Mueller & Dweck (1998) Classic Study: The Undermining Effects of Intelligence Compliments

One of the most influential papers by Carol Dweck and her colleagues is the Mueller & Dweck (1998) paper entitled Praise for Intelligence Can Undermine Children's Motivation and Performance. Over the years, this study has been criticized or disputed from various quarters, sometimes largely justified, sometimes largely unjustified. Knowing these criticisms, I argue that Mueller & Dweck (1998) is a classic publication that has lost little or nothing of importance. Here you can read a brief description of that research.

Background and purpose of the study

Context: popularity of the self-esteem movement: Mueller and Dweck did their research on the height of the so-called self-esteem movement. Supporters of this movement believed it was good to praise children for their abilities (such as their intelligence). This would improve their performance, their enjoyment and their motivation. Several prominent scientists adhered to this idea and so did many parents.

1st expected disadvantage of intelligence praise: focus on performance goals: Mueller and Dweck, however, expected that praising abilities such as intelligence could have several negative effects. The first negative effect is that it would evoke a performance goal orientation in them. By this is meant that they would become focused on showing their capabilities. Research at this time had already shown that such a focus on performance goals was associated with negative effects such as avoiding challenges, self-doubt, negative feelings and poorer performance after failure.

2nd expected disadvantage of intelligence praise: ability attribution after failure: A second negative effect that Mueller and Dweck expected from intelligence praise was that children would associate success with intelligence and failure with a lack of intelligence. According to this way of thinking, intelligence would be easy to read from how well you are doing a task. As soon as children made mistakes in a task, their tendency would increase to think: I apparently lack the capacities for this. This way of thinking makes one more likely to give up when doing new, difficult things.

The research

Mueller and Dweck conducted six studies in which they investigated the effects of intelligence praise and effort praise. Four of those studies worked according to a three-round design (the two remaining studies examined some additional detailed aspects).

Round 1: Easy Task: In the first round of the task, students in the highest grade of primary school were presented with 10 fairly simple problems from a well-known intelligence test for children (Raven Matrices). Because the exercises were relatively simple, almost all children scored well in this round.

Feedback: 3 conditions: Then they received feedback. There were three experimental conditions. In the first condition, the students received intelligence praise (“That's a really good score. You must be smart at these problems.” In the second condition, the students received effort praise (“That's a really good score. You must have worked hard at these problems. ”In the third condition (the control group), the children were only told: “That's a really good score.”.

Round 2: difficult task: In round two, the students were given 10 Raven Matrices exercises that the researchers knew were too difficult for children of this age. So they knew that students would make many mistakes and they did. They had few or no correct statements and were told this. The researchers wanted to find out how this failure would affect their motivation for a next task and their performance on that next task.

Round 3: Easy Task: The third round consisted of 10 tasks that were as easy as in round 1. Mueller and Dweck were particularly interested in what the post-failure performance would look like for the three groups. In other words, did students who received different types of feedback really score differently?


The students who were praised for their intelligence appeared to be relatively focused on achievement goals. They appeared to prefer to continue with simple tasks in which they would be successful and thus show their intelligence. This applied to a lesser extent for students from the control group and to a much lesser extent for students from the group of students who had received effort praise. These students preferred difficult tasks from which they could learn.

There were more differences. After round 2 (failure), the students who were praised for intelligence also had less pleasure in the task, perceived themselves as less intelligent, and performed less well than children who received effort praise. Students in the control group were somewhere in between these two groups. The figure below shows this.

Finally, it was also true that children who were praised for intelligence, saw intelligence more as a fixed trait (this is a fixed mindset). Children praised for effort saw intelligence more as something that can be developed (a growth mindset). Also, children who received intelligence praise more often lied about their score than children who received effort praise.


Praising abilities such as intelligence led to several negative effects compared to the control group. Praising for effort led to several positive effects compared to the control group. Taken together, this not only suggests that intelligence praise is unwise but also effort praise is sensible.