who endorsed learning goals agreed that their goal was to learn new things,
even if they might get confused, make mistakes, and not look smart.
The unit dealt with a scientific principle that cut across several types of
problems (i.e., pulleys, inclined planes, etc.). For the task itself, students
were trained on one type of problem (e.g., pulleys) and then given a transfer
test to see whether they could apply the same principle to another type of
problem (e.g., inclined planes). Looking at students with learning goals
versus performance goals, we found that even though both groups of stu-
dents learned the material equally well, students with learning goals: (a)
produced significantly more written work during their attempts to transfer,
(b) tried more different transfer strategies, and (c) were more successful in
transferring the principle to the new task. Transfer of training is a key part
of intellectual functioning (and creativity). This study showed that students
who are in a learning mind-set are more likely to search for and to find suc-
cessful transfer strategies than are those with concerns about validating
their ability.
In another study of students’ ability to display effective strategies in the
face of difficulty, Elliott and Dweck (1988) instilled different goals in late
grade-school students as they embarked on a challenging concept-formation
task. In addition, half of the children were led to believe they had high ability
and would probably do well on the upcoming task, whereas the other half of
the children were led to believe they had lower ability at the task. The con-
cept-formation task was one that allowed the researchers to assess the sophis-
tication of students’ problem-solving strategies on each trial and so allowed
them to monitor changes in the sophistication of their strategies as students
encountered a series of more difficult problems, ones that were somewhat too
difficult for children their age (see Diener & Dweck, 1978; cf. Gholson, Le-
vine, & Phillips, 1972).
Regardless of whether students had been given learning goals or perform-
ance goals, they performed equally well on the initial trials, prior to the diffi-
culty. However, the students with learning goals were able to maintain or
even improve their problem-solving strategies over the failure trials—regard-
less of whether they believed they had high or low ability at the task. In con-
trast, unless they believed they had high ability, those with performance goals
showed a steep decline in the sophistication of their problem-solving strate-
gies over the failure trials, with many of them falling into entirely immature
and ineffective strategies. Thus students with equivalent abilities and meta-
cognitive strategies on the early trials, diverged sharply in the level of strategy
they were able to use on a more difficult task.
Do students with different goals differ in the strategies with which they ap-
proach difficult course material? Much literature suggests they do (Ames &
Archer, 1988; Elliot, McGregor, & Gable, 1999; Graham & Golon, 1991;
Pintrich & Garcia, 1991). In a recent study, Grant and Dweck (2003) tracked
- MOTIVATION AND COGNITION 43