Christof Wolf

Volume 52 - Article 32 | Pages 1037–1050

This study investigates researcher variability in computational reproduction, an activity for which it is least expected. Eighty-five independent teams attempted numerical replication of results from an original study of policy preferences and immigration. Reproduction teams were randomly grouped into a ‘transparent group’ receiving original study and code or ‘opaque group’ receiving only a method and results description and no code. The transparent group mostly verified original results (95.7% same sign and p -value cutoff), while the opaque group had less success (89.3%). Second-decimal place exact numerical reproductions were less common (76.9 and 48.1%). Qualitative investigation of the workflows revealed many causes of error, including mistakes and procedural variations. When curating mistakes, we still find that only the transparent group was reliably successful. Our findings imply a need for transparency, but also more. Institutional checks and less subjective difficulty for researchers ‘doing reproduction’ would help, implying a need for better training. We also urge increased awareness of complexity in the research process and in ‘push button’ replications.

Three main approaches dominate the field of measuring individual social capital: the name generator, excelling in measuring strong ties; the resource generator, focusing on capturing social support; the position generator, reflecting social capital embedded in social hierarchies. Each of these measures has its unique scientific merit making it tempting to use all of them in the same study. Because these measurement instruments typically consist of a multitude of items this often will not be an option. Therefore it is of great importance to understand the core concepts behind these measurement approaches, their different variants and their strengths and weaknesses. While aiming at the greatest possible precision of measurement survey practitioners, nonetheless, face severe limitations, in particular concerning length and mode of the planned survey. Based on our experience of having implemented social capital measures in multiple surveys, we discuss different measures that balance difficulty and performance.

We assess the relative and joint contributions of genetic and environmental factors on health during childhood and assume that parental resources are part of the environmental factors shaping children's health. We discuss theoretical background and empirical evidence concerning the effects of parental resources and heritability on children's health. Based on these findings we formulate six hypotheses guiding our empirical analysis, using data from TwinLife, a nationally representative sample of same sex twin pairs in Germany. We analyze self-rated health of 1,584 twin pairs aged 4–18. We did find strong support for the idea that parental resources influence children's health: household income and fathers' education consistently show positive effects. In contrast to our expectation, we did not find that genetic factors influence the health of well-off children less than the health of children living in families with lower SES. We also did not find that the genetic influence on health increases during childhood and adolescence. On the contrary our results indicate that the role played by genetic factors diminishes whereas environmental factors gain importance for health of children while growing up. This finding is good news for those interested in improving health chances of children from lower SES backgrounds because it demonstrates the malleability of children's health.