Available online at www.sciencedirect.com

ScienceDirect

Procedia - Social and Behavioral Sciences 90 (2013) 914 - 922

6th International Conference on University Learning and Teaching (InCULT 2012)

A Study on the Significance of Students' Thinking Level to

Students' Performance

Zaid Mujaiyid Putra Ahmad Baidowia*,Norzaidah Md Nohb, Nor Akmal Md Nohc

aUniversiti Teknologi MARA, Kampus Pumcak Alam, 42300 Bandar Puncak Alam, Selangor bUniversiti Teknologi MARA Malaysia, 40450 Shah Alam, Selangor cUniversiti Teknologi MARA Johor, 85000 Segamat, Johor

Abstract

Students' level of thinking can be measured in terms of their problem solving, logical thinking and cognitive skills. The aim of this research is to prove the hypotheses, that the students' level of thinking is not a significant factor for students to achieve good result in C programming course. This course is a core subject offered to engineering foundation students in Centre of Foundation Studies (CFS), Universiti Teknologi MARA (UiTM), Malaysia. In Malaysian education system, students have been exposed to these three skills during their secondary school in Mathematics subject. Students are expected to possess problem solving, logical thinking and cognitive skills in order to acquire programming skill. In this research, we measure the students' thinking level as the independent variable. This is achieved by distributing a set of questionnaire to 170 respondents out of 700 students. The questionnaires were evaluated by using rubrics table and the score was based on Likert Scale. Students' score for the course, which includes the course work and final examination marks, were named as dependant variable. SPSS is used to explore the impact of students' thinking level as a significant factor to C programming course. In this case study, it is found that the students' thinking level is not a significant factor to the students' score in this course.

©2013TheAuthors.Published byElsevier Ltd.

Selectionand/orpeer-reviewunderresponsibility oftheFacultyofEducation,UniversityTechnologyMARA,Malaysia. Keywords: C Programming; Students' Level of Thinking; Significant Factor; Engineering Foundation Students; Skills in Programming.

1. Introduction

C programming is taught to engineering foundation students, UiTM Puncak Alam in the second semester. Teaching and learning process are implemented via mass lectures and laboratory sessions. Many approaches were carried out to teach students on C programming such as lectures, collaborative, problem solving

* Corresponding author.

E-mail address: zaidmputra@gmail.com

1877-0428 © 2013 The Authors. Published by Elsevier Ltd.

Selection and/or peer-review under responsibility of the Faculty of Education, University Technology MARA, Malaysia. doi:10.1016/j.sbspro.2013.07.168

as well as drill and practice. During mass lectures, students are taught theories and the conceptual of the subject matter. In lab sessions, students learned the subject by doing hands-on activities. In addition, mini-lectures were conducted during lab sessions before starting their lab exercises. The reason was to review the key points of the topic for the day to ensure the students understand the subject matter.

Observations by lecturers towards the students were carried out for the whole semester of their studies and it was found that students were facing difficulties in understanding the programming concepts and therefore they were unable to solve tasks given during lab sessions. The students took more than the time provided to complete a simple programming task with inaccurate answers. This is supported by a research that investigated the nature of the academic problem faced by novice programming students. It reported that students feel difficult in understanding and implementing both low level-programming concepts, such as syntax and variables, as well as high-level concepts, such as OOP principles and efficient program design (Butler & Morgan, 2007). This problem occurred due to zero knowledge in programming that they have never learnt during their secondary school.

Thus, the aim of this paper is to identify whether the students' level of thinking is a significant factor for students to achieve good results in C programming course. What is the common thinking level of the CFS students and how commonalities affecting their performance of C programming course is the research question to support the objective.

2. Literature review

One of the reasons why the students feel programming is difficult is that they are comfortable to learn and work with programming syntax rather than understand the programming concept (Butler & Morgan, 2007). This causes the students unable to solve other problems in real life environment which needs logical and conceptual thinking. In Malaysia, the main factor of the difficulty of programming course is due to lack of knowledge in programming. This course is new to the Malaysian students because they never learnt it during their secondary school. Therefore, when they entered this university and took this course, they feel that this was something hard for them to learn. In addition, many students also do not have enough mathematical, logical knowledge and problem solving skills (Gomes & Mendes, 2007). Sometimes, students are not sure how programming structure works in a computer and it could result in wasting time when computer run the programs. In fact, the computer can save the time to process other programs.

Before starting the hands-on activities on programming, lectures are conducted to ensure students get proper knowledge especially in understanding programming concept. However, most of the students fail to apply correctly the knowledge and the concept (Ismail, Ngah, & Umar, 2010) they have learnt. In order to make the students understand well the concept of the programming, students could learn the course by using visualization learning (Mierlus, 2006).

Programming requires students to analyze and solve problems. Thus, an effective cognitive strategy is needed for students to solve the problems (Ismail, Ngah, & Umar, 2010). Cognitive skills are skills that require human mind or mental to work (Liu, 2003). These skills are more complicated compared to physical skills because the skills require more time to be developed during learning process (Patel, Kinshuk, & Russell, 2000). According to (Sternberg, 1998), problem solving is one of the categories in cognitive skills. Therefore, the use of cognitive skills is very important in this programming course.

Computer engineering depends on mathematical logic and therefore computer engineers are expected to be fluent in logical thinking (Holvikivi, 2007). By possessing logical skill, students could think logically to solve problem and then converts into programming language in logic manner. By having good logical thinking skill too, students could gain superior performance (Al-Imamy, Alizadeh, & Nour, 2006) in this course. Technically, every desired output must be derived from input data which must be processed before it can be produced. In other words, the outputs are uniquely determined by inputs of the problems (VanLehn, 1996).

Besides logical thinking, students should posses another important skill in dealing with programming, that is problem solving. Problem solving is a mental activity and it helps students a lot in understanding programming concept (George, 2000). By having problem solving skill, students could analyze problems and synthesize a solution (Coles, Jones, & Wynters, 2011) and thus relate to the real world problem.

In general, we can say that there are three skills, problem solving, logical thinking and cognitive skills, that the students should have in order to be good in programming. In Malaysian education system, students have been exposed to these three skills during their secondary school in Mathematics subject.

3. Sample

In November 2009, Centre of Foundation Studies (CFS) UiTM was established to prepare students to pursue degree programs in critical areas in other public Higher Learning Institutions and UiTM itself. There are 5 pre-degree programs offered: 2 programs for Law, 1 program for Engineering, 1 program for TESL and 1 program for Science.

However, in CFS, UiTM Puncak Alam, there are two pre-degree courses offered; Engineering and Science. The minimum requirement for the students to get enrolled in the Pre-degree Engineering Program is two distinction A- grades and two credit B grades for the following 4 subjects: Physics, Chemistry, Mathematics and Additional Mathematics. Meanwhile, the minimum requirement for the students to get enrolled in the Pre-degree Science Program is two distinction A- grades and two credit B grades for the following 4 subjects: Physics, Chemistry, Biology and Additional Mathematics. A part from that, the students must score at least a credit B grade in English and a credit C grade in Bahasa Malaysia. The duration of the program is 2 semesters to ensure that the students acquire the basic knowledge to undertake any of the engineering or science and technology degree programs. The programs require students to complete six courses each semester. There are: Chemistry, Mathematics, Physics, Computing/Biology, English and Religious Study.

Students who are accepted into one of the programs would require completing their studies within 2 semesters with a total of 50 credit hours. In general, the students' assessment for each subject would be based on assignments, laboratory reports, quizzes, tests and final examination. Other than that, the students must also involve in co-curriculum activities as their merit points when applying degree programs.

Thus, this case study was implemented in UiTM Kampus Puncak Alam. The target respondents were engineering students in CFS. The total number of the engineering students is around 700 students. There were 170 questionnaires distributed to five groups of engineering students in CFS using the cluster sampling techniques. However, the number of papers returned back with complete answers was 145 forms.

4. Instruments and Methodology

The questionnaire was adopted from India BIX (2008-2012) as the instrument in this case study. There are a lot of questions developed in this website but a few questions were selected based on problem solving, logical and cognitive thinking categories. All the related questions were adopted and no changes were made to the questions. However, the questions are reliable based on reliability test value 0.97 by using Rasch Model. As for evaluation of the three thinking categories, we evaluated the students' answers based on our own scale (Trochim, 2006) as illustrated in Table 1.

The survey was divided into two sections: Demographic and Skills Evaluation. Demographic section was created to know the students' background in their previous education as well as their background in programming. Meanwhile, in skills evaluation, there are three sections: Problem Solving, Logical and Cognitive thinking skills. Skills evaluation section was to evaluate the students' level in Problem Solving, Logical and Cognitive thinking skills. In these sections, the questions were given with four possibility answers. The students were required to answer by giving the explanation to their answers. The students were evaluated based on their

answers and explanation but the explanation must be logical and acceptable. Likert scale (Table 1) was applied in this questionnaire to evaluate the students' level of thinking for the three skills.

In Problem Solving section, respondents were given questions to measure their skill in solving problems (see appendix A.1). Ideally, the questions in this section were quite similar to mathematical problem solving. The questions were constructed by using descriptive mathematical questions. The respondents were required to answer and give the explanation based on their understanding. The explanation enlightened the way of their problem solving thinking.

In contrast, Logical Thinking section, respondents were given a set of logical expression for each question (see appendix A.2). This section measured the respondents' skill in logical thinking. By reading the expressions, the respondents were expected to answer based on their logical thinking. Similarly, they were required to explain why they chose the answer based on their understanding. The explanation rationalized their answers.

Nevertheless, in cognitive section, respondents were given questions to measure their mental or mind thinking (see appendix A.3). Descriptive questions of four sentences were asked and the respondents were required to answer and explain why they chose the answer. The explanation enlightened their state of mental thinking.

Table 1. Likert Scale used in this Research Questionnaire to Evaluate Students' Level of Thinking.

Scale Description

1 No explanation OR the explanation does not make sense AND the answer most probably is wrong.

2 No explanation OR the explanation does not make sense BUT the answer might be right.

3 The explanation is not clear / accurate BUT the answer might be right.

4 The explanation is clear AND the answer might be right.

5 The explanation is very clear AND the answer might be right according to their understanding.

All scores were recorded and mean was calculated for the three categories: Problem Solving, Logical and Cognitive thinking Skills for each students. The mean would be used to determine the average level of all respondents. Table 2 below shows Likert Scale (Trochim, 2006) that was used in determining the level of the students' thinking skills as a whole.

Table 2. Likert Scale used in this Research Questionnaire to Determine Students' Level of Thinking. Scale Description

1 Very Weak

2 Weak

3 Average

4 Good

5 Excellent

Samples were analyzed quantitatively by using MS Excel and SPSS software package for descriptive statistics and the correlation between variables. The questionnaire is in Likert items (scale 5) but this study focuses more on score of each category in the thinking level. Thus, quantitative analysis was developed in this case study. The test of independency of variables was tested using ANOVA since the evaluation was based on four variables simultaneously. The impact of thinking level towards the C programming course performance was analyzed using the regression model. In this method, the dependent variable (Y) is result of C programming course which

combination of assessment marks and final exam result. While, the independent variables (X) are variables discussed previously: problem solving, logical and cognitive thinking.

5. Findings and Discussion

5.1. Students' Level of Thinking

Based on analysis of the data, we found that the average of students' level for Problem Solving is 3.48 and it shows that most of the respondents are extremely more than average in their Problem Solving. In the meantime, the average students' level for Logical Thinking is 3.04 and it shows that most of the respondents are more than average in Logical Thinking. In contrast, the average students' level for Cognitive is 3.53 which are almost good. The result is illustrated in Table 3.

The highest score for Problem Solving, Logical Thinking and Cognitive is 5.00. The lowest scores for Problem Solving, Logical Thinking and Cognitive are 1.33, 1.33 and 1.67 respectively. There are 5 (3.42%), 1 (0.68%) and 3 (2.05%) number of students scored 5 for Problem Solving, Logical and Cognitive respectively. The students who scored highest in Problem Solving are different with Logical Thinking and Cognitive. There are 3 (2.05%), 2 (1.37%) and 1(0.68%) number of students who got lowest score in Problem Solving, Logical Thinking and Cognitive. The result is illustrated in Table 4.

Table 3. Students' Level of Thinking - Average, Highest and Lowest.

Level Type of Thinking (Mean Score)

Problem Solving Logical Thinking Cognitive

Average 3.48 3.04 3.53

Highest 5.00 5.00 5.00

Lowest 1.33 1.33 1.67

Table 4. Students' Level of Thinking - No. of Students and Percentage.

Level Type of Thinking

Problem Solving Logical Thinking Cognitive

No. of % No. of % No. of %

Students Students Students

Highest 5 3.42 1 0.68 3 2.05

Lowest 3 2.05 2 1.37 1 0.68

5.2. On-Going Assessment and Final Exam Result

Total marks for on-going assessment (OGA) were 70% which consisted of laboratory exercises, quizzes and tests. The average mark for OGA was 51.69%. The number of students, who scored 5 either Problem Solving or Logical Thinking, was 4 people. However, there was 1 student who scored 5 in Problem Solving but did not managed to get over the average mark. Overall, out of 146 students, there were 77 (53%) students who scored more than the average mark. Meanwhile, 47% students scored less than the average mark.

As for final exam (FE), the total marks were 30% which consisted of two sections, objectives and structure. The average mark for FE was 20.80%. There were 3 out of 4 students that scored 5 in Problem Solving or Logical Thinking, managed to get marks more than the average mark. Overall, there were 73 (50%) students managed to get marks more than the average mark. However, there was only one student, who scored 3.5 in

Problem Solving and 4.0 in Logical Thinking, failed the course due to attitude problem. The students' performance in both on-going assessment and final exam can be illustrated in table 5 and table 6.

Table 5. Students' Performance for On-going Assessment and Final Examination.

Level OGA (70%) FE (30%)

Highest 64.41 28.95

Average 51.69 20.80

Lowest 30.12 6.30

Table 6. The Students Scored more than Average Marks for On-going Assessment and Final Examination.

More than 51.69 More than 20.80

Description out of 70% out of 30%

No. of Students 77 73

% of Students 53 50

5.3. Students' Performance after Attending C Programming Course

In general, 3 out of 4 students that scored 5 in either Problem Solving or Logical thinking scored A for the subject. However, the other one student scored B+ for the subject. Overall, there were 63 (43%) students scored A regardless of their level of thinking.

Table 7. Students' Performance after Attending C Programming Course

Grade No. of Students Percentage

A 63 43

A- 20 14

B+ 25 17

B 18 12

B- 10 7

C+ 6 4

C- 0 0

GA 1 1

Therefore, we can conclude that the students' level of thinking at the early semester does not determine their result of C programming course. During the course, the students learn numerous and valuable new knowledge in programming including thinking logically and solving problems. Therefore, problem solving and logical thinking skills can be developed by attending this course. This statement is supported by (Woods, 1997).

5.4. Correlation Analysis

The above four variables previously discussed (students' result, problem solving skill, logical thinking skill, and cognitive skill) were tested to see the distribution and the independency before the correlation are carried out.

All four variables are normally distributed (see appendix) for the sample size of 145 students. Besides that, the variables of problem solving, logical and cognitive thinking were treated as interval measurements. ANOVA was used to test the independencies between three variables; problem solving, logical, and cognitive thinking simultaneously. The three variables were significantly different and independent to each other since the F statistics value 754.799 with p-value 0 as shown in Table 8.

Table 8. ANOVA Result

Sum of Variation Degree of Freedom Sum of Square Mean Square F Statistics p-value

Between 1 239894.6 239894.6 754.7999 0

Within 578 183703.1 317.8

The correlation of the three variables has been tested towards the performance of C programming course using Regression model. In order to achieve the objective of this study, the relationship between students result, problem solving, logical and cognitive thinking were studied. As shown in Table 9 the p-value is 0.04717 which is more than a value (we set a at 1%). Therefore problem solving, logical and cognitive thinking skills are not significant factor to C programming course result with p-value 0.1072, 0.6118 and 0.0412 respectively. It can be concluded that, these two skills are not important prior to students enrolling in C programming course. These two skills might be developed during the course itself. However, it is not compulsory that students must possess these skills before they enroll in the course.

Even though problem solving skill could be categorized under cognitive skill, this study shows slightly different result. From the result, the cognitive variable is significant to the C programming result if the value is compared to a at 5% level of significance. Meanwhile, the impact of the cognitive skills toward the C programming result has weak positive linear association if we take the square root of R2 value (r = 0.2959).

Table 9. Regression Analysis Result

Variable Value Std. Error t value Pr(>| t |

(Intercept) 63.7948 2.8486 22.3951 0.0000

Problem 0.4313 0.2660 1.6211 0.1072

Logic -0.0793 0.1558 -0.5086 0.6118

Cognitive 0.6295 0.3055 2.0604 0.0412

Residual standard error: 8.615 on 141 degrees of freedom Multiple R-Squared: 0.08754

F-statistic: 4.509 on 3 and 141 degrees of freedom, the p-value is 0.04717

6. Conclusion and Future Work

In this research, we conducted a survey to prove that the students' level of thinking is not a significant factor for students to achieve good result in C programming course. Questionnaires were distributed to a few groups of students and evaluated based on their answers and explanation. The results were kept for record and run by using SPSS to prove the hypotheses. Reliability test was done using Rasch Model to check for the questionnaires reliability. The result obtained from this research proved that the students' level of thinking is not a significant factor for students to achieve good result in C programming course. It was also found that the skills could be developed by attending the course.

There are a few limitations in this research that can be improved in future work. Pre-questionnaire (pre-test) must be distributed to the students in early semester while the same set of questionnaire should be distributed as post-questionnaire (post-test) at the end of the semester. Hence, we could clearly see the improvement of thinking level for each student after attending the course.

Acknowledgements

The authors would like to acknowledge Centre of Foundation Studies, UiTM Kampus Puncak Alam for the financial support of the conference fee. A note of gratitude is addressed to the members of the research group for the commitment, support and encouragement throughout this research.

References

Al-Imamy, S., Alizadeh, J., & Nour, M. A. (2006). On the Development of a Programming Teaching Tool: The Effect of Teaching by

Templates on the Learning Process. Journal of Information Technology Education , 271 - 283. Butler, M., & Morgan, M. (2007). Learning challenges faced by novice programming students studying high level and low feedback

concepts. ICT: Providing choices for learners and learning, (pp. 99 - 107). Singapore. Coles, D., Jones, C., & Wynters, E. (2011). Programming Contests for Assessing Problem-solving Ability. The Journal of Computing Sciences in Colleges , 28.

George, C. E. (2000). Experiences with Novices: The Importance of Graphical Representations in Supporting Mental Models. 12th Workshop

of the Psychology of Programming Interest Group. Cozenza Italy: PPIG. Gomes, A., & Mendes, A. (2007). Learning to program - difficulties and solutions. International Conference on Engineering Education. Coimbra, Portugal.

Holvikivi, J. (2007). Logical Reasoning Ability in Engineering Students: A Case Study. IEE Transactions on Education. Finland.

India BIX. (2008-2012). Retrieved 11 2011, from www.indiabix.com:http://www.indiabix.com/logical-reasoning/logical-problems/026001

Ismail, M. N., Ngah, N. A., & Umar, I. N. (2010). Instructional Strategy in the Teaching of Computer. TOJET: The Turkish Online Journal of

Educational Technology , 125 - 131. Liu, M. (2003). Enhancing Learners' Cognitive Skills Through Multimedia Design. Interactive Learning Environments (pp. 23 - 39). Swets & Zeitlinger.

Mierlus. (2006). Codewitz Learning Objects. 6th International Educational Technology Conference , (pp. 1222-1230). Famagusta, North Cyprus.

Patel, A., Kinshuk, & Russell, D. (2000). Intelligent Tutoring Tools for Cognitive Skill Acquisition in Life Long Learning. Educational Technology & Society , 32-40.

Sternberg, R. J. (1998). Thinking and Problem Solving: Handbook ofperception and cofnition (2nd ed.). Burlington, MA: Academic Press. Trochim, W. M. (2006, 10 20). Likert Scaling. Retrieved 04 15, 2012, from Research Methods Knowledge Base:

http://www.socialresearchmethods.net/kb/scallik.php VanLehn, K. (1996). Cognitive Skill Acquisition . Annu. Rev. Psychol. Pittsburgh, Pennsylvania: Annual Reviews Inc. Woods, D. R., Hrymark, A. N., Marshal, R. R., Wood, P. E., Crowe, C. M., Hoffman, T. W., et al. (1997). Developing Problem Solving Skills: The McMaster Problem Solving Program. Journal of Engineering Education , 75-91.

Appendix A. Sample Questions

A.1. Question that measured Problem Solving skill of the respondents (India BIX, 2008-2012).

Three years back, a father was 24 years older than his son. At present the father is 5 times as old as the son. How old will the son be three years from now?

a. 12 years b. 6 years c. 3 years d. 9 years e. 27 years

Your Answer: Your Explanation:

A.2. Question that measured Logical Thinking skill of the respondents [ 13].

Fact 1: All dogs like to run.

Fact 2: Some dogs like to swim.

Fact 3: Some dogs look like their masters.

If the first three statements are facts, which of the following statements must also be a fact?

I: All dogs who like to swim look like their masters.

II: Dogs who like to swim also like to run.

III: Dogs who like to run do not look like their masters.

a. I only

b. II only

c. II and III only

d. None of the statements is a known fact.

Your Answer: Your Explanation:

A.3. Question that measured Cognitive skill of the respondents [13].

Ten new television shows appeared during the month of September. Five of the shows were sitcoms, three were hour-long dramas, and two were news-magazine shows. By January, only seven of these new shows were still on the air. Five of the shows that remained were sitcoms.

A. Only one of the news-magazine shows remained on the air.

B. Only one of the hour-long dramas remained on the air.

C. At least one of the shows that was cancelled was an hour-long drama.

D. Television viewers prefer sitcoms over hour-long dramas.

Your Answer: Your Explanation: