Graduate School of Communication – Universiteit van Amsterdam Master’s Thesis
Lecturer: Judith Möller Vera le Blanc [10578730] June 27th, 2014
Word count: 4691
Modern Family
Examining the effects of intergenerational transmission on political socialization in a digital age.
Subject
To what extent the traditional role of parents as intergenerational transmitters of ideological identification is still prevalent and talking about politics with their children can lead to an increased use of the internet for political activities.
Abstract
Up until the second half of the 1960’s, the Dutch political landscape was relatively stable. Verzuiling (pillarization) had made society divided into multiple groups, of which its
members shared the same religious views or ideology. School, social life and choice of news media were all within their own, closed pillar. This division of society made the political socialization process very orderly; children grew up to vote what their parents voted for. However, the ontzuiling (depillarization) and immense popularity of the internet have changed this stable landscape radically. Young adults can access all different kinds of information, people and opinions through the internet, possibly making the role of their parents as political transmitters less important.
At the same time, there are serious concerns about the political participation of today’s youths. Voting rates are declining, as well as political interest and –knowledge.
Given the fact that almost no research in this field has been conducted after the 1970’s, there is only sparse information on how prevalent intergenerational transmission is in political socialization nowadays, and whether this has implications for the political use of the internet by youths. This is a serious shortcoming in the academic world, as well as in (Dutch) society. We can not rely anymore on the results of studies which are more than half a century old; too much has changed. Furthermore, for parents, politicians and other policy makers, it is of vital importance to know how today’s young adults shape their political preferences.
This research makes use of the data of the first wave of a three-wave survey among Dutch youths aged 16 to 18 years old. A logistic regression indicated that youths having the same ideological preference as their father, was not dependent on whether the respondent knew his ideological preference, whether the respondent’s family talked about politics regularly or on internet use, social status or education level. Furthermore, a Mann-Whitney U-analysis showed that there was no significant effect of talking about politics with parents on the political use of the internet.
While the chosen models fitted the data well, it is likely that future research will find significant results for the same hypotheses. The reliability of the chosen operationalizations did not always meet the standard that is seen as appropriate. Also, focusing on party
preference instead of ideological preference could evoke different results.
Introduction
Decreasing turnout numbers as well as party membership and a lack of political interest and political knowledge; it is the concern of many that the young adults of the twenty-first century are not politically engaged anymore in the traditional sense of the word (Bennett, 1997; Delle Carpini, 2000; Norris, 2003; Bennett, 2008).
These developments are often ascribed to the severe changes that most first-world democratic countries have gone through in the past decades. According to some researchers, political participation should not only be measured by looking at voting turnout, but also by looking at less ‘hard’ characteristics, such as doing voluntary work or engaging in social causes
However, since democracies can not function without having a large and representative sample going voting, it is of vital importance that future generations will keep doing so as well. Since it is suggested that “early political participation predicts future electoral involvement” (Glasford, 2008), large amounts of scientific attention are geared towards youths and their political participation.
Political participation is an outcome of the political socialization process; “the process by which persons learn to adopt the norms, values, attitudes and behaviors accepted and practiced by the ongoing (political) system” (Sigel, 1965; p.1). The agents of political socialization vary throughout the different phases of life. Among the most important influences of political socialization on youths are, perhaps not surprising, the parents. The outcomes of the process (political participation, -attitudes, -knowledge and -norms/-values) are established trough the use of mass media and political talk.
Oddly enough, while the Dutch political landscape, as well as media and society have changed drastically, not much research on the subject of youths and the process of their political socialization has been done since the 1970’s. And even though there are numerous researches connecting political participation of youths with internet use, there is a lack of research which connects the role of parents and new media with political socialization.
Pillarization
Up until the late sixties of the twentieth century, Dutch parents were seen as the most
influential entity in the political socialization of their children, shaping their degree of interest and their ideological preference. Children were raised in a pillarized society, meaning they had no, or minimal, contact with people of different political beliefs. Even the newspapers
and tv-programmes were pillarized. Because of this societal distribution, there were only a few political parties, which all had a very stable base of voters.
Following the depillarization of the Dutch society, which was accelerated by growing wealth in the fifties of last century, the once so stable political landscape started to change. It was no longer a given that children would vote for the same party as their parents, as they now had the privilege and freedom to talk to whomever person they wanted, read whichever newspaper they wanted and watch whatever tv-channel they wanted.
In the current Dutch society, the political landscape is capricious; political parties do not have the security of a steady following anymore. In a way, politics are commercialized, because every party has to ‘sell’ its ideas through catchphrases, populism and other heuristics, in order to get votes of the electorate.
Whereas older generations might still have a strong party preference, youths nowadays usually do not. Before every new election, they decide what they will vote, often using heuristic cues from billboards, tv commercials or, more current, the internet. It is commonly thought that cognitive heuristics play an important role in making political decisions, because they allow people to simplify the often complex political affairs (Kahneman & Tversky, 1973; Ventura, 2001). However, others state that the simplification that comes with heuristics can be of danger for the quality of decision making. Lau & Redlawsk (2001) for example, find that heuristics only work for the people who already are well informed, while they lead to qualitatively poor decision making with the politically uninformed.
The invention and rise of the internet only amplified this development of growing ideological independence and instability, as it gives its users unlimited freedom to communicate with others and to search for information.
Today, almost every household in The Netherlands has access to the web and internet users are becoming increasingly younger.
Children are getting familiar with the use of the internet at an increasingly younger age. According to the Centraal Bureau voor Statistiek (CBS; the Dutch bureau for statistics), in 2013, more than 80% of the Dutch youths aged from 12-15 used the internet on a daily basis and of the 15-25 year olds, this is almost 100%. In 2005, these percentages were considerably lower (respectively 71% and 77%) (CBS, 2014).
To add to that, not only is the daily use of the internet increasing, also the amount of time that is spent on this medium is growing, just as the time spent on mobile phones with internet on them (Ridout, Foehr and Roberts, 2010; CBS, 2014).
However, it is not a matter of course that young people all use the internet to gain political information. While there are virtually no limits to communicating and information seeking anymore, the ways to avoid politics are countless as well. A person can spend all his time online and still not encounter any political discourse at all. It might as well be that they only use it for entertainment purposes, instead of more serious informative purposes. As Möller (2013) stated, in order for young adolescents to search for political information, there needs to be a certain level of political knowledge already.
Research
The transmission of political preferences from parents to their children has been the subject of research since the early sixties of the twentieth century, when Converse & Dupeux (1962) compared party identification between American and French citizens. Besides having found large differences between the countries regarding party identification, they found that respondents who could recall the political preference of their fathers, were also the
respondents who were most politically interested. This was the case with both American and French respondents.
The Converse & Dupeux study has served as a base for the model proposed by Percheron & Jennings, which stated that the nature of this transmission is related to the type of democracy in which a family lives. In multiparty systems, intergenerational transmission consists merely of preferences on a left-to-right scale, while in two-party systems, party preferences are passed on (Percheron & Jennings, 1981).
There is something peculiar about the existing research on political socialization. While in the 1950’s up until the 1970’s, the subject was quite popular amongst academics, in the decades after that, almost no research in this field was done. It is only since the early 2000’s or so, that sporadically researchers have been investigating political socialization again.
This makes for the odd situation that the current state of knowledge about political
socialization is quite dated. Today’s academics are dependent on outdated studies to base their perceptions on, which is far from ideal for a number of reasons.
One of the most serious shortcomings in the current body of research is the lack of knowledge about the influence of the internet on political socialization.
This medium has become one of the most important channels for political communication, be it through news websites, social media or political blogs, and it is important to understand to what extent this is an instrument in political socialization.
Whereas before the digital era, parents were seen as the most important influence on the political socialization of their children, it can be argued that the internet is taking up an important part in this process as well. However, it is not a given that children use the internet to find political information, or to discuss politics with others.
It is likely that the motivation of young adults to search for political information on the internet is to some extent dependent on the level op politicization of the family.
This leaves us with questions such as; Until what extent are parents still of influence on the political socialization of their children? Are youths still likely to have the same ideological preference as their father? Are youths from highly politicized families more likely to use the internet for politically related activities?
The Percheron & Jennings model is still used in current research on political preferences across generations. Children receive political labels that serve as voting cues from their parents, in the form of party identification or ideological position (Ventura, 2001). In The Netherlands and other multiparty systems, these voting cues usually do not result in affiliation with one specific party, like it does in two-party systems such as the United States, but in ideological identification, which can be measured on a left-right scale. In his research,
Ventura refers to the model proposed by Percheron & Jennings (1981), which states that “The more parties there are in the system […] the more dominant the ideological component will be, at the expense of the party-specific one.”
As Kent Jennings, Stoker & Bower (2009) found, intergenerational transmission regarding political preferences is most strong in families which are highly politicized. In their -US-focused- research, it is found that the amount of political cues that is provided from a young age onwards by parents, will increase their influence on the political identification of their children. This is of course an American study, which means that the party system is dual, instead of plural.
De Vreese & Möller (Forthcoming) state that children start to challenge their parents’
political beliefs during their teens, as they are turning to their peers and the media as agents of political socialization.
Many studies (Converse & Dupeux, 1962; Kent Jennings, 1984;) have been asking respondents to recall their parents’ political preferences or voting behaviour. This method could pose a threat to the validity of a study, since recalled data are usually biased or colored by time. However, De Vries, De Graaf & Eisinga (2008) addressed this problem and found that it is better to include this measurement error in the study, than to leave it out. The effects of political preferences of parents are just too strong to exclude.
The early political socialization studies have mainly focused on traditional media. This is not surprising, given the fact that it has only been in the last decennium or so that the internet has become a mainstream medium in most democracies. The internet is usually ascribed as a less trustworthy medium for political information compared to traditional media such as
newspapers and television news (Johnson & Kaye, 2000; Flanagin & Metzger, 2000), whereby information seekers must be very motivated in order to control their sources for credibility (Metzger, 2007; Flanagin & Metzger, 2000). However, it can not be denied that
this is the medium of choice for lots of citizens nowadays (Metzger, 2007). Therefore, I will investigate whether there is a difference in internet use for political purposes among youths who often talk about politics with their parents and youths who do not. Furthermore, I will analyze the extent through which intergenerational transmission still has an effect on ideological identification.
Research on political identification has in the past been largely focused on the effect of
traditional media consumption. However, the current generation of 18-year-olds has grown up using the internet and will probably receive most of its (political) information through this medium. In this study, I will analyze whether parents have influence on the ideological identification of their children and whether they have an influence on their online political behavior.
Hypotheses
When parents regularly talk about politics with their family, their children will be more likely to reflect the same political ideology later on in their lives (Ventura, 2001; Converse &
Dupeux, 1962). Studies on intergenerational transmission have found that the parental level of political involvement, as well as children’s ability to recall the political preferences of their parents, lead to similar political preferences of the children (Converse & Dupeux, 1962; Beck & Kent Jennings, 1975; Nieuwbeerta & Wittebrood, 1995). These studies have mostly
focused on party identification instead of ideological identification. This is traditionally seen as a more accurate measure of political preferences, since the perception of being left- or right-wing is fairly subjective, whereas a party preference is more absolute. Also, not all youths will be able to distinguish between left- and right-wing, whereas parties are more easy to identify with.
However, citizens of multiparty countries do not tend to have a strong affiliation with just one party; instead they rate themselves on an ideology scale from left- to right-wing. In The Netherlands, citizens tend to identify themselves not on a left-to-right-scale, but with multiple parties that are ideologically close to each other. However, because of this ideological
similarity in party preferences, the party identification of the Dutch can also be considered ideological identification (Ventura, 2001).
As De Vries, De Graaf & Eisinga (2008) addressed that it is better to include the recalled political preference of the father, than to leave it out, I will include those in my data analysis. Effects of time on the recall of the respondents will not be very severe, since the surveys were all conducted immediately after the elections.
Therefore, H1 will be as follows:
H1: Respondents who can recall the political preferences of their parents and are not
politically active on the internet, are more likely to reflect the same ideological identification as their parents, than the respondents who do not know this and are politically active on the internet.
Since there has to be a certain extent of political knowledge in order for a person to search for political information or to be politically engaged, I expect that respondents from highly politicized families will be more politically active on the internet, than respondents from less politicized families.
Therefore, H2 will be as follows:
H2: When a respondent regularly talks about politics with his parents, he or she is more likely to be politically active on the internet, than a respondent who does not.
Methods
Sample
To test the functions of the internet and of intergenerational transmission in the political socialization of today’s young adolescents, I will use a non-experimental comparative research design. The data were collected through a cohort study amongst a representative sample of Dutch adolescents, aged 15-18 years old, conducted by opinion research institute Intomart GfK. The survey asked them about their use of the internet and other media, as well as their political views and interest. Also, questions on the political preferences and behavior of their parents were included. Not every question was asked in every wave. The first wave was conducted immediately after the Dutch governmental elections of June 2010, with 1653 participants, of which 1186 participated in the second wave as well. For the second wave, conducted in June 2012, 200 extra participants were added to the sample. Only participants who had taken part in both the first and the second wave, were invited to participate in the third, and last, wave. This wave, conducted in June 2012, had a total of 888 respondents. For feasibility reasons, I will only use the data from the first wave of the survey.
Operationalization
The extent to which the respondents use the internet for political purposes, is measured through the computation of multiple questions regarding political activities (Which of these activities have you ever executed? “Posting a politically or societal relevant video or text on
your Hyves or Facebook account”, “Chatting or twittering about politics”, “Signing an online petition”, “Partake in an online political discussion”, “Use stemwijzer.nl, kieskompas.nl or a similar website”, “Start an online discussion about politics”, “Start an online petition”, “Forward an e-mail with a political message”, “write a blog post about a political or societal subject” and “Send an e-mail to a politician or other political organization”; Cronbach’s Alpha: .78).
All items on are measured on a three-point scale: 1, Never; 2, Sometimes; 3, Often.
Intergenerational transmission is measured through several questions asking the respondent about the party preference of the father, whether parents try to interest him or her for politics and how often his or her family talks about politics (“How often do you talk about politics with your parents?”, “Do your parents try to interest you for politics?” and “Do you know what the political preference of your father are?”; Cronbach’s Alpha: .37). Despite the
reliability of these measure being not ideal, they will be used in the analysis. All items for this variable are measured on, or recoded to, a binary scale: 0, No; 1, Yes.
The effect that political use of the internet and intergenerational transmission have on the political socialization of the respondents, is measured by whether the respondents have the same ideological preference as their father. This dependent variable is coded on a binary scale; 0, not the same preference; 1, same preference.
Covariates
In this analysis, the covariates education, social class, political interest and overall internet use are used. The first three covariates are all measured by single question, which the respondent
reported him- or herself. Overall internet use is measured through multiple questions on the usual online behavior of the respondents (Cronbach’s Alpha: .68).
Data analysis
Because the dependent variable of H1 has a dichotomous character and because not all independent variables are normally distributed, the first hypothesis will be tested by using a logistic regression model. This type of model is particularly suited for analyses in which the goal is to predict the membership of either one of two categories. In this case; the respondent having the same ideological identification as his or her father, coded as 0 in the data, or not having the same ideological identification, coded as 1.
In order to test the second hypothesis, a Mann-Whitney U-test will be conducted. This statistical test is particularly well suited for the hypothesis, since the variables are non-parametric.
Results
H1
The first hypothesis tested if knowledge of the father’s ideological preference, talking about politics with parents, social class, education, overall internet use and overall political interest, had an effect on whether the respondent would have the same ideological preference as his or her father. The logistic regression model which was used, indicated that the results are not statistically significant (χ2
Table 1 – Description of the data set used for the logistic regression
Same ideology?
Total sample (N)
Does not talk about politics w/parents (0) Talks about politics w/ parents (1) Yes (0) 615 338 277 No (1) 1035 546 489 Total 1650 884 766
According to the model, the log of odds of a respondent having the same ideological
preference as his or her father, was not related to talking about politics with family (p = .23) nor knowledge about the ideological preference of the respondent’s father (p = .269).
Furthermore, no connections were found for education level (p = .24), overall internet use (p = .667), overall political interest (p = .318) and social class (p = .780). This can be seen in table 2.
Table 2 – Logistic regression analysis of the similarities between the ideological preferences of respondents and their fathers.
Predictor ß S.E. Wald’s χ2 df p Exp(B)
Constant .772 .666 1.344 1 .246 2.164 Political Talk (0=no; 1=yes) -.139 .117 1.402 1 .236 .871
Knowledge of father’s ideology (0=no; 1=yes) .129 .116 1.223 1 .269 1.137 Internet use -.038 0.91 .174 1 .667 .963 Education level - - 2.849 2 .241 - Political interest - - 7.032 6 .318 - Social class - - 1.761 4 .780 -
The Hosmer-Lemeshow goodness-of-fit-test shows that the model fits the data well (p > .05). This is shown in Table 3.
Table 3 – Goodness-off-fit-test χ2 df p Hosmer & Lemeshow 6.082 8 .638
H2
From the data, which are illustrated by table 5 and table 6, it can be concluded that the
respondents who regularly talk about politics with their parents, do not display a significantly higher use of the internet for political activities (U = 337926, p = .85). Therefore, the null hypothesis can not be discarded.
Table 5 - Descriptives of the Mann-Whitney U-test on whether respondents who talk about politics at home, are also more politically active online.
How often do you talk about politics w/ your parents?
N
Are you politically active Never/Sometimes (0) 884
on the internet? Regularly/Often (1) 765
Total 1649
Table 6 – Test statistics for the Mann-Whitney U model
Are you politically active on the internet?
Mann-Whitney U 337926.000
Wilcoxon W 630921.000
Z -.188
Asymp. Sig. (2-tailed) .851
This data analysis was conducted in order to gain more insights in the status quo of the parental influence on young adults’ ideological identifications and their use of the internet for political activities. Given the growing concerns about politically ‘tuned-out’ youths, the decline of traditional media use and the lack of existing research on this situation, the choice for this data analysis was a logical one. Apart from researching the above-mentioned subjects, I have also controlled for variables such as education level, social status, overall internet use and overall interest in politics. Tests for the goodness-of-fit proved that the data fitted the model well.
The results did not support the hypotheses. No significant effects of parental influence on either ideological identification or political internet use were found.
Despite the model fitting the data well, there are critical points in this research that might have influenced the outcomes. For feasibility reasons, only the first wave of the data was used. However, a more longitudinal examination would have given more insight in the development of parental influence.
Furthermore, not all operationalisations achieved the ideal reliability value, making the executed tests less generalizable. This issue could be solved by creating a survey which is exclusively focusing on the subjects which are being investigated (though I am very grateful for the fact that Judith Möller provided me with the data set).
Finally, another suggestion for future research is that party preference is analyzed, instead of ideological preference. As stated in the theory before, children are not always able to
distinguish between left- and right wing, whereas they are more likely to be able to do this with political parties.
Conclusion
There is a serious need for more research on whether parents are still of influence on the ideological choices of their children. Most research is so old, that it does not even consider the use of the internet as a political mass medium. According to this data analysis, there is no significant effect of the influence of intergenerational transmission on the ideological preference, nor on the political use of the internet by young adults. However, there are some serious limitations to this research, which damage the validity of the results. Future
researchers should focus on long term effects and look at party preference instead of ideological preference.
Bennett, W. L. (2008). Changing citizenship in the digital age. Civic life online: Learning
how digital media can engage youth, 1, 1-24.
CBS (2014). ICT gebruik van personen naar persoonskenmerken.
http://statline.cbs.nl/StatWeb/publication/?VW=T&DM=SLNL&PA=71098NED&D1=0-
14,33-34,41-47,55-59,65-84,114-133&D2=7-8&D3=a&HD=090817-1415&HDR=G1,G2&STB=T
Converse, P. E., & Dupeux, G. (1962). Politicization of the Electorate in France and the United States. Public Opinion Quarterly, 26(1), 1-23.
Delli Carpini, M. X. (2000). Gen. com: Youth, civic engagement, and the new information environment. Political communication, 17(4), 341-349.
Flanagin, A. J., & Metzger, M. J. (2000). Perceptions of Internet information credibility. Journalism & Mass Communication Quarterly, 77(3), 515-540.
Glasford, D. E. (2008). Predicting voting behavior of young adults: The importance of
information, motivation, and behavioral skills. Journal of Applied Social Psychology, 38(11), 2648-2672.
Jennings, M. K., Stoker, L., & Bowers, J. (2009). Politics across generations: Family transmission reexamined. Journal of Politics, 71(3), 782-799.
Johnson, T. J., & Kaye, B. K. (2000). Using is believing: The influence of reliance on the credibility of online political information among politically interested Internet
users. Journalism & Mass Communication Quarterly, 77(4), 865-879.
Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological
review, 80(4), 237.
Lau, R. R., & Redlawsk, D. P. (2001). Advantages and disadvantages of cognitive heuristics in political decision making. American Journal of Political Science, 45(4), 951-971.
Metzger, M. J. (2007). Making sense of credibility on the Web: Models for evaluating online information and recommendations for future research. Journal of the American Society for
Information Science and Technology, 58(13), 2078-2091.
Möller, J (2013). Growing into citizenship: The differential role of the media in the political socialization of adolescents.
Nieuwbeerta, P., & Wittebrood, K. (1995). Intergenerational transmission of political party preference in the Netherlands. Social Science Research, 24(3), 243-261.
Norris, P. (2004). Young people & political activism. Harvard University, John F. Kennedy
School of Government.(32p).
Percheron, A., & Jennings, M. K. (1981). Political continuities in French families: A new perspective on an old controversy. Comparative Politics, 421-436.
Popkin, S. L. (1994). The reasoning voter: Communication and persuasion in presidential
campaigns. University of Chicago Press.
Rideout, V. J., Foehr, U. G., & Roberts, D. F. (2010). Generation M [superscript 2]: Media in the Lives of 8-to 18-Year-Olds. Henry J. Kaiser Family Foundation.
Sigel, R. (1965). Assumptions about the learning of political values. The annals of the
Ventura, R. (2001). Family political socialization in multiparty systems. Comparative
Political Studies, 34(6), 666-691.
De Vreese, C.H. & Möller, J. (Forthcoming). Communication and Political Socialization.
De Vries, J., De Graaf, N. D., & Eisinga, R. (2009). Biases in the effects of family
background characteristics on voting preference: The Dutch case. Electoral Studies, 28(2), 204-217.
