ATD, association for talent development

ATD Blog

What Do You Know: Does Feedback Improve Learning?

By

Wed Jan 25 2017

What Do You Know: Does Feedback Improve Learning?
Loading...

Content

Many learning practitioners find the topic of instructional feedback confusing, so I thought it might be a good topic review. Feedback is quite complex, so let’s start off with a basic question: Does feedback improve learning? You might think the answer is obvious, but actually, it’s more complicated than many L&D practitioners might realize.

Many learning practitioners find the topic of instructional feedback confusing, so I thought it might be a good topic review. Feedback is quite complex, so let’s start off with a basic question: Does feedback improve learning? You might think the answer is obvious, but actually, it’s more complicated than many L&D practitioners might realize. 

Content

To understand feedback, let’s define what it is and what it should do. The kind of feedback I’m talking about is written or verbal responses to answers or performance on instructional questions or activities. In Table 1, I show examples of questions and activities where we would typically supply feedback.

To understand feedback, let’s define what it is and what it should do. The kind of feedback I’m talking about is written or verbal responses to answers or performance on instructional questions or activities. In Table 1, I show examples of questions and activities where we would typically supply feedback.

Content

Table 1. Typical Questions and Activities Where Feedback May Be Given

Table 1. Typical Questions and Activities Where Feedback May Be Given

Content

Content

Content

Questions

Questions

Content

Activities

Activities

Content

Oral response

Oral response

Content

Case studies

Case studies

Content

Written response

Written response

Content

Simulations

Simulations

Content

Multiple choice

Multiple choice

Content

Scenarios

Scenarios

Content

Fill-in-the-blank

Fill-in-the-blank

Content

Realistic tasks

Realistic tasks

Content

Scenario questions

Scenario questions

Content

On-the-job practice

On-the-job practice

Content

The main purpose of feedback is to reduce gaps between current knowledge and performance and desired knowledge performance. In a Review of Educational Research article, Kulhavy describes a continuum of feedback similar to what’s described in Figure 1. Complexity increases as it moves further to the right.

The main purpose of feedback is to reduce gaps between current knowledge and performance and desired knowledge performance. In a Review of Educational Research article, Kulhavy describes a continuum of feedback similar to what’s described in Figure 1. Complexity increases as it moves further to the right. 

Content

Figure 1. Continuum of Feedback Complexity

Figure 1. Continuum of Feedback Complexity

Content

**

**

What Do You Know: Does Feedback Improve Learning?-57cdfeed51047bb67e1b2989fbfc857b4e9f171f17722026ef9af1c3b7a9b686

Content

**

**

Content

L&D practitioners mostly assume feedback helps learning. If it didn’t, we wouldn’t write it and quiz or test systems wouldn’t offer places to input it. But does feedback really have an impact? Before talking about what research says, I’d like you to consider the following question. Then we’ll discuss what the research says and its implications.

L&D practitioners mostly assume feedback helps learning. If it didn’t, we wouldn’t write it and quiz or test systems wouldn’t offer places to input it. But does feedback really have an impact? Before talking about what research says, I’d like you to consider the following question. Then we’ll discuss what the research says and its implications. 

Content

QUESTION: Does instructional feedback (in response to learning questions and activities) improve learning? Select the correct answer.

QUESTION: Does instructional feedback (in response to learning questions and activities) improve learning? Select the correct answer. 

Content

A. We don’t know if feedback improves learning.

A. We don’t know if feedback improves learning. 

Content

B. Feedback improves learning. That’s why we write it.

B. Feedback improves learning. That’s why we write it. 

Content

C. Feedback can improve learning, but it doesn’t always.

C. Feedback can improve learning, but it doesn’t always. 

Does Feedback Help Learning? 

Content

Feedback has been shown to help learning most when it specifically addresses forgotten information or strategies, difficult aspect of performance, or a faulty interpretation (misunderstanding), explain Hattie and Timperley in their Review of Educational Research article, “The Power of Feedback.” Feedback doesn’t help nearly as much when it addresses a lack of understanding. A lack of understanding implies that the instruction didn’t meet its goals or has one or more of the following problems:

Feedback has been shown to help learning most when it specifically addresses forgotten information or strategies, difficult aspect of performance, or a faulty interpretation (misunderstanding), explain Hattie and Timperley in their Review of Educational Research article, “The Power of Feedback.” Feedback doesn’t help nearly as much when it addresses a lack of understanding. A lack of understanding implies that the instruction didn’t meet its goals or has one or more of the following problems: 

  • Content

    The instruction didn’t consider the prior knowledge levels of participants (for example, we assumed they know more than they did)

    The instruction didn’t consider the prior knowledge levels of participants (for example, we assumed they know more than they did) 

  • Content

    The delivery of the instruction is problematic (for example, participants were unable to find or review parts they sought to review)

    The delivery of the instruction is problematic (for example, participants were unable to find or review parts they sought to review) 

  • Content

    Content, practice, or assessment elements are problematic (for example, there is inadequate practice to help remember or apply on the job)

    Content, practice, or assessment elements are problematic (for example, there is inadequate practice to help remember or apply on the job) 

Content

Trying to fix a lack of understanding in instruction is generally beyond its scope. Even well-written feedback, given in the right circumstances, cannot always help because participants don’t always understand or use it.

Trying to fix a lack of understanding in instruction is generally beyond its scope. Even well-written feedback, given in the right circumstances, cannot always help because participants don’t always understand or use it. 

Feedback Types and Conditions 

Content

Hattie and Timperley reviewed instructional feedback meta-analyses (a statistical approach to combine results from multiple studies), to show what types of feedback are likely to help the most and the least. The most powerful outcomes came from feedback about tasks and how to do them more effectively. Goal-oriented feedback and cues (hints) could also be effective. The least powerful outcomes came from praise, rewards, and punishment (extrinsic rewards).

Hattie and Timperley reviewed instructional feedback meta-analyses (a statistical approach to combine results from multiple studies), to show what types of feedback are likely to help the most and the least. The most powerful outcomes came from feedback about tasks and how to do them more effectively. Goal-oriented feedback and cues (hints) could also be effective. The least powerful outcomes came from praise, rewards, and punishment (extrinsic rewards).  

Content

They then looked at how to make effective types of feedback work well. Remember how in the beginning I said that the main purpose of feedback is reducing gaps between current knowledge and actual performance and desired knowledge and performance? Hattie and Timperley explain that to reduce this gap, feedback must answer three questions:

They then looked at how to make effective types of feedback work well. Remember how in the beginning I said that the main purpose of feedback is reducing gaps between current knowledge and actual performance and desired knowledge and performance? Hattie and Timperley explain that to reduce this gap, feedback must answer three questions: 

  • Content

    What are the goals?

    What are the goals? 

  • Content

    What progress am I making towards these goals?

    What progress am I making towards these goals? 

  • Content

    What do I need to do to make better progress?

    What do I need to do to make better progress? 

Content

Clear goals along with knowing where you are and how to move forward targets the right places to focus effort to reduce gaps between current knowledge and actual performance and desired knowledge and performance. Some feedback strategies work in the opposite manner. These include nonspecific or fuzzy goals, accepting poor performance, and not offering enough information. Research shows that when people don’t know what to do, feedback can be demotivating.

Clear goals along with knowing where you are and how to move forward targets the right places to focus effort to reduce gaps between current knowledge and actual performance and desired knowledge and performance. Some feedback strategies work in the opposite manner. These include nonspecific or fuzzy goals, accepting poor performance, and not offering enough information. Research shows that when people don’t know what to do, feedback can be demotivating. 

Content

Goals must supply actions and outcomes for a specific task or performance. They must also include success criteria that allow for consistent performance when facing common obstacles. Feedback cannot lead to a reduction in the “gap” if the goal and the criterion aren’t clear. Otherwise, people may rely on any method that works (for them), and their methods may have undesirable consequences.

Goals must supply actions and outcomes for a specific task or performance. They must also include success criteria that allow for consistent performance when facing common obstacles. Feedback cannot lead to a reduction in the “gap” if the goal and the criterion aren’t clear. Otherwise, people may rely on any method that works (for them), and their methods may have undesirable consequences. 

Content

Telling people how they are doing shouldn’t wait for formal assessment. People need specific feedback against specific goals (with success criteria) while learning in order to learn to self-correct. Learning to self-correct is one of the MOST critical aspects of learning and feedback. Metacognition (being able to assess your thinking and learning) and self-correction are often extremely more important aspects of learning, yet these approaches are often under used.

Telling people how they are doing shouldn’t wait for formal assessment. People need specific feedback against specific goals (with success criteria) while learning in order to learn to self-correct. Learning to self-correct is one of the MOST critical aspects of learning and feedback. Metacognition (being able to assess your thinking and learning) and self-correction are often extremely more important aspects of learning, yet these approaches are often under used. 

Content

Instruction often builds on things learned earlier (I’m seeing a lot of L&D practitioners discussing instruction like this doesn’t happen!). As Hattie and Timperley explain, too often when designing instruction, we keep piling on more information, one topic after another. One central goal of learning design should be helping people answer important metacognitive questions, such as: Do I remember what I need to remember (for work and more learning)? Do I understand how to apply this? Did I interpret this correctly? If these questions go unanswered and we simply pile on more content, we are building on faulty foundations.

Instruction often builds on things learned earlier (I’m seeing a lot of L&D practitioners discussing instruction like this doesn’t happen!). As Hattie and Timperley explain, too often when designing instruction, we keep piling on more information, one topic after another. One central goal of learning design should be helping people answer important metacognitive questions, such as: Do I remember what I need to remember (for work and more learning)? Do I understand how to apply this? Did I interpret this correctly? If these questions go unanswered and we simply pile on more content, we are building on faulty foundations. 

Content

For example, when learning to be a bank teller, there are specific account and math concepts that new hires need to master to understand later concepts. If people don’t understand these, they will not be able to correctly learn new tasks that are based on these concepts. This is true in most job training but too often, we train without considering what people must master before moving on to other, related topics.

For example, when learning to be a bank teller, there are specific account and math concepts that new hires need to master to understand later concepts. If people don’t understand these, they will not be able to correctly learn new tasks that are based on these concepts. This is true in most job training but too often, we train without considering what people must master before moving on to other, related topics. 

Content

I hope you can see that feedback is complex, and we shouldn’t write it only as an add-on response to instructional assessments. We need to better integrate feedback into the design of instruction. In some of my next articles I will answer more complex questions:

I hope you can see that feedback is complex, and we shouldn’t write it only as an add-on response to instructional assessments. We need to better integrate feedback into the design of instruction. In some of my next articles I will answer more complex questions: 

  • Content

    Where (else) does feedback fit into instruction?

    Where (else) does feedback fit into instruction? 

  • Content

    Are their different levels of feedback?

    Are their different levels of feedback? 

  • Content

    How should we design feedback to make it more valuable?

    How should we design feedback to make it more valuable? 

Content

The answer to the question near the beginning of this article is C. I’d love to know your questions about this article. Please share your thoughts in the Comments; they will help me write later articles on this topic so they meet your needs.

The answer to the question near the beginning of this article is C. I’d love to know your questions about this article. Please share your thoughts in the Comments; they will help me write later articles on this topic so they meet your needs. 

References 

Content

Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research , 77, 81- 112. http://www.columbia.edu/~mvp19/ETF/Feedback.pdf

Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77, 81- 112. http://www.columbia.edu/~mvp19/ETF/Feedback.pdf 

Content

Kulhavy, R. W. (1977). Feedback in written instruction. Review of Educational Research , 47(1), 211-232. http://journals.sagepub.com/doi/pdf/10.3102/00346543047002211

Kulhavy, R. W. (1977). Feedback in written instruction. Review of Educational Research, 47(1), 211-232. http://journals.sagepub.com/doi/pdf/10.3102/00346543047002211

You've Reached ATD Member-only Content

Become an ATD member to continue

Already a member?Sign In


Copyright © 2026 ATD

ASTD changed its name to ATD to meet the growing needs of a dynamic, global profession.

Terms of UsePrivacy NoticeCookie Policy