Record: Elias Dinas, “Why Does the Apple Fall Far From the Tree? How Early Political Socialization Prompts Parent-Child Dissimilarity” in British Journal of Political Science (April 2014): 1-26.
Introduction: This article is not explicitly about home education. Its central question, however, is an important one for many home educators. Many parents turn to homeschooling out of a desire to limit their children’s exposure to alternative views of life, hoping to secure allegiance from their children to the same religious and political values they hold themselves. Dinas’ argument, if correct, suggests that such parents are actually engaging in behaviors that are likely to promote their children’s rebellion against parental values once the children reach young adulthood.
Summary: Dinas, a lecturer in the School of Politics and International Relations at the University of Nottingham, here presents sophisticated statistical data to defend the thesis that children whose parents are more politically activist and partisan tend to move in young adulthood away from their parents’ views at a greater rate than do children whose parents are less doctrinaire.
He begins with a review of the literature on political socialization. Early studies found that the more concerted a parent’s efforts to inculcate political values in their offspring, the more likely those offspring were to take their parents’ views as their own. But these early studies had mostly been done on older children, not young adults. Anecdotal and some qualitative studies were finding anomalies once these same children got older. Many of them seemed to be repudiating the very parental views they had believed in as teenagers.
This qualitative and anecdotal literature has posited two theses:
- Parents who are more politically activist and partisan tend to produce offspring who at first become like their parents. By their teen years these children are much more politically aware than are children whose parents are less partisan. They hear their parents talking and adopt their views.
- But the same dynamic that leads these children to heightened political awareness also means that as they grow up they are more attuned to political debates, more interested in alternative points of view. Because of this many of them end up switching sides in young adulthood.
But why do some switch and some don’t? The anecdotal and qualitative literature has found that exposure to alternative points of view in young adulthood occurs primarily in two places. The first is college. It turns out that there is something to the complaint among many conservative parents that college turns their kids liberal. In many cases it does. The second venue where young adults are often exposed to alternative ideologies is work. A child who has spent her life being raised in an environment where anyone who believed differently than the parents was automatically suspect might suddenly find herself working next to someone who holds such views and discover that the enemy isn’t so bad after all.
A final factor that the qualitative and anecdotal literature has found to be important for reorientation of ideology during early adulthood is the broader historical context. Major events like the Civil Rights Movement, the Vietnam War, or, more recently, 9/11 or the push to legalize gay marriage, can often affect the political and ideological orientation of an entire generational cohort.
Having summarized all of this, Dinas gets to his actual question. Can we move beyond anecdote and small-scale qualitative life stories to get some actual statistical data based upon large samples to see if the trends this literature has uncovered are actually happening on a grand scale? Is the phenomenon of politicized families producing children who early on identify strongly with their parents’ views but shift to the other side when influenced by college, workplace, or current events something that can be documented scientifically, or are the examples of such behavior in the qualitative literature just outliers?
To try to get some quantitative leverage on this question Dinas used a data set called the Jennings-Niemi-Stoker Parent-Youth Socialization Study. This American longitudinal study surveyed the political attitudes of parents and their high school senior children in 1965 and then re-interviewed the children in 1973, 1982, and 1997, by which year the children were in their 50s. The sample size of the original 1965 study was 1,110 parents and 1,101 children, and these numbers stayed fairly constant in the later surveys.
Engaging in some sophisticated statistical analysis that I’ll confess I did not fully understand given my own limited background in quantitative methodology, Dinas did indeed find that the children in 1965 whose parents were the most politically zealous were most likely themselves to be political and to share the views of their parents. He likewise found that these same children were indeed more likely to abandon their parents’ beliefs in young adulthood, though the data does not allow him to say with certainty what factors contributed to this transformation.
There is an obvious potential problem for Dinas, however, in using data about children who came of age in the late 1960s and 1970s, a time of tremendous social and cultural upheaval in the United States. Dinas thus wants to find other data sets from more settled times to see if the generalization holds even then. He finds two. The first is another part of the Parent-Youth Socialization study. By 1997 the children of 1965 were old enough to have teen-aged and young adult children themselves, and the 1997 study interviewed this third generation as well (N=769). Though he didn’t have longitudinal data on this younger generation, Dinas split them into cohorts of younger children (age 16-21) and older children (age 23-29). If his thesis is correct, he’ll find that the younger cohort whose parents were more political will be more political themselves and will agree with the parents, but that the older cohort whose parents were more activist will be more likely than others to have moved away from their parents’ views. And that is exactly what he found.
A final source of data Dinas considers comes from the British Household Panel Study (BHPS), which followed a sample of 5,500 households, interviewing them annually from 1991 to 2008. Dinas hones in on individuals in the sample who were aged 16 to 20 in 1991 who still lived at home (N=710). Again he finds that as they move into early adulthood “offspring from more politicized homes are more likely to deviate from their parents’ partisan influences.” (p. 21) He notes that the change occurs mostly between ages 21 and 28.
Having surveyed all of this data, Dinas concludes, “although children who grow up in politicized homes are more likely than those from less politicized or apolitical homes to adopt their parents’ party identification as adolescents, they are also more likely to revise their partisan affiliations when they experience politics as adults.” (p. 21)
Appraisal: As noted above, I do not have the statistical chops to parse the complicated data Dinas presents here. I’ll just have to trust him on it. His graphs make it clear that the more politicized the family, the more likely the dynamics he describes here play out, but actual percentages I found hard to discern. At one point, for example (p. 20) he seems to be saying that children growing up in politicized environments are twice as likely as those who do not to adopt their parents’ views as adolescents. At another he finds that the single variable of objecting to the Vietnam War among young people accounts for 20% of the change in party identification from parent to child in highly politicized families but “exerts no significant effect” on offspring of less politicized families. (p. 13) And at another he finds that family politicization is responsible for 3% of the change in attitude toward liberal views on school integration and 8% on school prayer. (15) I personally found it hard to generalize from claims like these just how many children of highly ideological parents are likely to move away from the parental views as young adults. 50%? 20%? 3%? This concern may reflect more on my own statistical illiteracy than on Dinas’ presentation, though I do think he might have been clearer on this central question.
A more significant critique occurred to me as I was reading through his description of his first data set. The marker Dinas is using to determine if a child moves away from the views of her parents is party identification. But during the very time period of the survey, as many, many political scientists have described, the American parties were undergoing profound realignments. Thus when a Southern twenty-something switches from his parents’ Democratic party to the Republicans in the 1970s it is not necessarily because he as suddenly become a Yankee sympathizer. It is because the parties themselves shifted, as the Democrats moved from being for states’ rights (code for accepting segregation) to being for Civil Rights and as Republicans moved from being the party of Lincoln to the party of Nixon’ s”southern strategy” and Reagan. The story of this party realignment is long and complicated, and the generation whose voices are preserved in the interviews on which Dinas relies played the key role in speeding it along, but it does call into question the data Dinas presents. Many of these individuals would no doubt say, as Reagan himself did in 1962, “I didn’t leave the Democratic Party. The party left me.”
Having said that, Dinas’ later evidence, drawn from the late 1990s and from Great Britain, seems less open to such a critique, so perhaps he is correct after all. Currently homeschooling research on this question is also anecdotal and qualitative. Any number of memoirs written by children growing up in very ideological families describe their (often painful) move away from their parents’ views. Rob Kunzman’s book-length qualitative study Write These Laws on Your Children found that as they age many homeschooled children become less ideological than their parents. Braden Hoelzle’s recent study of four young adults likewise found a general drift away from parents’ strident views. And then of course there are the websites run by disgruntled former homeschoolers who are taking the lead in alerting the public to abusive situations and advocating for stiffer regulations. For years now I’ve wondered what percentage of children homeschooled in an ideologically rigid Christian Fundamentalism were likely to keep the faith and what percentage were likely to rebel. Dinas’ article is the closest thing I’ve yet seen to an answer. I still can’t say with any confidence that x percent of homeschoolers raised in very ideological contexts are likely to reject their parents’ views, but I can at least say that the more rigid and doctrinaire the ideology, the more likely rebellion will occur.
Milton Gaither, Messiah College