Statistics in Psychiatry (Arnold Applications of Statistics Series)

This book is essential reading for practicing research workers in psychiatry and related disciplines. It is intended for clinical, behavioral or social scientists, as well as applied statisticians with limited experience in the mental health field. It is problem-based rather than technique-based, and encourages thought in the use of statistical techniques, rather than a heavy reliance on formula. Great emphasis is placed on the evaluation of reliability, validity and other characteristics of measuring instruments and the importance of measurement errors and the fallibility of all psychological assessments is discussed. Quantitative methods of disorder and the use and correct analysis of double sampling strategies in the validation of psychiatric screening questionnaires are also examined. The role and importance of measurement in scientific psychiatry is can not be over-stressed and this book is a perfect example of the interplay between the ideas typical of the 'medical model' of disease and those that come from psychometric tradition.