Background: To translate the Modified Swedish version of Postural Assessment Scale for Stroke (SwePASS) into Norwegian (SwePASS-NV) and assess its intra-and interrater reliability. Methods:Translation and cross-cultural adaptation was accomplished according to international guidelines.Forty-seven adults, suffering a stroke within the last six months, were strategically recruited and allocated into groups according to five different levels of function. Their performance on SwePASS-NV was video-recorded and then scored by three raters twice, with a minimum of 4 weeks between sessions. Relative reliability was investigated by calculating intraclass correlation coefficients (ICC 1.1 and ICC 3.1 ). Absolute reliability was analysed using within-subject standard deviation (S w ) and smallest detectable difference (SDD). For individual items, Cohen's kappa (k) and percentage of agreement were calculated. Results: The study showed excellent intra-and interrater reliability (ICC 1.1 and ICC 3.1 ≥.99) for SwePASS-NV. K-values for the individual items ranged between 0.68-1.00.Percentages of agreement ranged from 77-100%. SDD at 95% confidence interval was ≤ 2 points for intrarater assessments and ≤ 3 points for interrater assessments. Conclusion: This study demonstrated very high intra-and interrater reliability of the SwePASS-NV in adults within the first six months after stroke. All items showed very high or high agreement.