Pre-college Electrical Engineering Instruction: Do Abstract or Contextualized Representations Promote Better Learning? Dr. Roxana Moreno, University of New Mexico Dr. Martin Reisslein, Arizona State University Dr. Gamze Ozogul, Arizona State University Frontiers in Education, October , 2009, San Antonio, TX

Pre-College Engineering Education  The K-12 school audience has been identified as a key target for improving engineering education.  Investigating methods that can help increase the performance and enthusiasm of pre-college students is a major focus.  How to help pre-college students develop problem-solving skills and positive perceptions towards engineering education?  A promising technique shown to promote problem-solving skills in well-structured domains such as physics or mathematics is worked-example instruction.

Two Conflicting Hypotheses  Contextualized Representations Promote Learning Realistic problem representations that are anchored in learners past experiences promote learning by activating prior knowledge that relates to the problem. Predictions: C group will show higher transfer, lower difficulty perceptions, higher perceptions of the program usefulness, especially of the problem representations.  Abstract Representations Promote Learning Abstract problem representations help learners focus on relevant (structural) rather than irrelevant problem information (superficial) Predictions: A group will show higher transfer, lower difficulty perceptions, higher perceptions of the program usefulness, especially of the problem representations.

Research Questions  Does contextualizing problems during worked-example instruction promote the near and/or far transfer of the principles learned?  Does contextualizing problems during worked-example instruction affect students’ ability to represent novel problems?  Does contextualizing problems during worked- example instruction affect students’ learning perceptions?

Method  Participants 86 pre-college students (54 females and 32 males). Age: M =15.4 years (SD = 1.43 years) Ethnicity  42 (48.8 %) students Hispanic American  24 (27.9 %) Caucasian  6 (7.0 %) African American  2 (2.3 %) Native American  2 (2.3 %) Asian American  10 (11.6 %) other ethnicities

Materials  Computerized materials demographic information questionnaire pretest instructional session problem-solving practice session program rating questionnaire  Paper-pencil materials posttest

Treatment Conditions  Abstract (A) Abstract text Abstract representations  Contextualized (C) Contextualized text Context representations

Results  Pretest No significant differences between groups  Abstract, M = 2.12 (max 6), SD = 0.87  Contextualized, M = 2.29, SD = 1.04  F(1, 84) = 0.65, p =.42  Research Question 1: Does Contextualizing Problems Promote the Near and/or Far Transfer of the Principles Learned? Treatment effect on near transfer  Abstract, M = 4.86 (max 9), SD = 3.78  Contextualized, M = 3.09, SD = 3.84  F(1, 83) = 4.98, MSE = 14.51, p =.03 No treatment effect on far transfer  Abstract, M = 1.61(max 9), SD = 2.69  Contextualized, M = 0.96, SD = 2.37  F(1, 83) =1.62, MSE = 6.41, p =.21

Results_ continue  Research Question 2: Does Contextualizing Problems Affect Students’ Ability to Represent Novel Problems? 15 % of the participants spontaneously produced graphic representations of posttest problems. Six of these students were in A group and 7 were in C group. Group A produced significantly better representations of the posttest problems than group C  Abstract, M = (max 60), SD =  Contextualized, M = 9.38, SD = 6.26  F(1, 10) = 5.39, MSE = , p =.04.

Results_ continue  Research Question 3: Does Contextualizing Problems Affect Students’ Learning Perceptions? No significant differences between the treatment groups on ratings of overall program usefulness (p =.60) No significant differences between the treatment groups on difficulty perceptions (p =.26) Marginally significant difference for representation usefulness ratings. Group C > group A, F(1, 84) = 2.84, MSE = 0.86, p =.10.

Theoretical Implications  Abstract representations help learners focus on relevant structural information underlying isomorphic problems  The findings support a coherence principle for worked- example engineering education according to which visual adjuncts that are not necessary to promote the learning objectives of a lesson should be minimized.  The marginal tendency in favor of group C on the picture representation usefulness suggests that realistic problem representations may create an illusion of understanding (they are perceived to be more useful but do not promote learning).

Practical Implications  Pre-college engineering instruction should focus on the development of abstract problem solving before tackling real-life problems independently  Pre-college students have reached the cognitive development necessary to engage in abstract thinking, development of abstract problem solving is appropriate for this age