Tested the hypotheses that evaluatively biased language influences attitudes and that the magnitude and persistence of attitude change depends on the amount of cognitive effort. 132 undergraduates participated in the experiment, which used material focusing on the issue of restricting adolescent driving over the weekends to reduce the number of fatal traffic accidents. Results indicate that evaluatively biased language can affect attitudes. Using words that evaluate the pro-position positively and/or the anti-position negatively changed Ss' attitudes in the pro-direction and vice versa. The amount of cognitive effort affected the magnitude of attitude change."