Methodological development for questionnaire studies

Topics covered by already published papers:

• Bias in sibling and co-twin control studies due to measurement error:

Gustavson, K., Ask Torvik, F., Davey Smith, G., Røysamb, E, Eilertsen, E.M.   (2024). Familial confounding or measurement error? How to interpret findings from sibling and co-twin control studies. European Journal of Epidemiology. https://doi.org/10.1007/s10654-024-01132-6.

To increase the usefulness of the analyses, we have developed a shinyapp- SibSim - where users can estimated expected bias due to measurement error in their sibling control study. Please feel free to try it out.

• Using negative control variables to handle unmeasured confounding:

Gustavson, K, Davey Smith, G, Eilertsen, E (2022). Handling unobserved confounding in the relation between prenatal risk factors and child outcomes: a latent variable strategy. European Journal of Epidemiology, 37(5).

• Preventing bias due to selective non-response:

Gustavson, K., Røysamb, E, Borren, I. (2019). Preventing bias from selective non-response in population-based survey studies: findings from a Monte Carlo simulation study. BMC Medical Research Methodology, 19,120.

• Implications of regression towards the mean in longitudinal studies:

Gustavson, K. Borren, I (2014). Bias in the study of prediction of change: a Monte Carlo simulation study of the effects of selective attrition and inappropriate modeling of regression toward the mean. BMC Medical Research Methodology, 14,133.

We are currently working on the following topics:

• Bias in factor analysis of psychological data: How to interpret divergent results from different studies.

To increase the usefulness of our findings for researchers, we are also working on R package for factor analyzing personality data.

• Measurement error in contingency tables.

The project includes collaboration between researchers at different institutions and collaboration with other research projects.