Taking seriously the claim that twins form a natural experiment
Switch branches/tags
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
RItools @ 4f220aa


Natural Experiments in Genetics

Classical twin designs compare the behavior of identical twins (monozygotic or MZ) with fraternal twins (dizygotic or DZ) in an effort to understand the influence of genes on behavior. This research attempts to take seriously the claim that nature has created an experiment in which twin pairs are randomly assigned to share 100% of their genetic material (MZ) versus 50% of their genes (DZ). If we take this claim seriously what would we expect to see in the data, and how would we go about analyzing it? What conclusions can we safely draw, and what conclusions are outside the scope of this design? What other data do we require to make stronger statements about the interaction of genetic and environmental factors?

As a political scientist, the data in this study and the substantive questions are political in nature, but the conclusions apply to a broad range of scientific questions that appeal to the intuitive natural experiment of twinning.


The data for this study come from a survey same-sex MZ and DZ twin pairs published by the University of Nebraska-Lincoln Political Physiology Lab. The original data can be downloaded from the UNL PolPhy Lab website. As per the request of the original data collectors, I am happy to acknowledge:

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.


To build this document, you will probably need a UNIX like command line environment. It may be possible to build the document on Windows, but I don't care to find out. Additionally, you'll need the following:

  • LaTeX (and the latexmk script should be in your $PATH)
  • R. You should install the following libraries (using install.packages from the R prompt):
  • optmatch
  • lattice
  • multicore
  • make

To build the document, cd to the directory and type:

$ make

It may take a while, but make is smart and subsequent builds will probably be faster. The heavy computations are all cached and will only run if their dependencies are updated.

Copyright, Licenses, and Citation

All text and code are copyright Mark M. Fredrickson (2011). All code is licensed under a BSD License. Please see LICENSE.txt for more information. The text is not licensed.

As this is a working paper, the author would appreciate an email if the work is cited. Contact mark.m.fredrickson at gmail.com.