Asch devised a procedure to assess to what extent people will conform to the opinion of others, even in cases where the answer is unambiguous.
Group size - he varied the number of confederates from 1-15 (so total group size was 2-16). Conformity increased with group size but only to a point. With 3 confederates, conformity to the wrong answer rose to 31.8%. The presence of more confederates made little difference; conformity rate levelled off. This suggests must people are sensitive to the views of others; one or two confederates caused change in opinions.
Unanimity - he introduced a confederate who disagreed with the other confederates. In one variation of the study they gave the correct answer and in another he said a different wrong answer. The participants conformed less often with the presence of the dissenter. The rate decreased to less than 1/4 of the level it was when the majority was unanimous. The dissenter allowed the participant to behave more independently; even in the second variation. This suggests that the influence of the majority depends on being unanimous to a large extent.
Task difficulty - he increased the difficulty of the line judging task by making the standard and comparison lines more similar lengths; therefore it became harder for participants to genuinely see the difference between the lines. Conformity increased; people look to others for guidance and assume they're right (informational social influence.) since they don't know any better.