Available online at www.sciencedirect.

com

Procedia Social and Behavioral Sciences 8 (2010) 403410

International Conference on Mathematics Education Research 2010 (ICMER 2010)

Assessing Cognitive and Metacognitive Strategies during Algebra Problem Solving Among University Students
Sahar Bayata,*, Rohani Ahmad Tarmizia,b
a

Laboratory of Innovations in Mathematics Education, Institute for Mathematical Research, b Faculty of Educational Studies Universiti Putra Malaysia

Abstract One of the most important goals of mathematics learning is to enhance mathematical knowledge, which includes mathematical concepts and skills in problem solving. Thus, efforts to achieve these goals have increased. The purpose of this study was to assess the cognitive strategies and Algebra problems solving performance among university students. The design adopted for this study was a descriptive correlation design. The subjects of this study were selected from First Year mathematics students who took Algebra course in a public university in Malaysia. The Cognitive Strategy Questionnaire, which comprised of 18 items, was used to assess the students specific cognitive strategy for solving the given Algebra problems. Algebra problem solving performance was measured using a test which included routine and non-routine problems, based on the topics covered in the course. The results showed that there is no significant correlation between Algebra problem solving performance with shallow cognition strategy (r = -.134, p>0.05). Similarly, there is no significant relationship between the students performance with deep cognitive strategy (r = .124, p>0.05). Results also showed that there is significant and strong relationship between students Algebra problem solving performance and overall performance in the course (r = .721, p<0.05). Findings also revealed that there was positive and moderate significant correlation between overall meta-cognitive strategies and performance of Algebra problem solving (r = 0.394, p<0.05). In addition, there was a significant positive and moderate relationship between overall metacognitive strategies with overall performance in the course (r = .390, p<0.05). Specifically, there is significant relationship between overall performance in the course and all three subscales of meta-cognition (knowledge, planning and evaluation). In conclusion, meta-cognitive strategies may have impact on mathematical performance among university students whilst cognitive strategies indicated minimal impact. 2010 Published by Elsevier Ltd.
Keywords: Cognitive strategy; Metacognitive strategy; Algebra problem solving performance; Routine problems; Non-routine problems

1.

Introduction

The teaching and learning of problem solving has always been a fundamental and vibrant field of research since early 1980 and to date in 2009. Since learning with understanding is generative and improves transfer, learners might apply the acquired knowledge to learn new topics and to solve new problems (Hiebert & Carpenter, 1992). According to De Corte, Verschaffel and Opt Eynde (2000) learning is a constructive, cumulative, self-regulated,

* Corresponding author. E-mail address: p4ry4_f@yahoo.com

1877-0428 2010 Published by Elsevier Ltd. doi:10.1016/j.sbspro.2010.12.056

404

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

goal-oriented, collaborative, and individualized process of knowledge building and construction of meaning (p. 690). Thus many researchers are interested in finding ways to enhance the effectiveness of students learning. Previous researches have shown that learning process or activity influences meaningful cognitive strategies, which in turn influences academic achievement (Greene & Miller, 1996). When students gain new information, they use various cognitive and meta-cognitive strategies to help them encode, organize and retrieve new information (Somuncuoglu & Yildirim, 1999). Cognitive strategies are classified into two types namely, surface cognitive strategies and deep cognitive strategies. Surface cognitive strategies refer to rehearsal, involving the repetitive rehearsal and rote memorization of information, which helps to encode new information into short-term memory, mainly through reading the course material over and over again. On the other hand, deep cognitive strategies pertains to elaborating, organizing and critical thinking, involve challenging the veracity of information encountered and attempting to integrate new information with prior knowledge and experience, which facilitates long-term retention of the target information for example making an outline of important concepts after a learning session. Some studies have been conducted to show the relationship between performance goals and the use of deep cognitive strategies among college students. The results indicated that performance goals are consistently unrelated to students use of deep cognitive strategies. However, Elliot et al. (1999), Pintrich (2000) and Wolters (2004) mentioned that most of the research on the relationship of performance goals and study strategies did not distinguish between approach and avoidance versions of performance goals. Wolters (2004) distinguished between approach and avoidance versions of performance goals and found that there is positive relationship between performance approach goals and deep cognitive strategies. The repetitive rehearsal and rote memorization of information are characteristics of surface approach processing. It is assumed by theorists that surface processing is integral to the development of a comprehensive knowledge base. Memorization of information is provided in academic exams of a course. Sometimes the surface cognitive strategies are necessary for examinations, such as multiple-choice tests or exam questions which are not meant to test deep understanding or integration of the material. Therefore, positive relationship exists between approach and avoidance performance-goals with students use of surface cognitive strategies. Several studies reported that performance goals are positively related to students use of surface cognitive strategies (Elliot et al. 1999). In addition, Dupeyrat and Martin (2005) indicated that performance-approach goals had positive relationship with students use of surface cognitive strategies. The concept of metacognition was introduced by Flavell as a concept of intelligent structuring and storage of input, of intelligent search and retrieval operations, and of intelligent monitoring and knowledge of these storage and retrieval operations - a kind of 'meta- memory' (Flavell, 1971 p. 277). Thus, he conceptualise metacognition as the notion of thinking about one's own thoughts. According to Flavell, metacognition includes the awareness about what one knows - metacognitive knowledge, what one can do - metacognitive skills and what one knows about his own cognitive abilities - metacognitive experience. Using Flavell's words, metacognition is "knowledge and cognition about cognitive phenomena" (Flavell, 1979 p. 906). Earlier studies forwarded meta-cognition as consisting of four important meta-cognitive strategies in mathematical problem-solving, including prediction, monitoring, planning and evaluation (Brown, 1980; Desoete, Roeyers, & Buyse, 2001). Prediction as a meta-cognitive strategy enables the student to differentiate difficult problem from easy ones. Monitoring ensures that students are closely following their plan and tracking whether the plan is helping to successfully solve the problem. Planning helps the student analyze the problem, retrieve relevant domain-specific skills, and properly sequence problem solving strategies. Evaluation of the problem is placed at the end of the process, in which the student judges the answer and the process taken to obtain the answer. Planning strategies enable problem solvers to determine which sub-goals must be obtained and in what order (Derry and Hawkes, 1993). Elaboration consists of summarizing, creating analogies, paraphrasing and generative note taking. They help to encode new information with the existing knowledge in long term memory(Ormrod, 1999; Woolfolk, 1998; McCown & Roop, 1992).Organization refers to selecting main ideas, outlining, networking, drawing up tables or charts and diagramming. They help to analyze new information in terms of the relationship among ideas and transferring this information into different modes of representation (Somuncuoglu & Yilidirim, 1999). Students who employ elaboration and organization engage at deep level of processing. They are able to recall the information

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

405

easily later as more elaborate encoding of information produces better learning and recall (Rogers, 1994). According to Cassidy (2000), cognitive strategies have been shown to be important components of performance. Miller (1996) in his research illustrated that there is a positive relationship between deep cognitive strategy and mathematics achievement (r = 0.26, p<.001). However, the correlation between shallow cognitive strategy and mathematics achievement did not reach any statistical significance. In addition, Albaili (1998) indicated that undergraduates with high academic achievement utilized shallow cognitive strategies. Although many studies exist on cognitive strategies and performance, only a few investigate the relations between these two components (Ainley, 1993; Graham & Golan, 1991; Nolen & Haladyna, 1988; Pintrich & Garica, 1991). According to Montague and Applegate (2003) mathematical problem solving is derived from research in general problem solving, mathematical problem solving, metacognition, and affective variables associated with problemsolving performance. The cognitive strategies and processes (i.e., specific problem-solving strategies) are read, paraphrase, visualize, hypothesize, estimate, compute, and check. The metacognitive strategies and processes that develop awareness and regulation of the cognitive strategies include self-instruction, self-questioning, and selfmonitoring. They further explained that the instructional application of mathematical problem solving has four components: (a) assessing student performance and identifying students for whom the instructional program is appropriate; (b) explicitly instructing students in the acquisition and application of strategies for mathematical problem solving; (c) process modelling; and, (d) evaluating student outcomes, with an emphasis on strategy maintenance and generalization. Thus far, in investigating performance in mathematical problem solving, the role and impact of the cognitive strategy and metacognitive strategy in mathematical activity may enlighten teaching and learning of mathematics. In pursuit of enhancing mathematics learning for, this study seek to investigate and assess: 1. 2. 3. 4. 5. 6. The students level of cognitive strategy (deep versus surface) related to mathematical learning. The students level of meta-cognitive strategy in) related to mathematical learning. The students level of each subscale of meta-cognitive strategy (knowledge, planning and evaluation) related to mathematical learning. The relationship between students level of shallow cognitive strategy with algebra problem solving performance. The relationship between students level of deep cognitive strategy with algebra problem solving performance. The relationship between students level of meta-cognition and its subscales with algebra problem solving performance.

2.

Methodology

The research design adopted for this study was the descriptive correlation design in examining the possible influences of both cognitive and metacognitive strategies of respondents on the mathematical problem solving ability. Cognitive strategy and metacognitive strategy were assessed by using self-report instruments. The mathematical problem solving measure was an Algebra Test consisting of routine and non- routine problems and overall scores obtained for the course. The subjects of this study were selected from first year students who took Algebra (MTH 3200) course in one of the public university in Malaysia. Three groups of students had taken this course. Two groups totalling of 86 students were then selected randomly (two out of three groups). The students involved were mathematics major and also mathematics majors with education. The cognitive strategy instrument assess two types of cognitive strategies: surface cognitive strategy (referring to rehearsal, i.e. highlighting, underlining, copying, repeating items in a list) and deep cognitive strategy (referring to elaboration, i.e. paraphrasing, summarizing, creating analogies, and generative note-taking, that help integrating new information with the existing knowledge in long-term memory, and organization such as selecting main ideas, outlining, networking, and diagramming, that help analyzing the information in a text, in terms of the interrelations between ideas and transferring this information into different modes of representation). The students were asked to respond to each statement with regard to their common practices during learning algebra and solving algebra

406

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

problems using a five-point Likert scale ranging from 1 - never to 5 = very often. The Metacognitive Awareness Inventory (MAI) comprised of 52 items was used to assess students opinions concerning the occurrence of metacognitive activities (such as self-instruction, self-questioning, and self-monitoring) while solving algebra problems throughout the MTH 3200 course. The students were required to either give a true or false response to each item. Algebra problem solving performance was measured using a test based on problems discussed in their tutorial class. This test included of seven algebra questions - four questions were routine problems and three questions were nonroutine problems. Based on some definition, non-routine algebra problems are problems that require creative and/or critical thinking. A solution to this type of mathematics item necessitates the application of knowledge, previous experience, and/or intuition. A non-routine problem can also be a verbal or word problem, for which an algorithm has not been learned (Barson, 1986, p.10). In this study, non-routine problems are problems which students previously were not familiar in their classroom and solving them requires them to be inventor, whilst the routine problems are common algebra problems used in the class. Algebraic mathematical achievement was also measured and based on the cumulative final score of the MTH 3200 course taken by the students during the semester. 3. Findings

A total of 88 sets of questionnaires and test were collected. From the 88 respondents 85.2% were females and 14.8% were males. Majority of the student respondents were Malay students (71; 80.7%) followed by Chinese students (15; 17%) and there were a few respondents as other ethnic (2, 2.3%). The respondents major was mathematics with one group enrolled in Bachelor of Science degree and the other enrolled in Bachelor of Science with Education. A total of 42 respondents (47.7%) were from Bachelor of Science and 46 respondents (52.3%) were from Bachelor of Science with Education. 3.1 Students level of cognitive strategy and algebra problem solving performance

Table 1 indicated descriptive measures of students level of cognitive strategy. Findings indicated that students shallow cognitive strategy with mean of 3.69 was higher than their deep cognitive strategy with mean of 3.46. This may indicate that the students prefer shallow or surface strategy compared to deep strategy when learning or solving algebra problems. Table 1: Students level of Cognitive strategy Mean 3.46 3.69 Std. Deviation .57 .59

Deep Cognitive Strategy Shallow Cognitive Strategy

The means and standard deviations of the students performance variable are provided in Table 2. The mean of overall performance in the Algebra Problem Solving (ALGPS) was 3.304. The students score of overall performance in the Algebra Problem Solving was computed from students total score of Algebra Problems SolvingRoutine (ALGPS-R) and students total score of Algebra Problems Solving-Non-routine (ALGPS-NR). For comparing the means of ALGPS-R and ALGPS-NR, both scores were changed to the same unit between 0 to 10. The mean of students performance on routine problems was found higher (mean=4.63) compared to performance in non-routine problems (mean=1.63). Table 3 provides details of the students Algebra problem solving performance.

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

407

Table 2: Algebra problem solving performance Routine(R) / Non-Routine (NR) problems R1 R2 R3 R4 NR1 NR2 NR3 ALGPS ALGPS-R ALGPS-NR Valid N 3.2

N 88 88 88 88 88 88 88 88 88 88 88

Minimum .00 2.00 .00 .00 .00 .00 .00 1.29 1.5 .00

Maximum 10.00 10.00 10.00 10.00 10.00 10.00 10.00 6.43 9.25 6.67

Mean 5.7045 8.5795 3.4438 .8131 3.7841 .4886 .5682 3.304 4.63 1.63

Std. Deviation 3.46712 2.39598 3.34917 2.11746 3.44563 1.70189 2.32822 1.1899 1.6497 4.601

Students level of meta-cognitive strategy

Table 3 shows descriptive measures of students level of meta-cognition. The range of summative scores of the meta-cognitive strategy survey was between 0 and 52. The mean of students level of overall meta-cognitive strategy is 1.7312 (ratings measured were from one to two). Students level for each subscale of meta-cognitive strategy is itemized in this table. Findings indicated that the mean level of students meta-cognitive strategy for knowledge subscale is 1.6832, which was rather low whilst the mean score for evaluation and planning subscales were higher. Table 3: Students level of Meta-cognitive strategy Minimum Overall Meta-Cognition Knowledge Planning Evaluation Valid N 3.3 1.46 1.29 1.41 1.39 Maximum 2.13 2.00 2.00 2.00 Mean 1.7312 1.6832 1.7253 1.7727 Std. Deviation .11427 .14393 .13310 .14434

Relationship between students level of cognitive strategy and Algebra problem solving performance with overall mathematical achievement

The Pearson correlation analysis was used to establish relationship between cognitive strategies, Algebra problem solving performance with overall mathematical achievement (refer table 4). The results showed that there is negative and very weak correlation between students ALGPS with shallow cognition(r = -.134, p = .212). Similar, there is no

408

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

significant relationship between ALGPS with deep cognition (r = .124, p = .251). Results in Table 4 also showed that there is significant and strong relationship between students ALGMA (r = .721, p = .000) and ALGPS scores. However, there is no significant relationship between ALGMA scores with level of deep and surface cognition. Table 4: Correlation Analysis ALGPS .721** .000 88 1 88 Deep Cognition .058 .595 88 .124 .251 88 Shallow Cognition -.030 .778 88 -.134 .212 88

ALGMA

ALGPS

Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N

3.4

Relationship between students level of meta-cognitive awareness with Algebra problem solving performance

Results from Table 5 indicate that there is a significant positive and moderate relationship between overall metacognitive strategy with mathematical achievement (ALGMA) (r=.390, sig=.000). However, there is a weak significant relationship between ALGMA with all three subscales of meta-cognition (knowledge, planning, and evaluation) with r= 0.278, sig=.009; r = 0.291, sig = .006; r = 0.273, sig = .010 respectively. Table 5: Correlation Analysis Overall metacognition .390** .000 88 .394** .000 88

Evaluation ALGMA Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N .273* .010 88 .299** .005 88

Planning .291** .006 88 .284** .007 88

Knowledge .278** .009 88 .284** .007 88

ALGPS

The result indicated that, there is a positive and moderate significant correlation between overall meta-cognitive strategy and performance of Algebra problem solving (ALGPS) (r= 0.394, sig=.000). In addition, there is weak significant correlation between ALGPS and meta-cognition subscales (knowledge, planning, and evaluation) (r= 0.284, sig=.007) (r= 0.284, sig=.007) (r= 0.299, sig=.005). 3.5 Differences between students with shallow cognition and student with deep cognition in Algebra problem solving performance and mathematical achievement

The results indicated that students mean mathematical achievement of the deep cognitive strategy students (2) (mean=7.2) is higher compared to students with shallow cognitive strategy (1) (mean= 6.91). However, there is no significant difference in mathematical achievement of students with shallow or deep cognitive strategy, [t(86)= 0.954, p>0.05]. Results presented in Table 6 indicate that mean score of ALGPS of deep cognitive strategy students (mean=3.66) is higher compared to students with shallow cognitive strategy (mean= 3.18). However, the difference is

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

409

not significant. This indicates that for both type of students that is, shallow processor and deep processor performed similarly in ALGPS and ALGMA. Table 6: students mathematical achievement and performance of Algebra problems solving N ALGMA ALGPS ALGPS-NR ALGPS-R 1 2 1 2 1 2 1 2 60 28 60 28 60 28 60 28 Mean 6.9127 7.2000 3.1883 3.6658 1.4111 2.0476 4.5212 4.8795 Std. Deviation 1.13194 1.39310 1.05142 1.40913 1.41470 1.70864 1.49990 1.93992 Std. Error Mean .14613 .26327 .13574 .26630 .18264 .32290 .19364 .36661

Table 7: T-test Analysis Sig. (2tailed) 86 44.277 86 41.568 86 44.934 86 42.647 .306 .345 .079 .118 .070 .093 .346 .392 Mean Std. Error Difference Difference -.28733 -.28733 -.47748 -.47748 -.63651 -.63651 -.35821 -.35821 .27921 .30111 .26904 .29890 .34631 .37098 .37780 .41461

F ALGMA ALGPS ALGPS-NR ALGPS-R Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed Equal variances assumed Equal variances not assumed 1.759 5.641 .077 3.576

Sig. .188 .020 .782 .062

t -1.029 -.954 -1.775 -1.597 -1.838 -1.716 -.948 -.864

df

4.

Conclusion

Performance of problem solving of university students in mathematics is often poor or just on average. This study seeks to identify whether students performance is related to the lack efficient cognitive strategies. Findings indicated that majority of them are shallow processors thus this may explain on their low performance in Algebra. On the whole, the study showed there is no significant correlation between Algebra problem solving performance and both kind of cognitive strategy (deep and shallow). In addition, there is no significant correlation between mathematical achievement and both kind of cognitive strategy (deep and shallow). This study indicated that there is positive, moderate and significant relationship between Algebra problem solving performance, mathematical achievement with meta-cognitive strategies and its subscales. References
Albaili, M.A. (1998). Goal orientations, cognitive strategies and academic achievement among United Arab Emirates college students. Educational Psychology, 18(2), 195-203. Artz, A., & Armour-Thomas, E. (1992). Development of a cognitive-metacognitive framework for protocol analysis of mathematical problem solving in small groups. Cognition and Instruction, 9, 137 175. Berardi-Coleta, B. Dominowski, R.L.,Buyer,L.S., & Rellinger, E.R.(1995). Meta-cognition and Problem Solving: A process-oriented approach. Journal of Experimentl al Psychology, 21, 205-223.

410

Sahar Bayat and Rohani Ahmad Tarmizi / Procedia Social and Behavioral Sciences 8 (2010) 403410

Desoete,A.Roeyers,H. & Buysse.A.(2001) Metacognition and mathematical problem solving in grade 3. Journal of learning Disabilities, 34,435-449. Derry, S.J. & Hawkes, L.W. (1993). Local cognitive modeling of problem-solving behavior: an application of fuzzy theory. In S.P. Lajoie and S.J. Derry, ed., Computers as cognitive tools, pp. 107-140. Lawrence Erlbaum Associates: Hillsdale, NJ. D Corte, E., Opt Eynde, P., & verschaffel, L.(2002). Knowing what to believe. The relevance of mathematical beliefs for mathematics education. In:B.k.Hofer, & P.R.Pintrich(Eds), Personal epistemology: The psychology of belifs about knowledge and knowing (pp.297-320). Elliot.F.S.& Dweck.C.S(1988).Goals: An approach to motivation and achievement. Journal of personality and Social Psychology.54(1).5-12. Fortunato, I., Hecht, D., Tittle, C. K., & Alvarez, L. (1991). Metacognition and problem solving. Arithmetic Teacher, 39(4), 38-40. FLAVELL,J.H.(1979)Metacognitive aspects of problem solving, in RESNICK, L.B.(ed.) The nature of Intelligence, Hillsdale, NJ: Lawrence Erlbaum Associates. Garofalo, J., & Lester, F. K. (1985). Meta-cognition, cognitive monitoring, and mathematical performance. Journal for Research in Mathematics Education,16, 163-176. Greene,B.A., & Miller, R.B.(1996).Influences on course performance: Goals, perceived ability, and self-regulation. Contemporary Educational Psychology, 21,181-192. Graham, S., & Golan, S. (1991). Motivational influences on cognition: Task involvement, ego nvolvement, and depth of information processing. Journal of Educational Psychology, 83, 187-196. Hiebert, J., Carpenter, T. P., Fennema, E., Fuson, K.C., Wearne, D., Murray, H., Human, P., & Olivier, A. (1997). Making sense: Teaching and learning mathematics with understanding. Portsmouth, NH: Heinemann. Horst, H. (2000). A problem solving strategy for gifted learners in South Africa. Gifted Education International. Vol. 15, 103-110. Hutchinson, N. L. (1993). Effects of cognitive strategy instruction on algebra problem solving of adolescents with learning disabilities. Learning Disability Quarterly, 16, 34-63. Hutchinson, N. L. (1993). Effects of cognitive strategy instruction on algebra problem solving of adolescents with learning disabilities. Learning Disability Quarterly,16, 34-63. Kramarski, B., Mevarech, Z. R., & Arami, M. (2002). The effects of meta-cognitive instruction on solving mathematical authentic tasks. Educational Studies in Mathematics, 49, 225-250. Montague, M, Warger, C., & Morgan, T. H. (2000). Solve It! Strategy instruction to improve mathematical problem solving. Learning Disabilities Research & Practice, 15, 110-116. Mayer, R. E. (1998). Cognitive, metacognitive, and motivational aspects of problem solving. Instructional Science, 26, 49-63. Mevarech, Z. R. (1999). Effects of metacognitive training embedded in cooperative settings on mathematical problem solving. Journal of Educational Research, 92, 195-210. Pintrich, P.R. (2000). Multiple goals, multiple pathways: The role of goal orientations in learning and achievement. Journal of Educational Psychology, 92, 544-555. Schoenfeld AH 1985. Metacognitive and epistemological issues in mathematical understanding. In: S ilver EA (ed.). Teaching and learning mathematical problem-solving: multiple research perspectives. Hillside, NJ: Erlbaum. Somuncuoglu, Y., & Yildirim, A. (1999). The relationship between achievement goal orientations and the use of learning strategies. The Journal of Educational Research, 92(5), 267-277. Teong, S. K. (2003). The effect of metacognitive training on mathematical word-problem solving. Journal of Computer Assisted Learning, 19(1), 46 55. Woodward, J., & Montague, M. (2002). Meeting the challenge of mathematics reform for students with LD. Journal of Special Education, 36 Wolters, C.A. (2004). Advancing achievement goals theory: Using goal structures and goal orientations to predict students motivation, cognition, and achievement. Journal of Educational Psychology, 96, 236-250.