Assessing metacognitive knowledge in 3–5 year olds: the development of a metacognitive knowledge interview (McKI) (original) (raw)

Abstract

Historically, early cognitive skills have been underestimated, largely as a result of the ways these competencies have been measured, which is particularly pervasive in the area of metacognition. Only recently have researchers begun to detect evidence of contextualized metacognition in 3–5 year old preschool children through the use of observational assessment tools (e.g., Whitebread et al. J Cogn Educ Psychol 3:433-455, 2007, Metacognition Learn 4:63-85, 2009). While these observational methods are a more sensitive way to capture metacognition in young children, their exclusive use may not result in a comprehensive depiction of early metacognitive competency. In the current study, we describe the development of a metacognitive knowledge interview (McKI) and what it reveals about metacognitive processes in 43 3–5 year olds (including investigating face validity). Findings indicate that the McKI (a) is a developmentally appropriate measure for 3–5 year olds, (b) is capable of eliciting articulated metacognitive knowledge when engaging in a contextualized problem-solving task, (c) shows the expected developmental trend (i.e., older children perform at a higher level and scores increased over the course of a school year), and (d) provides sufficient variation across children. Implications for future research are discussed, including the importance of using multiple measurement tools when studying early metacognitive development.

Access this article

Log in via an institution

Subscribe and save

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1

The alternative text for this image may have been generated using AI.

Fig. 2

The alternative text for this image may have been generated using AI.

Fig. 3

The alternative text for this image may have been generated using AI.

Similar content being viewed by others

Notes

  1. See Flavell 2004 for a comprehensive review; Wellman and Liu 2004 for validation studies of various ToM measurement for preschoolers; and Bemis et al. 2013, Lillard and Kavanaugh 2014, and Rhodes and Wellman 2013 for recent empirical studies investigating ToM in preschoolers

References

Download references

Acknowledgments

We would like to thank the preschool children and their teachers for their time and effort in making this research possible. This research—and the writing of the manuscript—was funded by the generous support provided to the first author by the National Academy of Education/Spencer Dissertation Fellowship Program, the University of Michigan Rackham Predoctoral Fellowship Program, and the American Psychological Association (APA) Dissertation Research Award. Part of the research presented in this paper was presented at the Annual Meeting of the American Educational Research Association, Vancouver, B.C., April 2012.

Author information

Authors and Affiliations

  1. Department of Human Development, Connecticut College, 121 Bolles House, 270 Mohegan Avenue, New London, CT, 06320, USA
    Loren M. Marulis
  2. School of Education, University of Michigan, 610 E. University Ave., Ann Arbor, MI, 48109, USA
    Annemarie Sullivan Palincsar
  3. Department of Psychology & Education, Albertus Magnus College, 811 Winchester Ave., New Haven, CT, 06511, USA
    Amanda L. Berhenke
  4. Faculty of Education, University of Cambridge, 184 Hills Road, Cambridge, CB2 8PQ, UK
    David Whitebread

Authors

  1. Loren M. Marulis
  2. Annemarie Sullivan Palincsar
  3. Amanda L. Berhenke
  4. David Whitebread

Corresponding author

Correspondence toLoren M. Marulis.

Additional information

This research was conducted as a part of the first author’s dissertation research while she was a doctoral student in the Combined Program in Education and Psychology at the University of Michigan. Her contact information whilst conducting this research was: University of Michigan, Combined Program in Education and Psychology, School of Education Building, Room 1406, 610 E. University Ave, Ann Arbor, MI, 48109, marulisl@umich.edu. Loren’s current affiliation and address as Martha Bennett ‘73 Assistant Professor of Human Development, is listed above.

Appendices

Appendix A. Wedgits problem-solving puzzle task

figure a

The alternative text for this image may have been generated using AI.

Appendix B. Metacognitive Knowledge Interview (McKI)

Once the Wedgits puzzle task is complete, tell child: “Thank you for working on those puzzles! I would like to talk to you about the puzzles you just did and about your thinking. My job is to learn about how kids learn and think and I have a few questions for you, Okay?” Once child assents, say: “Thank you. Remember, there are no right or wrong answers; I only want to know what you think. Just give your best answer.” (If they don't agree, try to prompt them by saying: “I really need your help and want to learn about how kids think”.)

    1. “Do you think you did a good job, an okay job or not so good of a job on the puzzles?” Circle child's response. If they say they did a good job, ask “What did you do to help you do a good job?” If they answer okay or not so good, ask “What do you think would have helped you do an even better job?
    1. “Did you think anything was hard?” If no, ask: “Why not?” If yes, ask “Why? What would have made it easier?”
    1. “Would these puzzles be hard for another kid your age? Why/why not?”
    1. How did you know if you were getting the puzzles right?”

Show child the ‘alien’ finger puppet and say: “I have another friend to show you. This puppet’s name is Gogi and he/she (use same gender as the child) is from another land. S/he does not go to a school like yours or have a teacher like yours and doesn’t know anything about puzzles like the ones you just did. Will you help Gogi learn about these kind of puzzles?” Wait for child to assent and say: “Thank you.” (If they don't agree, try to prod them by saying that ‘Gogi really needs your help and wants to learn about these kind of puzzles’.)

    1. “Would these puzzles be easier for Gogi or you? Why?”
    1. “What should Gogi do if s/he is having trouble with the puzzle?”
    1. “Would it be helpful for Gogi to talk to herself/himself about the puzzle while doing the puzzle? Why would/wouldn’t that be a helpful thing to do?
    1. “Gogi has some questions for you about puzzles like this one. Okay?” Have Gogi ‘speak’ directly to the child and ask the following:
    1. Would the puzzle be easier with bigger or smaller pieces? Why?”
    1. “If all of the puzzle pieces were the same color, like in this picture (show the Wedgits booklet of all purple Wedgits) will the puzzle be easier? If yes, ask: “Why?” If no, ask, “Why not?”
    1. “If I think about how the pieces would fit together before I try, will the puzzle be easier? If yes, ask: “Why?” If no, ask, “Why not?”
    1. “If I close my eyes while I do the puzzle, will it be easier? If yes, ask: “Why?” If no, ask, “Why not?”

“Thank you for sharing all of your ideas and how you think with Gogi!

Appendix C. Wedgits Puzzle Scoring

    1. Rate the accuracy of the child’s performance on the Wedgits task. These coding categories are designed to match the Metacognitive Knowledge Interview (McKI) in which the child is asked how well she or he did on the puzzle (Question #1).
    2. a.
      Good = Child accurately (e.g., the puzzle looked exactly like the picture card) finished the first and second puzzles within the time allotted (4 min). The child may have started (or completed) a third or even fourth puzzle, but this is not required to receive a score of “Good”.
    3. b.
      Okay = Child accurately (e.g., the puzzle looked exactly like the picture card) finished the first puzzle within the time allotted (4 min) and accurately completed at least half of the second puzzle (i.e., completed the bottom half—that looks like a pyramid—and the green piece that is placed vertically in the pyramid).
    4. c.
      Not so good = Child accurately (e.g., the puzzle looked exactly like the picture card) finished the first puzzle within the time allotted (4 min) and started the second puzzle but did not accurately complete half of the second puzzle (i.e., did not complete the bottom half that looks like a pyramid).

Appendix D. Metacognitive Knowledge Interview (McKI) Annotated Scoring Codebook

The Metacognitive Knowledge Interview (McKI) is used to assess children’s metacognitive knowledge (or knowledge about individuals, tasks and strategies) individually using a series of 11 questions related to the Wedgits task [in which children are shown design cards of increasing difficulty and are asked to make the Wedgits building blocks look exactly like the picture on the card.

TO SCORE:

Rate responses to questions on a 0–2 scale for each question where:

The full set of questions is below including example responses and scoring. The actual scores given for this child are indicated along with what would qualify for the other levels of scoring.

Metacognitive Knowledge Interview_CODED SAMPLE

Once the Wedgits puzzle task is complete, tell child: “Thank you for working on those puzzles! I would like to talk to you about the puzzles you just did and about your thinking. My job is to learn about how kids learn and think and I have a few questions for you, Okay?” Once child assents, say: “Thank you. Remember, there are no right or wrong answers; I only want to know what you think. Just give your best answer.” (If child doesn’t agree, try to prod by saying “I really need your help and want to learn about how kids think”.)

    1. “Do you think you did a good job, an okay job or not so good of a job on the puzzles?” Circle child’s response. If they say they did a good job, ask “What did you do to help you do a good job?” If they answer okay or not so good, ask “What do you think would have helped you do a better job?” My brain-that controls my whole body and helped me think about the puzzle.

2 points. His knowledge of how well he performed on the task was accurate. He performed above his age level by completing the second challenging puzzle accurately and quickly and nearly completed the 3rd puzzle (designed for children older than preschool and chosen to pose a challenge sufficient enough for children this age to be unable to complete accurately, particularly within four minutes). Further, he showed metacognitive knowledge about himself and what helped him/would help him do a good job on puzzles-his brain.

1 point. A score of 1 would be given either for an inaccurate assessment of one’s performance on the puzzle or an accurate response to this with a non-metacognitive response to what would help do a good/better job such as “being good” or “doing a good job” (it’s a 2 part question). If a child accurately assessed his/her Wedgits performance and gave a partial metacognitive response (such as: “liking the puzzle” because it is possible that higher enjoyment/interest in a task leads to higher performance but this response does not fully spell this out. Or responding to the question “What did you to do help you do a good job” with: “I tried hard” without elaboration), he/she could get 1.5 on this question.

0 points would be given if the child was metacognitively inaccurate about his/her performance along with a non-metacognitive follow-up response.

    1. “Did you think anything was hard?” If no, ask: “Why not?” If yes, ask “Why? What would have made it easier?” My brain focuses when I’m doing puzzles

1 point. He said nothing was hard but other comments he made during the third puzzle (see notes below) indicated that there were parts he found very difficult. However, he verbalized metacognitive knowledge regarding why it didn’t seem difficult-he was aware that by being able to concentrate and maintaining focus (with your brain), the task will likely be/seem easier.

2 points. A child whose response to whether anything was hard matches his/her comments/emotional response to the puzzle would get a full point for that part of the question and another full point for why he/she didn’t find it hard-e.g., the type of response given here. If the child said something was difficult (and this matched his/her response to the puzzle) and also responded with a metacognitively aware answer regarding what would have made it easier-e.g., having help from an adult/older child or getting a hint, she/he would receive 2 points.

A score of 0 would be given if the child was metacognitively inaccurate about his/her performance along with a non-metacognitive follow-up response to Why/Why not.

    1. “Would these puzzles be hard for another kid your age? Why/why not?” No, I don’t know.

1 point. His answer matches the one above as to whether the task would be difficult depending on age but didn’t give a response as to why he thought this.

In order to receive 2 points, a child would have to either match the response to Question #2 (e.g., if they said the puzzle was/wasn’t difficult for them) and follow with a metacognitively aware response as to why such as “these puzzles are hard for 4 year olds” or “we have these in our classroom so they’re not hard for us” OR, the child could have a different response with a metacognitively aware answer as to why such as “Yes, they would be hard for other kids my age because they don’t have them at home like I do” or “No, they wouldn’t be hard for other kids my age because they are better at puzzles than me”.

0 points would be given if the child gave an answer that didn’t match #2 and gave a non-metacognitive follow-up response to Why/Why not.

    1. “How did you know if you were getting the puzzles right?” I looked at the picture.

2 points. He was aware not only of what strategy would be helpful in accurately completing the task but of which strategy he actually used while doing the task (this can be discerned by watching the video or noting whether the child actually used this strategy during the task).

A child would be given 1 point if gave a partially metacognitive response such as “I thought about it hard”.

0 points were given for non-plausible/non-metacognitive responses such as “I just knew” or “because I’m smart/good”.

Show child the ‘alien’ finger puppet and say: “I have another friend to show you. This puppet’s name is Gogi and he/she (use same gender as the child) is from another land. S/he does not go to a school like yours or have a teacher like yours and doesn’t know anything about puzzles like the ones you just did. Will you help Gogi learn about these kind of puzzles?” Wait for child to assent and say: “Thank you.” (If they don’t agree, try to prod them by saying that ‘Gogi really needs your help and wants to learn about these kind of puzzles’.)

    1. “Would these puzzles be easier for Gogi or you? Why?” A lot; I focus a lot

1 point. He indicated metacognitive knowledge in asserting that he (would had experience with puzzles and was from a school/had teachers who taught about puzzle and puzzle-related tasks) but not in his response as to why. It is metacognitive to understand that focusing helps improve performance, but in this instance, the question was about why he would have an easier time than Gogi.

In order to receive 2 points, a child would have to say something like “I have already done those puzzle” or “I know all about puzzles and Gogi doesn’t know about them at all”.

0 points would be given if the child chose Gogi without a metacognitive explanation. The child could receive 1 point by choosing Gogi but giving a metacognitive response as to why such as “He looks smarter with that big brain (the toy had a brain external to his head).”

    1. “What should Gogi do if s/he is having trouble with the puzzle?” Ask someone.

2 points. He his answer reflected metacognitive awareness of a good strategy to use when encountering trouble (i.e.., help-seeking).

To receive 1 point, a child could respond with an answer that indicates some awareness of cognitive states but not of a (potentially) successful strategy, such as “Try it” (if the child had added “harder” or “again” after “Try”, she/he would receive the full 2 points indicating that he/she is aware that trying again, using a different strategy, or putting forth more effort may result in more success).

A score of 0 would be a response such as “Be good” which is not at all indicative of awareness of cognitive strategy.

    1. “Would it be helpful for Gogi to talk to herself/himself while doing the puzzle? Why would/wouldn’t that be a helpful thing to do? Yes-he has to focus and get concentration. Talking to yourself does that.

2 points. He verbalizes metacognitive knowledge that talking to oneself about a task while performing it is likely to be cognitively helpful as well as why (because it helps you focus and concentrate).

1 point would be given if the child answered Yes but didn’t know why or gave a non-metacognitive response as to why such as “It helps” or “It’s good”.

0 points would be given for an answer of No without a metacognitive explanation. The child could receive 1 point by responding “No” but giving a metacognitive response as to why such as “Because if you talk you might get distracted and do a bad job on the puzzle”.

“Gogi has some questions for you about puzzles like this one. Okay?” Have the puppet ‘speak’ directly to the child and ask the following:

For the last 4 questions, there was an intended ‘correct’ answer in that one response would make the puzzle task easier (as confirmed by adults answering these same questions). Thus, the child would get 1 point for answering correctly/metacognitively accurately and another 1 point for giving a metacognitively aware response as to why this was true. However, it became apparent after conducting several interviews that children sometimes viewed the strategies differently than adults. Thus, if they responded differently than the adults to which situation would make the puzzle easier but gave a metacognitive response that appropriately made the first response defensible, he/she received the full 2 points (if there was no response given to back up their answer or a non-metacognitive response given, the score would be 0. In contrast, if the child responded as adults did to the first part but did not respond to the second part or responded non-metacognitively, she/he would receive 1 point).

    1. “Would the puzzle be easier with bigger or smaller pieces? Why?” Easier to hold in your hands.

2 points. He responded like adults for bigger pieces and gave a response that, while not as cognitively related as adults answered-e.g., “less pieces to have to figure out how to fit together or less intricate designs”, it was a plausible reason as to why bigger pieces make doing a puzzle easier.

1 point would have been given if there was no response to “Why” or he had said something like “It’s better”.

0 points would have been given if he had said smaller was easier and given no response to “Why” or he had said something like “It’s better”. However, 2 points would have been given if he had said “smaller” along with a metacognitive response such as “Smaller pieces would be easier for Gogi to hold/see” (Gogi is a small hand puppet).

    1. “If all of the puzzle pieces were the same color, will the puzzle be easier? If yes, ask: Why? If no, ask, “Why not?” I can’t figure out which one goes where.

2 points. He responded like adults and gave a metacognitive response to “Why not”.

1 point would have been given if there was no response to “Why” or he had said something like “It’s harder like that”.

0 points would have been given if he had said “Yes” with no response to “Why” or he had said something like “It’s just easier”. However, 2 points would have been given if he had said “Yes” along with a metacognitive response such as “Then you would be able to sort by size” (indicting less cognitive load because you don’t have two dimensions on which to sort).

    1. “If I think about how the pieces would fit together before I try, will the puzzle be easier? If yes, ask: Why? If no, ask, “Why not?” Because it is (demonstrated ‘thinking’ and putting the pieces in the right places).

1 point. He responded as adults would but did not give a metacognitive response to “Why” (though his demonstration came close, but even after being prompted after this enactment, he did not respond with any indication of metacognitive awareness.

2 points would have been given if his second response was something like “because it helps me focus on the puzzle” or “I’d have more time to figure out the right place”.

0 points would be given for a response of “No” with no response as to “Why not” or a non metacognitive response such as this one “Because it isn’t”.

    1. “If I close my eyes while I do the puzzle, will it be easier? If yes, ask: Why? If no, ask, “Why not?” No, he can’t see what he’s doing! He couldn’t see if he had the right piece or the right place on the card (pointed to the design picture card)”.

2 points. He responded like adults and gave a metacognitive response to “Why not”.

He would have been given 1 point for not responding to “Why not” or giving a non metacognitive response such as this one “Because it isn’t easier”.

0 points would be given for a response of “Yes” with no response to “Why” or without a metacognitive response to defend this (see above-adults were unable to come up with a way to metacognitively defend this answer except possibly by discussing how other senses may become more acute, but this still wouldn’t make the puzzle easier though a case may be able to be made for equally easy”).

“Thank you for sharing all of your ideas and how you think with Gogi!

Notes: While he was building the puzzle, he made a lot of metacognitive (evaluative) comments such as “I built this one before so I’m good at it.” And for the most challenging puzzle, he said “OOOh, I can’t do that one! It’s so hard!!” Then as he worked on it, he said “I’m having trouble with this part” (he didn’t successfully complete it but was close).

Rights and permissions

About this article

Cite this article

Marulis, L.M., Palincsar, A.S., Berhenke, A.L. et al. Assessing metacognitive knowledge in 3–5 year olds: the development of a metacognitive knowledge interview (McKI).Metacognition Learning 11, 339–368 (2016). https://doi.org/10.1007/s11409-016-9157-7

Download citation

Keywords