BACKGROUND: Ankle proprioception deficit is a major factor causing balance dysfunction in subacute stroke survivors. However, there is still no commonly-agreed ankle proprioception evaluation method for these patients. Whether ankle proprioception tested by the active movement extent discrimination apparatus (AMEDA) when participants are standing (AMEDA-standing) or sitting (AMEDA-sitting). OBJECTIVES: This study aimed to investigate the variations in ankle proprioception measures between the AMEDA-standing and AMEDA-sitting in subacute stroke survivors and assess the test-retest reliability of these measures. STUDY DESIGN: This study analyzes sitting and standing proprioception (AMEDA scores) and their associations with static/dynamic balance in 24 subacute stroke patients using ANOVA, Pearson's correlation, and ICC analyses in SPSS 26. METHODS: The battery of tests administered included the AMEDA, the Timed Up and Go Test, the Single-leg Stance, and the Limit of Stability, as measured by the NeuroCom® Balance Manager® VSRTM. Pearson correlation was applied to discern the relationship between the acuity of ankle proprioception and balance. The intraclass correlation coefficient was used to gauge the test-retest reliability of the measures. Furthermore, an analysis of variance was conducted to scrutinize any differences between the proprioception scores obtained from the AMEDA-standing and AMEDA-sitting protocols. RESULTS: Area under the curve values for the AMEDA-sitting during the initial test and retest were 0.665 (0.090) and 0.665 (0.080), respectively. For the AMEDA-standing, the area under the curve values were 0.697 (0.069) for the initial test and 0.699 (0.075) for the retest. Evaluating the reliability, the intraclass correlation coefficient was calculated as 0.704 for AMEDA-sitting, while for AMEDA-standing, it was 0.752. Upon conducting a Pearson correlation analysis, a statistically significant relationship was observed between AMEDA-sitting and several balance variables: Reaction Time-affected, Endpoint Displacement-affected, Max Excursion-affected, Endpoint Displacement-unaffected, Max Excursion-unaffected, and Endpoint Displacement-back. CONCLUSIONS: In the context of subacute stroke survivors, both AMEDA-standing and AMEDA-sitting methodologies prove to be appropriate, demonstrating commendable test-retest reliability.