The early childhood sector in Singapore has witnessed vast changes in the past two decades. One of the key policy aims is to improve classroom quality. To ensure a rigorous evaluation of the quality of early childhood environments in Singapore, it is important to determine whether commonly used assessments of quality are valid indicators across different national and community contexts. This study investigated the validity of the Early Childhood Environment Rating Scale-Revised (ECERS-R) in Singapore focused on three major aspects: response process, structural, and criterion validity. Data were collected from 108 classrooms of the main providers of kindergarten programs in Singapore. Initial analyses showed that the quality indicators in most of the ECERS-R items were disordered, which indicated problems with the hypothesized response process underlying the standard ECERS-R scoring system in which high-level indicators are only rated if low-level indicators are met (stop-scoring). To deal with the problem of the "stop-scoring" method, we used the data from a full-scoring method (in which every indicator on every item was rated) in subsequent factor analyses to investigate the internal structure of the ECERS-R. Twenty-two of the 34 items were retained and found to load on two factors, one relating to activities/materials and the other related to language/interactions. Both of the identified subscales and the combined scale showed good internal consistency. The factor score on the combined scale mirrored the ECERS-R full score and could discriminate between classrooms in the top and bottom quartiles of quality scores. Findings on the relationship between the identified factors to the teacher-child ratio and to an alternative measure of classroom quality provided further evidence for criterion validity of the ECERS-R. Implications for the future development and use of the ECERS-R are discussed.