OBJECTIVE: To translate, adapt and verify the validity of the motor criteria and the construct (internal consistence, discriminatory validity, correlation with other tests and predicted validity) of the Portuguese version of the Alberta Motor Infant Scale. METHODS: Cross-cultural validation study that enrolled 21 professionals and 561 children from the South of Brazil aged zero to 18 months (291 boys). The Alberta Infant Motor Scale, the Developmental Scale of Child Behavior and a questionnaire to control variables were used. Alberta Infant Motor Scale was translated, adapted and the content validity was analyzed. Professionals were then trained and the data were collected in daycares and participants' home. Scale's objectivity, reliability, criterion and construct were analyzed. RESULTS: The Portuguese version of Alberta Infant Motor Scale contained clear and pertinent motor criteria; it was reliable (total score, p=0.88; prone, p=0.86; supine, p=0.89; sitting, p=0.80 and standing, p=0.85), it had discriminative power (typical versus atypical development; score, p<0.001; percentile, p=0.04; classification criteria, chi-square: p=0.047) and, also showed temporal stability (p=0.07; rho=0.85; p<0.001) and predictive power (p<0.001). The scale was moderately correlated with the Developmental Scale of Child Behavior (rho=0.342; p=0.03), considering that the McNemar-Bowker proof showed differences between both methods (p=0.047). CONCLUSIONS: The Portuguese version of Alberta Infant Motor Scale showed validity and reliability. It is a suitable tool for the assessment of motor acquisitions of Brazilian children, being valuable for researchers and practitioners.
validation studies; motor skills; child development
