Recent research on causal inference suggests that common actions tend to be
attributed to goals, whereas difficult actions, if obstructed are attribut
ed primarily to preconditions. The present studies examine the way that the
framing of causal questions influences ratings of goals and preconditions
for common actions. The studies test the view that 'why' questions favour g
oal explanations, by presenting causal questions framed as 'why' questions
or 'explain' questions. Structured and free-response measures were used The
y show that when the question is expressed as asking why an action occurs,
goals are rated better than preconditions, regardless of the presence of ob
stacles, whereas if the question is framed as requesting an explanation of
the action, preconditions are deemed better explanations than goals for obs
tructed actions. Goals remain better explanations when the action is unobst
ructed. These findings confirm the importance of the framing of causal ques
tions for research on causal explanation, and suggest that the phrasing of
causal questions influences the focus of explanations. (C) 1998 John Wiley
& Sons, Ltd.