Improving statistical power in studies on event occurrence by using an optimal design

Kasia Jozwiak 
Department of Methodology and Statistics, Faculty of Social Sciences, Utrecht University

Project: Project financed by NWO

Project running from: 1 january 2009 – 1 January 2013

Supervisors: Dr ir M. Moerbeek and prof. P.G.M. Van der Heijden (Utrecht University)

Summary:
The main research question in studies on event occurrence is whether and when subjects experience a particular event, such as the onset of daily smoking or the shift to adulthood. The experience of such an event and its timing can be related to explanatory variables such as gender, socio-economic status, educational level, and, in the case of an experiment, treatment condition. Such a variable’s effect should be identifiable with sufficient probability, so the power of a study on event occurrence should be controlled in the design phase.
In studies on event occurrence subjects may be monitored continuously, or be measured at intervals. Interval measurement is often used in the behavioural sciences but sample size formulae for such trials are not readily available. The proposed research aims to remedy this deficiency by providing guidelines for the indices governing the number of subjects, the number of measurements per subject, the placement of the measurement points in time and the duration of the study. Where possible, mathematical formulae that relate sample size and duration to statistical power will be derived analytically.
Otherwise, the effect of these design factors on statistical power will be studied on the basis of simulation studies taking into account realistic conditions such as drop-out rates and the varying costs per treatment condition.
A study that is not carefully designed is a waste of resources. Therefore, ethical review committees and organizations funding scientific research frequently require research proposals to include power calculations. The proposed research will provide guidelines for efficient study-designs for use in event occurrence studies – ensuring that the financial cost and the number of subjects are minimized and sufficient power is guaranteed. From a scientific point of view this proposed research project is fundamental since it will enable future researchers to plan their research more efficiently.
Keywords: statistical power, cost-efficient designs, survival analysis, hypothesis testing.

Date of defence:   12 April 2013

Title of thesis:   Improving statistical power in studies on event occurrence by using an optmal design

ISBN:   978-90-393-5918-1