Short Biography

Melanie Revilla is a postdoctoral researcher at the Research and Expertise Centre for Survey Methodology (RECSM) and an adjunct professor at Universitat Pompeu Fabra (UPF, Barcelona, Spain). She received her PhD from UPF in 2012, in the areas of statistics and survey methodology, under the supervision of professors Willem Saris (UPF) and Peter Lynn (Essex University). She graduated in statistics and economics at the 'Ecole nationale de la statistique et de l'administration économique' (ENSAEParitech, Paris, France) and holds a Master of Science in Economics from the Barcelona Graduate School of Economics (Spain). She has published around 30 papers or book chapters related to the following topics: measurement errors, data quality, questionnaire design, web surveys, mobile devices, experiments, passive measurement.

https://www.upf.edu/web/survey/entry/-/-/67233/adscripcion/melanie-audreyrevilla

Course description

This crash-course to correction for measurement error in survey data starts with defining measurement error and highlighting why it is important to take it into account in statistical analyses using survey data.

We will move on to discuss the challenges of estimating measurement error, briefly revising different procedures. We will focus on the approach behind the prediction of measurement error using the free-access Survey Quality Predictor (SQP) software (http://sqp.upf.edu/).

Once we know what measurement error is, why it is important to correct for it and how we estimate it, we can learn the basics of correction for measurement error in regression analysis.

Software

We will use R software environment for statistical computing (https://www.rproject.org/). We will use “lavaan” (http://lavaan.ugent.be/) and “semTools” (https://github.com/simsem/semTools/wiki) packages. We will use SQP software to predict measurement error http://sqp.upf.edu/.

Prerequisites

Basic R knowledge is desirable, although not mandatory.

Useful to revise what old friends variance and variance-covariance matrix mean. 

References

Course largely based on:

Saris, W.E. and Gallhofer, I.N. (2014) Design, Evaluation and Analysis of Questionnaires for Survey Research. Second Edition. New York: Wiley.

Additional reference:

Saris, W.E. and Revilla, M. (2015). "Correction for measurement errors in survey research: necessary and possible". Social Indicators Research.

We will use data from the European Social Survey http://www.europeansocialsurvey.org/