of population commutes to work in another municipality, usually metropolitan municipalities 38 (13.1)

Large cities 50,000-200,000 inhabitants, more than 70% of the population live in urban areas 31 (10.7)

Suburban municipalities to > 50% of population commutes to work 22 (7.6)

large cities in a large city

Commuter municipalities > 40% of population commute to work in another municipality 51 (17.6)

Tourism and travel N of guest nights is higher then 21 nights 20 (6.6)

industry municipalities per inhabitant, N of holiday homes is higher then.20 per inhabitant

Manufacturing > 34% of population is employed 54 (18.6)

municipalities in manufacturing, mining, energy, environmental and construction industries

Sparsely populated Municipalities where less than 70% of the 20 (7.2)

municipalities population lives in urban areas and less than eight inhabitants per km2

Municipalities in densely > 300,000 inhabitants within a 112.5 km 35 (12.1)

populated regions radius

Municipalities in sparsely < 300,000 inhabitants within a 112.5 km 16 (5.5)

populated regions radius

Source: Swedish Association of Local Authorities and Regions (english.skl.se/).

have been adamant in providing the infrastructure and financial resources necessary to provide digital amenities for considerable parts of the population (Facht and Hellingwerf, 2011; Gibson, 2004).

For administrative as well as for historical purposes, Sweden is divided into 290 municipalities. Enjoying relative amounts of self-governance from the national parliament, municipal governments are elected every four years. Given geographic and regional differences between different sections of the country, most notably between northern and southern parts of Sweden, the Swedish Association of Local Authorities and Regions classify the municipalities into several categories, as shown in Table 1.

Table 1 shows clear differences between different types of municipalities, mirroring various aspects of a rural-urban dichotomy. There is also considerable variation in terms of the distribution between the categories, further motivating the decision to employ these categories in the following analyses in order to map out the spread of social media use.

Various aspects of Internet use by Swedish municipalities have been assessed by previous research. For example, Wiklund (2005) employed the Habermasian model of deliberative democracy, finding that while properly designed ICT services clearly have

such potential, the services found on the web pages of the majority of the municipalities could be described as 'marginal supplements to the established institutions' (2005: 264) - suggesting that the online realm had brought with it limited change from more traditional practices.

Providing more in-depth findings from a case study of a single Swedish municipality and its initiative to improve representative democracy through stimulating citizen participation between elections, Svensson (2008) focused on the civic committees arranged by the Helsingborg municipality. Results indicated that some areas of municipal-citizen interaction lent themselves particularly well to these types of platforms, Svensson also noticed a 'discursive struggle' (2008: 260) between official and citizen conceptualizations of civic engagement.

Gauging political Twitter practices in the Swedish context, Larsson and Moe (2012) focused on analysing the practices emerging from the #val2010 hashtag used to indicate tweets relevant for the simultaneous national, regional and local elections. While local content was largely absent in the analysis, the results suggested that official political activity employing the specified hashtag was limited at best. The present study, then, attempts to provide a somewhat different picture of how social media practices take place on the municipal level only.

Method

This article takes a structural approach to the study of online political communication. While early structural studies on this topic tended to provide more descriptive accounts of how various aspects of technology were being used, later work usually mapped different types of functionalities found on political actors' websites onto different indices, gauging the degree to which parties were utilizing the different functionalities of the Internet (e.g. Larsson, 2011; Lilleker and Malagon, 2010; Schweitzer, 2005, 2008; Vaccari, 2008a, 2008b). The present study, then, expands on the latter of these two approaches and uses various statistical techniques to assess the factors behind social media employment in Swedish municipalities. The approach is described below.

Data collection

Inspired by Suen (2006), the study makes use of both primary and secondary data. First, in order to assess the degree to which Swedish municipalities employ social media services, the websites of all 290 municipalities were visited and searched to find links to officially sanctioned social media presences (as suggested by Kalnes, 2009). Municipality social media profile pages were archived on 4 September 2012, using a combination of the Web Snapper application for Mac OS X and PDF copies. Besides checking for presence or not on the platforms under scrutiny, quantity of activity was gauged by assessing the number of posts (on Facebook), tweets (on Twitter), videos and photos (on YouTube and Flickr respectively) that had been made or uploaded by each municipality at the time of the archiving. Moreover, the length of presence on the respective platforms was noted.

Following previous, similar studies (e.g. Schweitzer, 2005, 2008), intra-coder reliability was calculated by recoding a random 5% of the original material. To accommodate

this, the Krippendorf alpha macro for SPSS was used (Hayes and Krippendorff, 2007). Levels of reliability for the variables measuring presence and degree of activity were all reported above the suggested .70 level of acceptance for conservative tests such as the one employed here (e.g. Lombard et al., 2002).

Moreover, secondary data were gathered in order to test the influences of variables identified by previous research on the uses of social media as discussed above. Data for these variables were gathered from the Swedish Post and Telecom Authority (www.pts. se/), Statistics Sweden (www.scb.se/) and Swedish Association of Local Authorities and Regions (english.skl.se/). The reason for inclusion and their operationalization for analysis are discussed below.

Ideology. Previous research has found that left-leaning parties or governments tend to utilize their web presences in more innovative ways than more conservative political actors (e.g. Borge et al., 2009; Larsson, 2011; Vaccari, 2008a; Wohlers, 2009). Taking the focus of the present study into account, a dummy variable indicating whether or not the municipality was governed by a left-leaning party or combination of parties was included.

Broadband reach. As pointed out by Haug, a 'high level of access to broadband infrastructures might put pressure on the municipality to go online' (2008: 84). Other studies similarly suggest that the infrastructure of high-speed broadband needs to be in place for advanced Internet use by municipality officials to take place (e.g. Gibson et al., 2008; Jensen et al., 2007; Suen, 2006). For the present study, this was gauged as the percentage of the population in each municipality with access to high-speed Internet connections.

Higher education. Following previous studies on citizen use of the Internet for various political purposes, we can conclude that level of education has some influence over such practices (e.g. Jensen et al., 2007; Jouet et al., 2011). Similarly, the percentage of citizens in a given region or municipality with higher education appears to influence the degree to which the local government employs the Internet (e.g. Wohlers, 2009). This was assessed by following the definition offered by Borge et al. (2009) - the percentage of the population in each municipality with a university degree.

Number of citizens. Gibson and Rommele suggest that 'the local context also appears to play a part' (2005: 105-106) in processes such as the ones studied here. One of the more common variables used in these types of studies, the number of citizens in a specific municipality often serves as a proxy of sorts for city size (e.g. Dahl and Tufte, 1973). Such a measurement was also included. Indeed, city size might serve as a good indicator of the community needs of a municipality, something that appears to hold true also in more recent studies (e.g. Wohlers, 2009).

Voter turnout. While a reverse tendency might have been expected, Borge et al. (2009) suggest that municipalities with higher voter abstention rates cultivate more communicative initiatives, for example using the Internet in order to communicate with citizens. Perhaps such declines in voter participation puts the pressure on the local political system to perform in novel ways (as suggested by Haug, 2008). For the

present purposes, this was measured as the percentage of citizens casting their votes in the 2010 elections.

Data analysis

In order to facilitate analyses, a series of indices were created (inspired by Larsson, 2011, 2012a; Schweitzer, 2005, 2008). Specifically, activity indices were calculated by dividing the number of posts, tweets, videos and photos posted on respective sites by the number of days that each municipality had maintained a profile on each of these networks. This process allowed for some insight into the 'day-to-day' activity of each municipality. A similar overall index was calculated to evaluate overall activity, adding all aforementioned indices and dividing by the same amount. As such, the data detailed presence or non-presence, but also provided a measurement, albeit crude, of the degree of activity on each platform.

Given the characteristics of the collected data, a series of different analyses were performed. First, descriptive statistics are used in order to provide an overview of the uses of social media in the studied municipalities. Second, mean comparisons between the activity indices created were performed, assessing the differences between municipalities in different categories (as discussed earlier). Finally, the explanatory power of the variables discussed in the previous section were assessed by means of a series of multiple regression analyses. The results from these analyses are presented in the following section.

Results

Using a 2010 survey performed by the Swedish Association of Local Authorities and Regions regarding municipal use of social media as a baseline of sorts (see www.skl. se/press/nyheter_2/nyheter_2010_1/over_60_procent_anvander_sociala_medier), the results from this particular study are reported in comparison with the data gathered for the article at hand. This is presented in Table 2, which also presents data regarding activity on the four services under scrutiny.

Perhaps as a result of the novelty aspect of emerging online platforms like social media, the overall picture presented in the first two columns of data in Table 2 is one of diminishing presence. In comparison with the data collected for this study, the self-reported data from information officers in Swedish municipalities, gathered in 2010, suggest that these actors have taken a step back when it comes to using social media. For example, while 92% of Swedish municipalities claimed to have an official page on Facebook, this statistic was 42% according to the data presented here. While some of these differences are indeed larger than others, and while the data reported differ with regard to method of collection, the differences between the data from 2010 and 2012 are striking.

Furthermore, it could be argued that simply instating an official municipal Twitter profile or Facebook page without also employing some variety of continuous maintenance is not enough. While it is hard to designate a specific suitable or even 'best practice', some degree of activity in maintaining communication between the governments

Table 2. Descriptive statistics regarding use of social media by municipalities.

2010 - 2012 - 2012 - Profile 2012 - Mean

Municipalities Municipalities with no activity N of days with

with profile N (%) with profile N (%) N (%) profile N (SD)

Twitter 154 (53) 137 (47) 26 (19) 817 (366)

Facebook 267 (92) 122 (42) 4 (3) 637 (302)

YouTube 180 (62) 76 (26) 30 (39) 823 (425)

Flickr 35 (12) 14 (5) 1 (7) 777 (295)

and their respective electorates should be necessary in order to uphold the suggestions offered by the 2.0 rationale (e.g. O'Reilly, 2005). As visible in the third column of data in Table 2, while the majority of municipal actors appear to be active in some way on all four services studied here, large variations can be discerned in this regard. For example, while a mere 3% of municipalities on Facebook appear to have posted nothing on their respective pages, 19% of those on Twitter and 39% of municipalities on YouTube have not posted or uploaded content to those respective services. This can perhaps be interpreted as part of what could be referred to as the novelty or even 'bandwagon' effect of social media - the notion that it is necessary for actors such as municipalities to maintain some sort of presence on these services. As the results in Table 2 would suggest, establishing such profiles can be relatively easily done - while sustaining one through regular activity probably demands more effort than the veritable click of a button. Moreover, as Twitter and Facebook appear more frequently used than YouTube and Flickr, we must keep the differences between the social media studied in mind. For example, while maintaining an active Twitter account could be expected to take up time, it seems reasonable to assume that the maintenance of a municipal YouTube account - involving the production of video - would be even more time-consuming.

Finally, the fourth column in Table 2 details the mean number of days that municipalities have been present on the services studied. As the standard deviations would indicate, there is considerable variation here. With such discrepancies in mind, the average municipality appears to have utilized the services discussed here slightly longer or slightly shorter than a two-year period. Following Lilleker et al., this would place Swedish municipalities as slight laggards, since 'Web 2.0 reached the mainstream on the World Wide Web in 2007/8' (2011: 196). While there is a possibility that previous social media profiles of Swedish municipalities have come and gone, this is arguably beyond the scope of this study.

Next, while descriptive statistics like the ones presented in Table 2 provide us with a means to grasp the overarching predispositions of the social media practices of Swedish municipalities, they tell us very little about more specific variations between different types of municipalities. Employing the typology described earlier, Table 3 presents mean comparisons between each municipality type regarding the aforementioned activity indices.

While no clear trends appear in the analysis presented in Table 3, we can point to at least three tendencies. First, while the standard deviations reported are generally large,

Table 3. Mean comparisons of activity indices.

Municipality Overall Twitter Facebook YouTube Flickr

category index index index index index

Metropolitan Mean .25 .84 .20 .17

municipalities N 1 2 2 1

SD .19 .16

Suburban Mean .37 .57 .36 .03 .05

municipalities N 38 22 16 16 3

SD .29 .83 .27 .02 .05

Large cities Mean .56 1.1 .29 .05 1.79

N 30 20 22 23 4

SD .27 .88 .14 .05 2.05

Suburban Mean .27 .20 .38 .01 1.45

municipalities N 22 11 8 4 1

to large cities SD .27 .12 .21 .04

Commuter Mean .23 .57 .31 .01 .36

municipalities N 51 24 17 5 1

SD .11 .05 .22 .01

Tourism and Mean .19 .32 .32 .02

travel industry N 19 6 5 3

municipalities SD .12 .22 .11 .01

Manufacturing Mean .27 .36 .41 .01 .13

municipalities N 54 25 22 9 4

SD .19 .30 .22 .01 .10

Sparsely Mean .11 .24 .65 .03

populated N 21 4 3 2

municipalities SD .09 .16 .34 .02

Municipalities Mean .35 .70 .29 .02 .38

in densely N 35 18 18 12 1

populated SD .28 .66 .12 .02

regions

Municipalities Mean .18 .34 .27 .01 .12

in sparsely N 16 5 6 1 .01

populated SD .14 .31 .16 .01 .01

regions

Total Mean .29 .57 .34 .03 .71

N 287 137 119 76 14

SD .19 .43 .20 .02 1.2

Kruskal- .000 .004 .121 .041 .293

Wallis

the mean comparison indicates that (the municipality category of) Large cities emerge as the most ardent overall users of social media, scoring the highest means in the Overall, Twitter and Flickr activity indices. While there are exceptions to this - for example, the fact that Metropolitan municipalities scored the highest on the YouTube activity index

Table 4. Multiple regression analyses predicting social media index levels.

Predictor Overall Twitter Facebook YouTube

variable index index index index

Structural Governed by -.05 .09 .23* -.02

predictors the lefta

Large cityb 22** .10** -.09* .19*

Broadband .16* .19* -.20 .03

Sociodemographic Higher .08 -.09 .18* -.11

predictors education

N of citizens .09 .10 -.20 .49

2010 voter -.06 -.04* -.11 .07

turnout

R2 .18 .16 .12 .28

Adj. R2 .15 .11 .07 .21

Note: Standardized beta values presented. **p < .01, * p < .05.

aDichotomous variable: 1 = governed by the left, 0 = other government. bDichotomous variable: 1 = large city, 0 = other types of municipalities.

- it would appear that comparably, the most ardent municipal use of social media is performed by information officers and the like in cities with up to 200,000 inhabitants.

Second, the other exception to the above stated tendency can be seen in the column containing the means for the Facebook activity index. Here, Sparsely populated municipalities emerge as the most enthusiastic when it comes to utilization of said service. Perhaps due to Facebook's popularity in Sweden (e.g. Nordicom, 2012), this has become the social media of choice for many municipalities.

Third, due to the quality of the data, the Kruskal-Wallis one-way analysis of variance ;procedure for SPSS was used to test the significance of the differences shown in Table 3. The results for the test are visible in the bottom row of Table 3, indicating that significant differences were found with regard to means for the Overall, Twitter and YouTube indices. While the Flickr index is clearly skewed due to extreme values, the non-significance of the Facebook index could perhaps again be interpreted as a token of Facebook's popularity in Swedish society.

As the category of Large cities proved to score the highest means overall, a dummy variable was constructed indicating whether or not a municipality belonged to this particular category. This variable was then employed together with the aforementioned independent variables in a series of multiple regression analyses, as reported in Table 4. Given the comparably limited number of municipalities using Flickr, this service was excluded from further analysis.

First, we turn to what is introduced in Table 4 as structural predictors - variables pertaining to the city itself. While previous research has indicated that ideology would play a part in the degree to which parties involve themselves in online practices such as the ones studied here, the mostly non-significant results presented for the dummy variable

assessing left-leaning political affiliation in the municipalities would suggest otherwise. This contrast to previous studies is further confirmed by the pan-European study by Lilleker et al., as their data suggested that 'ideology played only a minor role' (2011: 206). As such, the 'deideologization in party campaigning' (Schweitzer, 2008: 460) can be seen also in the local, Swedish context.

The dummy variable included to test for the effect of municipality size, emerged as a significant predictor for all indices employed. As such, the influence of city size noted in Table 3 remains - albeit with rather small effect sizes. This would indicate that social media practices are most plentiful in larger cities, but not in metropolitan areas. Perhaps the Large city category, although admittedly a rudimentary definition, serves to create a critical mass for use and need of these communication services in the context of municipal government. Smaller cities might not have the resources to follow suit, and metropolitan areas might have divided their communicative efforts into smaller administrative units.

The third independent variable, testing for access to high-speed Internet connections, proved to be significant in two of the cases analysed here - the overarching index and the Twitter index. While the reported beta values are modest at best, this still says something about the role of information technology as infrastructure - shaping communication practices in certain areas, while apparently leaving others without such influences.

Finally, the three sociodemographic predictor variables are assessed. As previously suggested, measures such as Higher education and N of citizens could be considered essential in this regard; but only the former emerged as a significant predictor for one of the indices employed. Moreover, the variable gauging 2010 voter turnout emerged as a negative predictor. Such an effect might stem from pressure applied on politicians to try novel means of connecting with the populace when faced with declining voter turnout. The results presented here seem to indicate that such a suggestion is also at least partially valid in the Swedish context.

Discussion

While the findings presented by Lilleker et al. (2011) indicate a slow but steady uptake of social media features by European national governments, the results offered here would suggest somewhat dissimilar developments on the local level - at least for the Swedish case. In the following, some of these issues are raised and suggestions for future research are provided.

One such point of divergence can be observed by focusing on rural and urban settings for municipal social media use (e.g. Polat, 2005). As pointed out by Van Dijk (2005), the dichotomous approach to the study of online political communication has largely omitted issues of 'digital divides'. Indeed, debate about such divides has primarily focused on the reinforcing of such gaps in the populace of structures that generally favour citizens of higher socioeconomic status. While Sweden boasts leading statistics regarding Internet penetration, the results presented in this article suggest that the digital divide makes itself present also in this context. While these are indeed complex synergies of cause and effect, the data indicate that much as larger political parties make use of online innovations to a larger extent than their smaller counterparts, comparable patterns can be found

with regard to city sizes. Similar to the findings reported by Borge et al. (2009), the processes at play here can at least partly be understood along a centre-periphery axis (e.g. Bonfadelli, 2002; Warschauer, 2003). Of course, the use of social media by municipal governments is not in itself an absolute necessity - local governments are perhaps best left out of techno-utopian ideals since they are probably the foremost judges of the communicative needs of their respective societies (Suen, 2006). However, as parliamentary political apathy spreads among the young, and as this particular group is supposedly prone to engaging in civic issues on various social media platforms, the use of these services could prove to be an interesting opportunity. Future research might address this issue by providing case studies looking into these uses as manifested in different types of municipalities - belonging to the high or low end as reported here, perhaps. Such qualitative analysis could help in providing more in-depth insights than those presented here. Moreover, while the independent variables employed here provided some insights into the more structural qualities under scrutiny, future quantitatively oriented efforts should combine what could be seen as 'classical' predictors with more novel approaches to variable selection - such as looking at the degree of party competitiveness in local political contexts, or the use of other communication channels by local politicians. Such approaches, ideally combined with qualitative explorations as previously discussed, could help increase our collective knowledge on these issues.

Indeed, the perceived necessity of use could perhaps be important here. As usage rates for municipalities appear to have dropped from 2010 to 2012, perhaps due to difficult legal requirements (e.g. Gibson et al., 2008; Klang and Nolin, 2011), this signals that these emerging platforms are not as important or malleable for political legwork as some might suggest. Contrariwise, the interest from citizens to gather information regarding their local community on novel platforms might not be sufficient to mobilize the necessary resources (e.g. Haug, 2008; Winswold, 2007). Perhaps the use of what Kleis Nielsen labelled 'Mundane Internet Tools' (2010), online staples such as email, holds strong also in the local context. As suggested by Jackson and Lilleker, 'Email may be leading more to evolutionary than revolutionary change' (2004: 526). As this particular online practice has gone from a 'killer app' (e.g. Downes and Mui, 2000) to a mundane tool, it is possible that social media uptake will fare similarly. Future research should thus study online developments in a longitudinal fashion, perhaps gauging the 10-year lag period of technology uptake in local governments suggested by King (1982). The argument is made here that such studies, providing over time comparable results, will prove helpful in the ongoing discussions regarding normalization and equalization in political communication.

Such adoption might take longer than a decade. These processes are indeed difficult to isolate from 'the properties of the political system they are embedded in' (Polat, 2005: 446), and as these institutions change rather slowly, we should not expect a radical change in use of social media at the hands of politicians any time soon. While the 'vote-maximizing' view of political campaigning might be overtly pessimistic, politicians and public administrators alike are experiencing difficulties in adapting to the online world, 'struggling with the baggage of history' (Negroponte, 1995: 230) - as shown here, and in other studies (e.g. Larsson, 2012b). Future research into these matters might find possible theoretical explanations for these issues by adopting perspectives discussed by scholars like Couldry (2012) or Flichy (2007, 2008).

In closing, perhaps the activity sought here is not necessarily taking place on different official Facebook pages or Twitter feeds. While this study provided insights into how such web presences are used, future research should attempt to track similar activity on pages other than the ones sanctioned by the government. While most research on this topic - including this effort - tend to give weight to the less positive side of the aforementioned dichotomous perspective, it has mostly been focused on official profiles and pages. Placing the gaze elsewhere could provide complementary insights into these practices, perhaps transgressing the boundaries of the often employed normalization-equalization perspective.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Barnett S (1997) New media, old problems: New technology and the political process. European

Journal of Communication 12(2): 193-218. Bentivegna S (2006) Rethinking politics in the world of ICTs. European Journal of Communication 21(3): 331-343.

Bonfadelli H (2002) The Internet and knowledge gaps: A theoretical and empirical investigation.

European Journal of Communication 17(1): 65-84. Borge R, Colombo C and Welp Y (2009) Online and offline participation at the local level: A quantitative analysis of the Catalan municipalities. Information, Communication and Society 12(6): 1-30.

Castells M (2007) Communication, power and counter-power in the network society. International

Journal of Communication 1(1): 238-266. Chadwick A (2003) Bringing e-democracy back in: Why it matters for future research on e-

governance. Social Science Computer Review 21(4): 443-455. Chadwick A (2006) Internet Politics: States, Citizens, and New Communication Technologies.

New York: Oxford University Press. Couldry N (2012) Media, Society, World: Social Theory and Digital Media Practice. Cambridge: Polity.

Dahl RA and Tufte ER (1973) Size and Democracy. Stanford, CA: Stanford University Press. Dahlberg L (2011) Re-constructing digital democracy: An outline of four 'positions'. New Media

and Society 13(6): 855-872. Downes L and Mui C (2000) Unleashing the Killer App: Digital Strategies for Market Dominance.

Boston, MA: Harvard Business School. Facht U and Hellingwerf K (2011) Nordicom-Sveriges Internetbarometer 2010. Goteborg: Nordicom.

Flichy P (2007) Understanding Technological Innovation: A Socio-technical Approach.

Cheltenham: Edward Elgar. Flichy P (2008) The Internet Imaginaire. Cambridge, MA and London: MIT Press. Gibson R (2004) Web campaigning from a global perspective. Asia-Pacific Review 11(1): 95-126.

Gibson RK and Rommele A (2005) 'Down periscope': The search for high-tech campaigning at the local level in the 2002 German federal election. Journal of E-Government 2(3): 85-112.

Gibson RK, Lusoli W and Ward S (2008) Nationalizing and normalizing the local? A comparative analysis of online candidate campaigning in Australia and Britain. Journal of Information Technology and Politics 4(4): 15-30.

Giddens A (2002) Runaway World: How Globalisation is Reshaping our Lives. London: Profile Books.

Hara N and Jo Y (2007) Internet politics: A comparative analysis of U.S. and South Korea presidential campaigns. First Monday 12(9).

Haug AV (2008) Local democracy online: Driven by crisis, legitimacy, resources, or communication gaps? Journal of Information Technology and Politics 4(2): 79-99.

Hayes AF and Krippendorff K (2007) Answering the call for a standard reliability measure for coding data. Communication Methods and Measures 1(1): 77-89.

Hilbert M (2009) The maturing concept of e-democracy: From e-voting and online consultations to democratic value out of jumbled online chatter. Journal of Information Technology and Politics 6(2): 87-110.

Jackson NA and Lilleker DG (2004) Just public relations or an attempt at interaction? British MPs in the press, on the web and 'in your face'. European Journal of Communication 19(4): 507-533.

Jaeger P, Paquette S and Simmons S (2010) Information policy in national political campaigns: A comparison of the 2008 campaigns for president of the United States and prime minister of Canada. Journal of Information Technology and Politics 7(1): 67-82.

Jensen MJ, Danziger JN and Venkatesh A (2007) Civil society and cyber society: The role of the internet in community associations and democratic politics. The Information Society: An International Journal 23(1): 39-50.

Jouet J, Vedel T and Comby JB (2011) Political information and interpersonal conversations in a multimedia environment: A quantitative and qualitative examination of information practices in France. European Journal of Communication 26(4): 361-375.

Kalnes 0 (2009) Norwegian parties and Web 2.0. Journal of Information Technology and Politics 6(3): 251-266.

King JL (1982) Local government use of information technology: The next decade. Public Administration Review 42(1): 25-36.

Klang M and Nolin J (2011) Disciplining social media: An analysis of social media policies in 26 Swedish municipalities. First Monday 16(8). Available at: firstmonday.org/ojs/index.php/fm/ article/view/3490/3027.

Larsson AO (2011) 'Extended infomercials' or 'Politics 2.0'? A study of Swedish political party web sites before, during and after the 2010 election. First Monday 16(4).

Larsson AO (2012a) Interactivity on Swedish newspaper websites: What kind, how much and why? Convergence: The International Journal of Research into New Media Technologies 18(2): 195-213.

Larsson AO (2012b) 'Rejected bits of program code': Why notions of 'politics 2.0' remain (mostly) unfulfilled. Journal of Information Technology and Politics 10(1): 72-85.

Larsson AO and Moe H (2012) Studying political microblogging: Twitter users in the 2010 Swedish election campaign. New Media and Society 14(5): 729-747.

Lilleker DG and Malagon C (2010) Levels of interactivity in the 2007 French presidential candidates' websites. European Journal of Communication 25(1): 25-42.

Lilleker DG, Koc-Michalska K, Schweitzer EJ et al. (2011) Informing, engaging, mobilizing or interacting: Searching for a European model of web campaigning. European Journal of Communication 26(3): 195-213.

Lombard M, Snyder-Duch J and Bracken CC (2002) Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research 28(4): 587-604.

Loveland M and Popescu D (2011) Democracy on the web. Information, Communication and Society 14(5): 684-703.

Negroponte N (1995) Being Digital. New York: Knopf.

Nielsen RK (2010) Mundane internet tools, mobilizing practices, and the coproduction of citizenship in political campaigns. New Media and Society 13(5): 755-771.

Nordicom (2012) Nordicom-Sveriges Internetbarometer 2011. MedieNotiser 2012, Vol. 2. Gothenburg: Nordicom.

Norris P (2000) A Virtuous Circle: Political Communications in Postindustrial Societies. Cambridge: Cambridge University Press.

O'Reilly T (2005) What is Web 2.0? Design patterns and business models for the next generation of software. Available at: www.oreillynet.com/lpt/a/6228.

Polat RK (2005) The internet and political participation: Exploring the explanatory links. European Journal of Communication 20(4): 435-459.

Resnick D (1998) Politics on the internet: The normalization of cyberspace. In: Toulouse C and Luke TW (eds) The Politics of Cyberspace. New York: Routledge, pp. 48-68.

Schweitzer EJ (2005) Election campaigning online: German party websites in the 2002 national elections. European Journal of Communication 20(3): 327-351.

Schweitzer EJ (2008) Innovation or normalization in e-campaigning? A longitudinal content and structural analysis of German party websites in the 2002 and 2005 national elections. European Journal of Communication 23(4): 449-470.

Schweitzer EJ (2011) Normalization 2.0: A longitudinal analysis of German online campaigns in the national elections 2002-9. European Journal of Communication 26(4): 310-327.

Suen I-S (2006) Assessment of the level of interactivity of e-government functions. Journal of E-Government 3(1): 29-51.

Svensson J (2008) Kommunikation, medborgarskap och deltagardemokrati -En studie av medbor-garutskotten i Helsingborg. PhD thesis, Lund University, Lund.

Tichenor PJ, Donohue GA and Olien CN (1970) Mass media flow and differential growth in knowledge. The Public Opinion Quarterly 34(2): 159-170.

Vaccari C (2008a) Research note: Italian parties' websites in the 2006 elections. European Journal of Communication 23(1): 69-77.

Vaccari C (2008b) Surfing to the Elysee: The internet in the 2007 French elections. French Politics 6(1): 1-22.

Van Dijk J (2005) The Deepening Divide: Inequality in the Information Society. Thousand Oaks, CA: Sage.

Warschauer M (2003) Technology and Social Inclusion: Rethinking the Digital Divide. Cambridge, MA and London: MIT Press.

Wiklund H (2005) A Habermasian analysis of the deliberative democratic potential of ICT-enabled services in Swedish municipalities. New Media and Society 7(2): 247-270.

Winswold M (2007) Municipal websites in the local public debate: Supplying facts or setting agenda? Nordicom Review 28(2): 7-23.

Wohlers TE (2009) The digital world of local government: A comparative analysis of the United States and Germany. Journal of Information Technology and Politics 6(2): 111-126.

Copyright of European Journal of Communication is the property of Sage Publications, Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.