Research Methodology and Statistics

Psychiatry and behavioral sciences present unique challenges to quantitative research sciences with underlying biopsychosocial interactions. The broad discipline of psychiatry necessitates the widespread application of statistical methods in studies that address challenging questions concerning diagnosis, treatment monitoring and outcome, identifying causal risk factors, evaluation of mental health programs, etc. while simultaneously accounting for the intricate combination of patient characteristics (i.e., risk factors, activity levels, stressful events, family relationships, signs and symptoms, duration of stay, medication effects, etc.). As the field moves toward greater implementation of evidence-based practice, knowledge of study design and measurement is becoming increasingly important. Given that statistical methods are invariably tied to research design, this chapter provides an overview of issues in study design and measurement, combined with a summary presentation of standard statistical procedures commonly used in the literature with the following objectives: (1) guiding quantitative-oriented practitioners who wish to implement statistical methods to data they have collected; (2) aiding psychiatrists who are interested in reviewing the literature and evaluating findings with promising clinical applications; (3) providing a basis for conceptual understanding of common statistical analytical techniques without reliance on mathematic formulas; (4) elucidation of the most salient issues confronted by nascent readers of the psychiatric literature with a minimal statistical background. Finally, this chapter provides some practical suggestions for assisting readers to evaluate study conclusions and weigh the practical significance for application to clinical practice.