Aim: To provide further validation of the Environmental Audit Tool (EAT) by describing data on scores from 56 facilities and comparing the scores of facilities with a purpose-built dementia environment with those with non-purpose-built designs. Methods: Fifty-six facilities were assessed with the EAT. EAT scores for 24 purpose-built environments were compared with 32 non-purpose-built environments using a Wilcoxon rank-sum test. Results: Descriptive data on EAT scores are presented across all facilities. Facilities scored well on safety/security, familiarity, highlighting useful stimuli and privacy. Purpose-built unit scores were significantly higher than those for non-purpose-built environments for nine of 10 subscales of the EAT and the overall EAT score. Conclusion: The EAT can assess the quality of homelike environments in residential aged care facilities for people with dementia, differentiate between the quality of design in various types of facilities and provide an evidence basis for devising improvements.