Genetic influence on political discussion: Results from two twin studies

ABSTRACT

Two twin studies are used to explore genetic influence on political discussion. Results from both studies demonstrate latent genetic traits account for, on average, 40.63% of variance in traditional and online political talk, discussion with agreement and disagreement, and political conflict avoidance. Taken together, the findings suggest a heritable genetic mechanism may partly explain why individuals vary across multiple dimensions of political discussion and differentially experience discussion effects. Implications for the political discussion effects literature and for reconceptualizing the etiology of political discussion are discussed.

KEYWORDS:

• Political discussion
• individual differences
• behavior genetics
• twin study
• genes

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Chance York (Ph.D. Louisiana State University) is an Assistant Professor in the School of Journalism and Mass Communication at Kent State University.

Notes

1. Polderman and colleagues’ meta-analysis of over 17,000 traits taken from more than 2700 twin studies gathered from respondents in 39 countries showed genetics explain, on average, 49% of trait variance.

2. It should be recognized that more recent twin research corroborates earlier studies by Beatty, McCroskey and colleagues in the “communibiological” paradigm. This earlier work suggested genes affect personality and temperament, which in turn influence communication traits such as verbal aggressiveness (Beatty & McCroskey, 1997), tolerance for disagreement (McCroskey, Heisel, & Richmond, 2001), communication apprehension (Beatty, McCroskey, & Heisel, 1998), communication adaptability (Beatty et al., 2001), and interpersonal affiliation (Beatty, Heisel, Hall, Levine, & La France, 2002).

3. The TDF is held annually in the town of Twinsburg, Ohio. Twinsburg is located near the cities of Akron and Cleveland, Ohio. For more information, see: http://www.twinsdays.org/.

4. For a partial list, see http://www.twinsdays.org/research-opportunities/. Private organizations such as Proctor & Gamble and University Hospitals have conducted research at the TDF. Academic researchers from Kent State University, Case Western University, the University of California-San Diego, and West Virginia University have also collected data at the TDF. See Cesarini et al. (2008) and Horvath (1995) for additional discussion on the use of TDF data.

5. The only demographic variables collected for the 2017 TDF were sex, age, and education. Additional demographics would have been useful to describe the sample. However, demographics are rarely collected or reported in twin research unless they are integral to the analysis – e.g., if they are explicitly modeled as a covariate in multivariate analysis. Sex and age are the exceptions since they can be used to test for biologically based differences that may impact ACE estimates.

6. The MNTPS data come with the following acknowledgement: “The data employed in this project were collected with the financial support of the National Science Foundation in the form of SES-0721378, PI: John R. Hibbing; Co-PIs: John R. Alford, Lindon J. Eaves, Carolyn L. Funk, Peter K. Hatemi, and Kevin B. Smith, and with the cooperation of the Minnesota Twin Registry at the University of Minnesota, Robert Krueger and Matthew McGue, Directors.”

7. It is important to note that Funk et al.’s (2010) unpublished conference paper presented an analysis of two of three discussion items available in the MNTPS: political discussion frequency and conflict avoidance. Discussion with disagreement was not analyzed. As such, the present study can be considered a partial replication of Funk et al. (2010).

8. In ACE modeling, the total variance in a phenotype is explained by the three latent factors in the model. Importantly, ACE modeling has been extensively described and validated (e.g., Neale, 2009; Neale & Cardon, 1992; Neale & Maes, 2004). It represents one of the few analytical techniques used to examine the simultaneous impact of genes and environment on behavior. However, Smith and Hatemi (2013) and York (2017) discuss alternative modeling techniques using regression.

9. Put differently, ACE modeling assumes that “identical twins are not more similar on [a] characteristic simply as a function of being treated more similarly [than are fraternal twins]” (DiLalla, 2004, p. 6). If the equal environments assumption (EEA) is violated because identical twins experience more similar environments than do fraternal twins, “this violation would inflate estimates of genetic influence” (Knopik et al., 2016, p. 86), although it should be noted that the EEA has undergone extensive testing, even in the specific context of political behavior with few and random violations (see Littvay, 2012).

10. The OpenMX user guide can be found at: https://vipbg.vcu.edu/vipbg/OpenMx2/docs//OpenMx/latest/OpenMxUserGuide.pdf.

11. A series of independent samples t-tests revealed no significant differences in means between identical and fraternal twin pairs. The largest mean differences were for the Social Media Discussion (M = −0.19, t = −1.13) and the Frequently Talk items (M = 0.19, t = 1.07) in the TDF and neither were significant. All t-test results are available from the author upon request.

12. Both the TDF and MNTPS offer relatively small and demographically homogenous twin samples. For example, TDF respondents tended to be highly educated compared to the population while the MNTPS twins were all same-sex pairs over 55. Thus, it may be easier to find genetic influence on political discussion among these demographically uniform samples because fewer environmental differences leave only genetic variation to explain differences in political discussion variables.