Objectives
The use of normative data has become well-accepted and a common strategy to interpret individual's health outcome scores, which can help in making decisions. The objectives of this study were to obtain population normative data for the domains and component summaries of the 36-item SF-36® Health Survey (SF-36), and to evaluate its reliability and construct validity.

Methods
This study was conducted using population-based data from the Welsh Health Survey (WHS; 2011–2015). This study used version 2 of the SF-36 (SF-36v2® Health Survey). The descriptive statistics and normative data for the eight domains and two summaries, physical component summary (PCS) and mental component summary (MCS), were calculated. Reliability assessment used internal consistency methods and construct validity assessment used known group comparisons and item–scale correlations.

Study design and sample
We performed a secondary analyses of data from the Welsh Health Survey (WHS).

Results
This study included 74,578 participants aged 16 years or older (53.6% were women). Participants aged 16–24 years scored higher on SF-36 scale than older groups on all domains. The SF-36 profiles by age group demonstrated lower scores for older age groups, with the most pronounced differences shown on the physical-related scales. Across the age groups, men had higher PCS and MCS scores than women. All SF-36 domains and PCS and MCS achieved a good to excellent internal consistency reliability exceeding 0.7. The scales demonstrated construct validity by showing associations with a range of factors known to be related to health.

Conclusions
This study provides SF-36 normative data for Wales based on a representative data and confirms the construct validity and reliability of the SF-36.