External evaluation is often necessary for students to increase their competence level. Although they may be able to determine the correctness of their spelling words by consulting a master list, they usually need assistance in judging whether the imagery in their writing is more vivid or their argument more persuasive than previously. Teachers' feedback can direct future efforts as well as point out newly developing competencies. Knowing what they need to do to improve gives students a perception of control over achievement outcomes which enhances effort and learning. However, some kinds of evaluation are more likely to accomplish these goals than are others.

  1. Make evaluation criteria clear. Students do not feel in control of their academic outcomes when the criteria for evaluation are vague. Rubrics are often helpful. In addition to clarifying the criteria and standards for evaluation, rubrics help students make independent judgments of their work and determine on their own when they need to make corrections or revisions.
  2. Give students different ways to demonstrate what they know. When students are given diverse opportunities to demonstrate their understanding, teachers gain a more complete picture of their skills and students have more chances to demonstrate competence. Understanding of mathematical concepts, for example, might be expressed by finding correct solutions to a calculation problem, explaining why a strategy for solving a problem is efficient, representing the problem and solution in a drawing, or assisting another student.
  3. Point out what is good, right, or shows improvement. Students need external validation of the competencies they have developed, as well as information on how to improve. This is particularly true on assignments that students are likely to have difficulty evaluating, such as a task that is new for them (e.g., writing a book report or writing up a science experiment for the first time). Commending what is good or shows improvement lets students know that they are developing competencies.
  4. Provide dear, specific, and informative feedback.

    a. Avoid global, uninformative comments. "Nice job," "well done," or "good," on the cover of a paper doesn't give a student the message that her work met clear, well-defined standards. Students do not always consider such global feedback credible (Damon, 1995).

    b. Focus on the behavior, not the person. Dweck (2000) and her colleagues have found that person-praise (as well as criticism) promotes an entity theory of ability and engenders concerns about demonstrating ability. Accordingly, it makes children more vulnerable to losing their self-confidence and becoming helpless when they encounter difficulty (see also Kamins & Dweck, 1999; Mueller & Dweck, 1998). This line of research suggests that teachers should avoid comments that focus on personal traits, such as, "You are so smart." It is better to focus attention on the students' effort and behavior ("You used some really creative strategies"), or their level of mastery ("You really understand this well").

    c. Provide written, substantive comments when possible. Specific commendations (e.g., "The paper is well organized"; "The transitions are smooth"; "Your strategy is creative and shows that you really understand the math ideas"; "Your summary of the results of the experiment is clear and concise") provide clear, credible evidence of competence. Substantive and specific feedback also helps students develop criteria that they can use to assess their own skills and identify their strengths and weaknesses.

  5. Base rewards (including high grades) on achieving a clearly-defined standard or set of criteria or on personal improvement. Some of the decline in enthusiasm for schoolwork in junior high or middle school is probably caused by the stricter and more social comparison-based standards for grading. The decline in grades caused by the competitive standard lowers some students self-confidence and, consequently, their interest in learning. Indeed, the magnitude of a drop in grades following the transition into junior high or middle school is a major predictor of leaving school prior to graduation (see Eccles & Roeser, 1999).

    Ironically, a competitive standard for rewards does not necessarily build self-confidence in the "winners." This is because some high-achieving students perform at a high level, relative to classmates, without much effort and without developing their competencies. As mentioned above, the perceptions of high competence developed by students who are constantly rewarded for their schoolwork, regardless of their effort, are very fragile and fall apart in a situation in which success or rewards do not come easily.1

    Rewards, including good grades, stars, happy faces, or privileges should therefore be based on a set of clearly-defined criteria or standards, or on improvement, rather than on relative performance.

  6. Give students multiple opportunities to achieve a high grade. Teachers can allow students to keep working (within some reasonable limit) until they have achieved a satisfactory level of mastery or performance. One teacher I interviewed developed the simple but ingenious method of marking incorrect responses on written assignments with a dot. Students continued to work on assignments until all answers were correct. Dots could easily be changed to check marks, which she used to indicate correctness without leaving any evidence of the original error. Thus errors were treated as a natural step in mastering new material. Students can also be given opportunities to rewrite papers or redo assignments to achieve a higher grade or demonstrate a higher level of mastery.

    Providing substantive feedback en route to task completion is also helpful. This is a critical feature of the current popular approaches to teaching writing. Students write several drafts of their papers, concentrating on correcting different aspects of the writing (e.g., getting thoughts on paper, developing and organizing ideas, mechanics) at different points in the process. Feedback along the way provides useful corrective information and gives students a feeling of control over their performance outcomes.

  7. Teach students to celebrate their classmates' successes, at whatever level they occur. The teacher is not the only evaluator in a classroom. Classmates' reactions can be just as important, and peers can undermine the teacher's efforts to reinforce effort and improvement.

    The values of the teacher, therefore, need to be internalized by every student in the classroom. This will occur, to some degree, if the teacher models appropriate behaviors. But students can also be instructed to support the teacher's goals. They can, for example, be encouraged to applaud their classmates for significant improvements, regardless of the student's relative skill level. This conveys to students that improvement, which is genuinely available to all students, is valued by everyone, and it gives students an example of concrete behaviors that communicate this value to each other.2

  8. Minimize public evaluations. Wall charts and other public displays of students' performance may enhance the motivation of a very small group of top performers, but they can threaten and discourage other students, especially if evaluative criteria are uniform despite varying skill levels among students. For all students, these public displays orient attention to relative performance rather than personal improvement or mastery.3

    a. If evaluation is public, it must be a "fair contest." All students must have a realistic chance of "winning." This can be done by charting progress toward achieving goals that are adjusted to be realistically attainable by all students. I observed an example of this in a classroom in which paper boats with students' names on them "raced" across a blue sheet of paper on the bulletin board. The position of each student's boat was determined by how much progress that student had made in mastering an increasingly difficult set of spelling words. All students began at the same point on the sheet of blue paper, even though their actual beginning level of mastery varied considerably. Consequently, the student who won the race did not necessarily attain the highest level of mastery. Rather, the winner made the most progress from his or her personal starting point.

    b. Have students keep personal progress records. A better method than public displays for helping students monitor their progress is to have them keep records, including charts, in their desks. A personal chart, unlike a class chart, focuses students' attention on their own improvement and mastery rather than on how they compare to classmates. It is also useful for communicating student goals and progress with parents.

    c. Give students an opportunity for private interactions with the teacher. A classroom structure in which students' interactions with the teacher are either private or in small groups can also minimize the public quality of performance and evaluative feedback and therefore reduce the risk of embarrassment.

  9. Teach students to evaluate their own work. An authority figure is not always available to give students feedback on their developing competencies. Students should be encouraged and taught to evaluate their own work and to monitor their own progress. This provides them with more opportunities to experience a sense of developing competencies as well as strategies to guide their own efforts to improve. Self-evaluations also have been shown to promote self-regulation skills (Perry, 1998).

    a. Encourage students to use their own judgment. When students ask for feedback (e.g., "Is this good?"), encourage them to venture a judgment on their own (e.g., "What do you think, does it look good to you?") and to consider strategies for making a personal judgment (e.g., "Do you have any ideas about how you could figure out whether it is good, besides asking me?").

    b. Give students opportunities to check their own work. Students can check their own solutions to math problems or the accuracy of their spelling words against an answer sheet or by making comparisons with a classmate. Asking classmates to resolve discrepancies provides opportunities for collaborative learning. Students can check facts by consulting a book or using the Internet.

    c. Give students explicit instructions on how to evaluate their own work. Students learn criteria for making independent judgments in part by receiving specific, substantive feedback on their work from teachers. They also need explicit instructions and reminders of evaluation techniques. Models of good work can be useful, although they need to be used carefully to minimize the potential for students developing an unnecessarily narrow view of a good product. As mentioned above, rubrics are often helpful, such as those used to assess writing samples in standardized tests.

    d. Link evaluation criteria directly to instruction. Students will not be able to assess their own progress (and thus experience a sense of mastery) if the evaluation criteria they are given are not aligned with instruction. If, for example, math instruction is focused on providing clear and persuasive explanations for solutions, students should be given a set of criteria and strategies for determining whether an explanation is clear and persuasive. If writing instruction is focused on teaching students to use paragraphs, they should be given a set of criteria for writing a good paragraph.


lThis "unraveling" of self-confidence is common among my graduate students who previously received perfect marks without having to work very hard. I have witnessed their self-confidence evaporate with the first critical feedback they receive on their work.

2A teacher once shared with me a story about a child with learning disabilities who was reading far below the level of most of his third-grade classmates. When he announced that he had read his first chapter book, his classmates secretly asked the teacher to buy a copy that they could all sign and present to him. They wanted to celebrate this milestone with him, even though most of them had passed it long ago. This teacher had clearly created a "community of learners."

3 I experienced the negative effects of public displays of performance outcomes during a year in a French university. It was standard procedure to hand back papers and exams in the order of students' scores. The assumption, I am sure, was that the humiliation of being among the last to receive one's paper would foster future effort to avoid such humiliation. What I observed instead, in other students and myself, was withdrawal and discouragement. My solution, contrary to the instructor's intentions, was to miss class on the day that the next exam was returned.