Evidence of Metacognition Hierarchy in Sophomore Engineering Students

Evidence of Metacognition Hierarchy in Sophomore Engineering Students Alex Davies & Dr. Hector Medina School of Engineering Abstract Conclusion FIGURE 1: MAI 2018 RESULTS 63% 86% Abstract: • Ongoing research aims at developing an evidence-based framework for self-regulated learning. • Various measuring tools of metacognition have been proposed such the Metacognition Awareness Inventory (MAI). The MAI, developed by Schraw and Dennison in 1994, consists of a survey of 52 true-or-false questions and it is aimed at measuring the various metacognition constructs or levels: three in the subarea of awareness and five in the subarea of regulation or control. • Test Group: sophomore-level engineering course during two consecutive semesters. • Results: . Overall, students were better at understanding their cognition, which represents the basic levels of metacognition. However, they had a more difficult time implementing ways to control their knowledge and regulate their learning. • Subcategory of “evaluation” ranked the lowest in both semesters. Trailing behind “evaluation” was “planning”, which confirms that both are at the highest hierarchical levels of the metacognitive constructs. • The results provide guide to the researchers in future implementation of treatments aimed at developing self-regulated learner Conclusions: • The results give evidence to significant conclusions about hierarchy trends in metacognition • The results can be modelled similar to that seen in Bloom’s Taxonomy and this metacognition pyramid can be seen in Figure 7. • Given that Planning and Evaluation ranked lowest in the MAI results in both 2018 and 2019 these can be assumed to be the highest on the pyramid • The awareness of knowledge traits are the foundation to metacognition and a basic understanding must be fulfilled before regulation of knowledge can grow. • Although these traits can be shown in a pyramid, it doesn’t necessary mean that a student needs 100% in one category before they move on to the next one. • Instead this pyramid should be read as building blocks, where the higher you get in one trait the more likely you are to succeed in the next • Given that these were self evaluated there could be some bias in the answers particularly if students gave themselves near a 100% score in every trait 60% 85% 69% 82% 75% 76% Declarative knowledge Conditional Knowledge Procedural Knowledge Information Management Debugging Strategies Comprehension Monitoring Evaluation Planning FIGURE 3: MAI 2019 RESULTS 62% 76% ow led 58% of on ati gul Re 78% 71% Declarative knowledge Conditional Knowledge Procedural Knowledge Information Management Debugging Strategies Comprehension Monitoring Evaluation Planning Aw Kn aren ow ess led ge of 70% { Figure 5: Data Distribution MAI 2018 Comprehension Monitoring Debugging Strategies Information Management Procedural Conditional ng Pl an ni n ua tio al on M sio n Ev ito rin g gie s in en gg Future Work Co m pr eh n m at io fo r In De bu M an g. S ag tra te em en t dg e le Kn al ed ur oc Co nd Pr iti on al Kn ow ow le le dg dg e e Declarative kn ow e tiv ra Results: • Overall, students were better at understanding their cognition, which represents the basic levels of metacognition. However, they had a more difficult time implementing ways to control their knowledge and regulate their learning. • Subcategory of “evaluation” and “planning” ranked the lowest in both semesters • The highest in both semesters were the awareness of knowledge traits including “declarative knowledge”, “conditional knowledge”, and procedural knowledge” • Based on the results from the surveys and the percentages observed there is a clear trend from high to low. • The trends observed in 2018 are almost identical to that received in 2019 • In 2019 the procedural knowledge was the highest which is different to 2018 when it was the third highest • The 2019 data was on average lower than what was seen in 2018 • The 2019 data had a larger spread of data then 2018, and the standard deviations were much higher 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% cla • The test group consisted of two consecutive semesters of sophomore level engineering students taking a dynamics class • The MAI, developed by Schraw and Dennison in 1994, consists of a survey of 52 true-or-false questions and it is aimed at measuring the various metacognition constructs or levels: three in the subarea of awareness and five in the subarea of regulation or control. • Each “True” answer was given a value of ‘ 1’ and each “False” response was given a value of ‘ 0’ • The total number in each category was totaled and then divided from the total possible to get a percent of success in that trait. • Students were assigned this project for a small amount of class credit to ensure that they spent time completing the survey • The surveys were self evaluated but students were encouraged to be as honest as possible, and were told that their answers would not affect their grade • Students were told how this project was for their own benefit to learn where the struggled and where they excelled. 71% De Methods Evaluation Kn 76% Research Question • Are Sophomore-level mechanical engineering students evenly distributed in relation to the various levels (or constructs) of metacognition? If not, then, what are the levels were they show strength/weakness the most? Planning ge Introduction and Research Question Introduction: • Metacognition: the awareness and regulation of cognition. Put in simple words, metacognition measures how much one is aware of one’s own knowledge as well as how much control of it one has • The two categories split up the different levels of cognition, to measure how people can go from just understanding what they know to regulating and controlling it • Goal is to find the hierarchal levels of metacognition and which one's students struggle with the most • A hope to identify difficult traits with the plan to develop ideas on how to help future classes • The various questions asked in the MAI survey were meant to encourage students to reflect on the way they study and learn best. { Figure 7: Metacognition Hierarchy Pyramid Figure 6: Data Distribution MAI 2019 90% 1. Repeat exercise and add a control cohort 2. Use the results from this study to help create an evidence-based framework for self-regulated learning 80% 70% 60% 50% References 40% 30% 1. 20% 10% 0% Declarative knowledge Conditional Knowledge Procedural Knowledge Information Management Debugging Strategies Comprehension Monitoring Evaluation Planning 2. 3. Medina, Hector & Davies, Alex “Metacognition Theory and Trends Observed from Mechanical Engineering Undergraduate Students” submitted to present at 2020 ASEE International Conference and Exposition, Montreal, Canada, June 2020. Case, Jennifer, et al. “Students’ Metacognitive Development in an Innovative Second Year Chemical Engineering Course. ” Research in Science Education, 2001, doi: 313– 335. Shcraw & Dennison “Assessing Metacognition Awareness” 1994