The aim of the present study was to screen the effects of the formulation variables - $POLYOX^{(R)}$ molecular weight $(X_1)$, the ratio of $POLYOX^{(R)}$/$Avicel^{(R)}$ PH102 $(X_2)$ and the amount of $POLYOX^{circledR}$ and $Avicel^{circledR}$ PH102 $(X_3)$, hardness $(X_4)$, HPMCP amount $(X_5)$, $Eudragit^{(R)}$ L100 amount $(X_6)$, and citric acid amount $(X_7)$ - on the paroxetine hydrochloride release from $POLYOX^{(R)}$ matrix tablet using the Plackett-Burman screening design. Paroxetine hydrochloride matrix tablets were prepared according to a 7-factor-12-run statistical model and subjected to a 8-h dissolution study in Tris buffer at pH 7.5. The regression results showed that $POLYOX^{(R)}$ molecular weight $(X_1)$ and $POLYOX^{(R)}$/$Avicel^{(R)}$ PH102 ratio $(X_2)$ had significantly influence on the drug release mechanism and drug release rate as main effects. Hardness $(X_4)$ had an insignificant effect on the drug release mechanism but a significant effect on the drug release rate. On the other hand, HPMCP, $Eudragit^{(R)}$ L100 and citric acid had an insignificant effect on the both responses. The information obtained by screening design study can be expected to be useful for further formulation studies.