Development and validity data of the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP)

Background: The internet provides a research opportunity for psychiatry and psychology. This article presents the development and preliminary data of a large web-survey created to study how temperament relates to other psychological measures, behavior and psychiatric disorders. Methods: We used the Affective and Emotional Composite Temperament Scale (AFECTS) to evaluate temperament and we selected several self-report instruments to evaluate behavior, psychological constructs and mental disorders. The system provides anonymous psychological (phase 1) and psychiatric (phase 2) feedback and includes questions to assess the validity of the answers. Each phase has around 450 questions. This system was broadcast utilizing Brazilian media. Results: After the exclusion of 21.5% of the volunteers (those who failed the validation questions), 41,427 participants concluded the first part of the system (mean age = 31.2 +/- 10.5 yrs, 26.9% males), and 21,836 (mean age = 32.5 +/- 10.9 yrs, 25.1% males) completed phase 2. Around 25% have received a psychiatric diagnosis from a mental health professional. Demographic and temperament profiles of those who completed either only 80 questions, only phase 1, or the whole system were similar. The rate of non-serious answers (e. g. on bizarre behaviors) was very low and congruency of answers was very high. The internal consistency of classical trait scales (TCI-R and PANAS) was high (Cronbach's alpha > 0.80) for all dimensions. Limitations: Relatively high dropout rate due to the length of the process and an overrepresentation of female, young and well-educated subjects. Conclusions: The BRAINSTEP provides valid and abundant data on psychological and psychiatric measures. (C) 2012 Elsevier B. V. All rights reserved.