The broad aim of this study on father-child attachment was to verify whether the Risky Situation (RS) procedure is a more valid means than the Strange Situation (SS) procedure of predicting children's socio-emotional development, and to evaluate the moderator effect of day-to-day involvement on attachment and activation. Participants were 53 father-child dyads. The RS and the SS were conducted when children were 12-18 months old to measure attachment and activation, and a questionnaire on fathering was administered at the same time. Childcare workers rated children's socio-emotional development at 30-36 months. Regression analyses revealed that the RS predicted children's socio-emotional development, while the SS did not, even when controlling for paternal involvement. This study advances the field by empirically testing the predictive relationships from attachment and activation to social emotional outcomes, and the moderator effect of fathering. The results underscore the value of the activation relationship theory and the RS procedure as a means for comprehending and capturing the essence of father-child attachment. (Contains 2 tables and 2 figures.)