Background: Insecure attachment in infancy is associated with a range of socioemotional problems later in life. It is important therefore to identify at-risk children so that appropriate support can be provided. However, the two most well-established measures of attachment, the Attachment Q-sort (AQS) and Strange Situation Procedure (SSP), are both time-consuming and costly to administer. The aim of this study is to create a valid, short version of the AQS using modern psychometric techniques. Method: Data was used from the NICHD Study of Child Care and Youth Development (NICHD SECCYD) (n=1,364). First, the factor structure of the AQS was explored using Q-factor analysis. Item response theory (IRT) was then used to create shortened scales containing the subset of items which provided the most information. The validity of the shortened scales was then examined. Results: Q-factor analysis indicated two clear factors relating to harmonious interaction with the caregiver and proximity-seeking behaviours. Two scales of eight items each were created based on these factors. The shortened measure showed comparable convergent, discriminant and predictive validity to the full AQS. Conclusion. This brief version of the AQS shows good potential as a screening measure for insecure attachment in infancy.