Which Factors Influence the Academic Performance of Students?

TABLE OF CONTENTS

CONTENTS

ABSTRACT

ABBREVIATION

TABLE OF CONTENTS

CHAPTER ONE

INTRODUCTION

CHAPTER TWO

METHODS
Data Collection
Statistical Methods used in the study
Multivariate Analysis

CHAPTER THREE

RESULTS AND DISCUSSION

Performance of students in the eight departments
Performance of students who followed Crop Science modules
Factors Relating to the Performance
A/L Z-Score
Gender
Ethnicity
Socio Economic Status
Attempts
General Knowledge
A/L English
Agriculture/Physics at A/L
Vacancy Filling/ 1st selection
District
Multivariate Analysis
General Linear Model
Principle Component Analysis
Subject Classification
Advanced Crop Production Technology
Plantation Management and Forestry

CHAPTER FIVE

CONCLUSION

REFERENCES

FACTORS RELATING TO PERFORMANCE OF STUDENTS IN THE CROP SCIENCE MODULES (2002-2012)

ABSTRACT: As the leading agriculture higher education institute, Faculty of Agriculture, University of Peradeniya offers three valuable degree programs to fulfill the agriculture-based knowledge requirement of the country. Factors relating to academic performance might be different in Universities as well as in different departments of a given Faculty. This study aims to investigate the factors relating to performance of students who followed the Crop Science modules in the Faculty of Agriculture, University of Peradeniya from 2002 to 2012. Secondary data were collected from the Senate Office of the University and the Administrative office of the faculty. For data analysis, Box plot, Two sample t-test, Spearman correlation coefficient, Pearson’s correlation coefficient, General Linear Model (GLM) of SAS, Simple Linear Regression, Principle component Analysis and Cluster Analysis were used. The performance of students in the Crop Science modules was also compared with other department. Using Cluster Analysis, the subjects of the two Crop Science modules were classified.

According to the results, the different performances were found in different departments and performance of students in the Crop Science department was at an average level. The performance of students of the Crop Science department was lower than other departments. However, this was mainly due to large number of students in the Crop Science modules. Due to this the average performance gets pulled down. When only the top 15 students were considered the performance of students in the Crop Science modules was very high. A significant difference was found between the two Crop Science modules in terms of GPA and the better performance was observed in Advanced Crop Production Technology Students. Correlation analysis showed that the A/L Z-Score, Gender, District, Socio-Economic status, number of A/L attempts, A/L English grade and Agriculture/Physics as an A/L subject were all significantly related to academic performance of the Crop Science students while Vacancy Filling or 1st Selection, Ethnicity, General Knowledge at A/L did not affect the performance significantly. GLM analysis indicates that English was the highest related factors to performance followed by Z-score and Gender. Basically, social status of the students, cognitive capacities of the students and the preference of the students to the Agriculture subject are the common factors relating to the performance of the students in Crop Science Majoring module.

Key words: Variables relating to students’ performance, GPA, multivariate analysis

ABBREVIATIONS

Abbildung in dieser Leseprobe nicht enthalten

CHAPTER ONE

INTRODUCTION

Education is one of the essential factors needed to accomplish dynamic and challenging future opportunities. Sri Lanka's education structure is divided into three parts: primary, secondary and tertiary. Tertiary education is given at universities in Sri Lanka. Admission to the university system is based on the highly competitive GCE Advanced Level examination and on successful completion of the secondary exams, students can move on to tertiary education. Out of the education system in Sri Lanka, The higher education system plays a big role to fulfill the social aspects.

Being an agricultural country, there are seven Universities in Sri Lanka offered Agriculture-based degree courses and out of those, Faculty of Agriculture, University of Peradeniya is the oldest, largest and leading agricultural Faculty in Sri Lanka. The faculty offers three BSc (Agriculture) courses namely Agriculture Technology and Management (ATM), Food Science and Technology (FST) and Animal Science and Fisheries (ASF). Out of the other B.Sc. (Agriculture) courses, Agricultural Technology & Management offered by Faculty of Agriculture, University of Peradeniya is unique in term of the curricular and its own recognition. Being a four year special degree, the ATM degree course intakes 200 students proposed by University Grants Commission (UGC) in order of Z - Scores ranked merit wise and according to the district quota.

The ATM degree is a four year special degree and the course is conducted as a semester based course unit program. The ATM students follow the common subjects up to 3100 semester and thereafter follow an Advance Program in a particular department in accordance with the preferences of students during 3200 to 4200 semesters. This selection may be influenced by various factors such as CGPA, employment opportunities, districts they come from and personal preference.

Literature reveals that the interactions of gender with the department that the students were most competent and the students’ selections of majoring departments are significant. It indicates specially that the associations are evident between the gender and the department students were most competent for the AB, AE, EB, EX, SS, and AS departments (Thattil and Nalaka, 2009). This evident suggests that students select a majoring module based on their competency in such subjects. This may ultimately lead to increased performance during the Advance Program.

Being one of eight departments, the Crop Science department comprise two majoring module namely Advance Crop Production and Technology and Plantation Management and Forestry. The courses of Advance Crop Production and Technology have been designed to impart knowledge and skills required to practice advance crop production and technology and the courses of Plantation Management and Forestry have been designed to impart knowledge and skills required to manage a plantation more efficiently and use forestry concepts in land and environmental managements (Anonymous, 2012). One fourth of ATM students come to the Crop Science department to follow Crop Science majoring modules in each year. It indicates the importance of the Crop Science department in the Faculty. Hence, it is worth to investigate the factors relating to academic performance of crop Science students.

Quality of a course depends on content of that course. Even though the content is good it does not ensure that those knowledge and skills are effectively transferred to students. The transformation is measured at the faculty examination and it indicates by a single value called as GPA. Different subjects need different knowledge and skills to be performed well. In view of the variability of subjects in the curricular, application of the same curriculum changes (e.g. strategies) to the entire set of subjects is inappropriate. Hence, in order to apply the same strategy generally to a large number of subjects, grouping is necessary (Thattil and Nalaka, 2009). It clearly shows that subjects in the same group require similar strategies and skills to perform well in those subjects. it also helps students to exert differently in accordance with subjects in different groups to obtain academic success at exams.

A great deal of heterogeneity in terms of university entrance Z score, gender, number of attempt in A/L for university entrance, English proficiency, district, ethnicity and socio-economic back ground exit among the students. The performance of students may relate to those entry qualifications and their Socio economic back ground. Investigation of those factors is necessary to enhance the performance of students. Those complementary factors will help students to achieve their potential performance.

Many studies have been carried out on the factors affecting the students' academic performance in university examinations in Sri Lanka. Many of these studies based upon university entrance selection criteria. Some studies argue that those criteria do not affect the academic performance while other studies argue that there is an effect of those criteria on academic performance of undergraduates. Hence, factors relating to academic performance might be different in Universities as well as in different departments of a given Faculty. Therefore, this study was conducted to determine the factors relating to the academic performance of students following the Crop Science module in the Faculty of Agriculture University of Peradeniya.

CHAPTER TWO

METHODS

The study was done to investigate the factors relating to the academic performance of the student doing their final year project in the crop science department using information collected on the student who entered the Faculty of Agriculture, University of Peradeniya in the period 1998/1999 to 2007/2008 academic years. Data were collected on gender, ethnicity, university entrance as vacancy filling or 1st selection, number of attempts, agriculture or physics as a A/L subject, A/L English, Z-score, general knowledge marks at A/L, District, CGPA at 3100 semester, CGPA during majoring module and Final GPA of selected students. CGPA at 3100 semester, CGPA during majoring module and final GPA were collected on each student in each department.

For subject classification, data were collected on students who entered the university the period 2005/2006 to 2007/2008 academic years and followed the new curricular. There were 90 and 36 students who followed Advanced Crop Production Technology and Plantation Management and Forestry modules respectively. Subject grades taken for the crop subjects during the advanced Programme were collected for subject classification.The performance of students in the Crop Science modules was also compared with other department.

Data Collection

The variables used in the study and the codes used in entering data in a Microsoft excel sheet are given in table 3.1.

Table 3.1: Variables and codes used

Abbildung in dieser Leseprobe nicht enthalten

Statistical Methods used in the study

Table 3.2: Statistical Methods used in the study and their purpose

Abbildung in dieser Leseprobe nicht enthalten

Multivariate Analysis

Principle component analysis (PCA)

In this study Principle Component Analysis was done using standardized data to categorize the important variables into uncorrelated common factors.

Principle Factor Analysis (PFA)

Principle factor analysis was performed to identify the common factors which are related to the academic performance of the students.

Cluster Analysis

Crop subjects of advanced crop production Technology and Plantation management majoring module were categorized using PCA and cluster analysis.

CHAPTER THREE

RESULTS AND DISCUSSION

Performance of students in the eight departments

The Crop Science Department is the largest department of the faculty in term of Numbers of the students followed the Advanced Programme. Numbers of the students in each department during 6th (3200) to 8th (4200) semesters are given in Figure 4.1. About 26% of students had come to the crop science department to follow advanced program during 1998/1999 – 2007/2008 academic years. It was over one fourth of each batch in each year. The lowest number of students entered the Soil Science Department.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.1: Students Distribution among the departments

CGPA at 3100 semester is one of the factors that the selection of departments for advanced Programme depends on. This is disclosed by General linear model (GLM) analysis of CGPA. The result (p = <.0001) has shown that CGPA up to 3100 semester were significantly different among the eight departments (Table 4.1).

Table 4.1: Results of the Duncan’s n mean separation method for CGPA up to 3100 semester

Abbildung in dieser Leseprobe nicht enthalten

*Means with the same letters are not significantly different.

The highest average CGPA was found in Agricultural Biology, Agricultural Economics and Business and Food Science & Technology departments while the lowest average CGPA was found in Crop science and Agriculture Engineering departments. The box plot diagrams obtained for the eight departments visualized that CGPA distributions were not similar (Figure 4.2).

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.2 : Box Plot of the CGPA distribution in each department

The distribution of the CGPA of students’ in the Crop Science Department was more symmetrical compared to the other departments. Hence lower CGPA holders as well as higher CGPA holders were present in the Crop Science department and it lead to the decrease in the mean CGPA of the department. Because of the highest number of students were in Crop Science department, average CGPA can be reduced due to the low performers.

When analyzing CGPA of the 15 highest CGPA holders in each department in each year, the result (p = <.0001) showed that the highest CGPA holders follow the advanced Programme in Agricultural Economics and Business, Crop Science, Agricultural Biology and Food Science & Technology departments respectively while the lowest CGPA holders were in Animal Science, Agricultural Extension, Agriculture Engineering and Soil Science departments respectively (Table 4.2).

Table 4.2: Results of the Duncan’s mean separation method CGPA of the 15 highest CGPA holders

Abbildung in dieser Leseprobe nicht enthalten

*Means with the same letters are not significantly different

During the advanced Programme students’ performance generally increase. But the Increment of the performance during the advance Programme differs from a department to department. GLM analysis indicated that CGPA during the advance Programme in each department are significantly different (p = <.0001). The results are shown in table 4.3.

Table 4.3: Results of the Duncan’s mean separation method for CGPA during the advance Programme

Abbildung in dieser Leseprobe nicht enthalten

*Means with the same letters are not significantly different

Performances of the students of the Agricultural Biology and Soil Science departments were significantly different from the other departments. The Agricultural Economics and Business Management, Crop Science and Agricultural Extension department have performed moderately while Food science and Technology, Agriculture Engineering and Animal science departments’ students have performed poorly compared to the other departments. Mean CGPA of the students during the advanced Programme are comparatively good in the Crop Science department.

FGPA of each department were also significantly different (p=<.0001). FGPA analysis results were also same as CGPA up to 3100 semester with some exceptions (Table 4.4). Students who followed the Crop Science majoring modules increase their performance and finally, it lead to increase FGPA, as seen by table 4.1 and 4.4. This is generally true for all departments. The highest increase was seen in the soil science department.

Table 4.4: Results of the Duncan’s mean separation method for FGPA

Abbildung in dieser Leseprobe nicht enthalten

*Means with the same letters are not significantly different

Performance of students who followed Crop Science modules

The two modules in the Crop Science Department are not the same in all aspects. The performances of the students in these two modules are also different. Therefore, the two sample t-test was performed to analyze the CGPA, Majoring CGPA and FGPA of the students in Advanced Crop Production Technology and Plantation Management and Forestry modules.

Results of t-test between CGPA and the Crop Science modules (p <.0001) shows that there is significant difference between the two modules. Mean CGPA of Advanced Crop Production Technology and Plantation Management and Forestry modules were 3.07 (SD=0.34) and 2.75 (SD = 0.39) respectively. Performance of the students who followed Advanced Crop Production Technology modules was better than Performance of the students who followed Plantation Management and Forestry module. The majoring results were significantly different in the two module and the students in Advanced Crop Production Technology obtain higher GPA than Plantation Management & Forestry (p=0.0094). Mean CGPA during the Advanced Programme of the students in Advanced Crop Production Technology and Plantation Management and Forestry modules were 3.37 (SD=0.28) and 3.16 (SD=0.42) respectively.

Results of t-test between FGPA and the Crop Science modules (p<.0001) indicate that the performance of the students were also significantly different in the two modules. Mean FGPA during the Advanced Programme of students in Advanced Crop Production Technology and Plantation Management and Forestry modules were 3.15 (SD=0.32) and 2.87 (SD=0.38) respectively.

Mean CGPA of the students follow the Crop Science modules has been changing from year to year (Figure 4.3). The mean performance during the Advance Program has been decreasing with the time with the peak in 2001. The gaps between mean CGPA up to 3100 semester, mean CGPA during the advanced program and mean FGPA have been decreasing with time.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.3: Performance of the Crop Science students over the time

Factors Relating to the Performance

Correlation of nine factors to CGPA up to 3100, CGPA during Advanced Program, and FGPA were analyzed and the results are given in table 4.5. The results of the correlation analysis showed that A/L Z-Score, Gender, Socio-Economic Status, A/L attempt, A/L English marks and Agriculture/Physics as an A/L subject were all significant predictors of academic performance of Crop Science students of Faculty of agriculture, University of Peradeniya.

Table 4.5: Correlation between the Factors and GPA

Abbildung in dieser Leseprobe nicht enthalten

A/L Z-Score

Minimum and Maximum Z-Score values of the sample were 0.5994 and 1.9849 respectively and Average Z-Score of the students of the Crop science students was 1.5087(SD=0.22102). According to the results given in Table 4.5, Z-Score is positively but weakly correlated with GPA.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.4: Result of Regression Analysis between FGPA and Z-Score

FGPA = 2.5575 + 0.2909*Z-Score (eq. 4.1)

Regression Analysis also showed a positive effect of Z-Score on FGPA significantly (p= 0.0043) (eq. 4.1). When the Z-Score increase performance of students increase and vice versa. Correlation shows that the effect of Z-Score on GPA fades over the time during the eight semesters.

When average Z-Score of students in a course increase, the effect of Z-Score on performance increase and vice versa. This is proved with the evidence of past literature. Hewage et al. (2011) shows a significant positive relationship between A/L results and performance medical undergraduates who have mean Z-Score over 2.00. But it does not affect the performance of the undergraduates in the Faculty of Agriculture, University of Ruhuna (Sandika et al., 2012).

It is said that performance of undergraduate students do not correlate with merit, district basis or underprivileged as selection criteria of the Computer Science and Statistic students in University of Kelaniya (Hewapathirana, 2003) and this result justifies the selection of students to the universities based on the three criteria used at present. Hewapathirana (2003) argues that when the students are given equal opportunities as for those of the group with high G.C.E A/L aggregate, they perform equally well.

Gender

Overall, 59% of the study population was female. Gender was the second highest correlating factor with GPA. During the Degree program, it was negatively correlated to the GPA. It means that female students had performed at the exams better than male students. Two sample t-test analysis indicate that academic performance of male and female were significantly different (p<.0001). Mean FGPA of male students was 2.86 (SD = 0.37) while mean FGPA of female students was 3.05 (SD = 0.32). The reason for this difference is that time spent on academic activities by the female students might be comparatively higher than by the male students. This result goes parallel with findings of Ranjani et al. (2013 and 2007), Ediriweera and Weerakkody (2008), Sandika et al. (2012). Hewage et al. (2010) but does not follow the findings of Dilini (2007) and Fernando and Weerahandi (1979).

Ethnicity

About 98% of the study population was Sinhalese while 2% were of other ethnicities. Ethnicity was not significantly correlated with GPA. Irrespective of ethnicity all ethnic groups had performed equally.

Socio Economic Status

Out of the sample, “Mahapola” scholars, Bursary higher, Bursary lower were 56%, 1%, and 4% respectively. 39% of the students did not receive any scholarship. Socio economic status was based on “Mahapola” Scholarship and Bursary Scholarship data. Bursary and “Mahapola” Scholarship are based on parental income of students, number of brothers and sisters, and educational/employment status of brothers and sisters of the students. Hence it shows Socio-Economic status of the students. The research finding revealed that Socio-Economic status positively and weakly correlates with FGPA. The students having high Socio economic status perform better than the students having low Socio Economic status. Similarly, the finding of Nithlavarnan and Sinnathamby (2012) is compatible to this study. But, this finding is incompatible with the finding of Sandika et al. (2012) and the argument of that is family income had no significant effect on the performance of the students. However, this effect has been reducing over the time (Figure 4.5).

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.5: Correlation between FGPA and Socio Economic Status over the time

This may be because of new scholarship programs and/or psychological impact of the challenging future. The effect of Socio-Economic Status on academic performance can be expected no longer in accordance with the result of this study given in Figure 4.4.

Attempts

The most students had been selected for admission following the 2nd attempt at the GCE A/L examination: 8% from the 1st attempt, 48% from the 2nd attempt and 44% from the 3rd attempt. There was a significantly negative correlation of attempt at A/L with GPA (1st attempt had a lower code). Number of attempt of the students had adverse effect on the academic performance of the students.

According to the GLM analysis, the performance of the students in each attempt were significantly different (p=0.0011). The students who came in 1st and 2nd attempt at A/L exam had performed better than those who came in 3rd attempt. Similarly, a research finding of Hewage et al. (2010) goes parallel to this study. But Sandika et al. (2012) has showed the result in contrast to this finding.

Table 4.6: Comparison of number of Attempts

*Means with the same letters are not significantly different

Abbildung in dieser Leseprobe nicht enthalten

General Knowledge

Average General Knowledge Marks of the population was 65.16 at A/L examination. Correlation of those marks with FGPA shows that those marks do not impact on academic performance of students Following Crop Science modules.

A/L English

Among the students for whom A/L General English grades were available, 35% had failed the paper; 27% had a simple pass; 23% had a “C” grade; 10% had a “B” grade; and 5% had an “A” grade. Of the variables, A/L English had a much higher correlation coefficient with academic performance at the faculty Examination than the other variables. A/L English grades showed significant positive correlation with all Examinations in the Faculty. A/L English was the main predictor of academic performance of students. Even though the students have the required knowledge, if they cannot show it at exams, the outcome is failure. The same result has shown in the studies of Kottahachchi (1992), Gunawardene (1993) Sandika et al. (2012) and Karuanarathne et al. (2007 and 2013). But this is incompatible with findings of Hewapathirana (2002).

Agriculture/Physics at A/L

For A/L Bio-Science, students can select either agriculture or Physics as a subject. About 26% of the population was those who had followed Agriculture as an A/L subject while the rest of the students had followed Physics as an A/L subject. The result of t-test shows that the performance of the students was significantly different (p <.0001). According to the result of the study shows that the students who took physics as a subject performed better at the Faculty exams than who took Agriculture. Even though the students had followed agriculture in A/L, the performances at university exams are comparatively lower. Mean FGPA of Students who took Agriculture in A/L was 2.85 (SD = 0.34) and mean FGPA of Students who took Physics in A/L was 3.03 (SD = 0.34).

Vacancy Filling/ 1st selection

Most of students are recruited to the faculty in 1st selection of UGC while other students are recruited to the faculty for vacancy filling. 59% of the sample was those who have followed the degree as 1st selection. Other students were taken to fill vacancy. There was no significant difference between performance of students in these two groups irrespective of vacancy filling or 1st selection.

District

About 19% of the population had come from the Kurunegala district while from Killinochchi, Mannar, Mulative, Vavuniya, and Trincomalee districts students had not come to the Crop Science department of the faculty. The “Others” category includes Jaffna, Killinochchi, Mannar, Mulative, Vavuniya, Trincomalee, Batticaloa, Ampara, Puttalam, Polonnaruwa, Badulla and Monaragala (Figure4.6).

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.6: Student Distribution among districts

According to the UGC categorization there are two district categories namely educationally advantaged districts and educationally disadvantaged districts. Educationally advantaged districts include Colombo, Gampaha, Kaluthara, Mathale, Kandy, Galle, Matara, Kurunegala and Kegalle while educationally disadvantaged districts include the other 16 districts. 74.6% of the population had come from educationally advantaged districts and others were from educationally disadvantaged districts. t-test analysis was performed to analyze the means of FGPA of two district-category; educationally advantaged districts and educationally disadvantaged districts. There was a significant difference between these two groups (p=0.0021). The students coming from educationally advantaged districts performed better than those who coming from educationally disadvantaged districts. Mean FGPA of the students who come from educationally advantaged districts was 3.00 ± 0.35 and mean FGPA of the students who come from educationally disadvantaged districts was 2.89 ± 0.38. Two district categories show positive correlations with Z-Score, Socio-Economic status and English (p = 0.0007, p = 0.0020 and p = 0.0024 respectively). It indicates that the other factors also mainly depend on district and this disparity in the society also determine the performance of students following the Crop Science module in the Faculty of Agriculture.

Table 4.7: Correlation between district and FGPA

Abbildung in dieser Leseprobe nicht enthalten

Correlation between each district and FGPA was analyzed and the results in Table 4.7 show that the correlation of Colombo and Jaffna districts with FGPA were significantly positive while Nuwara-Eliya district showed significantly negative correlation with FGPA. This result supports the findings of Ranasinghe et al. (2012) but not the finding of Dilani (2007). Positive significant Correlation exist between English results with Colombo district (p=0.0004) and this ultimately lead to higher performance of the students who come from Colombo district. Significantly negative correlation is present between Z-Score and Nuwara-Eliya District (p <.0001) and it points to the low performance of the students coming from Nuwara-Eliya district.

Multivariate Analysis

General Linear Model

Analysis of General Linear Model was done and the result shows that English (p<0.0004) (p=0.0098) Gender (p=0.0167), Z-Score and were the best predictors of the academic performances of the students. Because of large number of missing values, other factors did not show significance. English is the most important factor that affects academic performance of students following the Crop Science modules. R[2] of the model is 0.222.

Principle Component Analysis

Principle Component Analysis was performed to group the variables and three Principle Components were retained with Eigen value over 1.0 (Table 4.8). Cumulatively, 63.2% of the variance was explained by first three principle components.

Table 4.8: Principle Component Analysis of the Variables

Abbildung in dieser Leseprobe nicht enthalten

Eigenvectors of Principle Component Analysis of Variables are given in table 4.9.

Table 4.9: Eigenvectors of Principle Component Analysis of Variables

Abbildung in dieser Leseprobe nicht enthalten

Principle factor analysis was performed to identify the common factors which are related to the academic performance of the students. Results of principle factor analysis of variables are given in table 4.10.

Table 4.10: Principle factor analysis of variables

Abbildung in dieser Leseprobe nicht enthalten

Principle factor analysis was performed with varimax rotation option to identify the variables belong to each factor clearly. Results of Principle factor analysis with varimax rotation option are given in table 4.11.

Table 4.11: Principle factor analysis with varimax rotation option

Abbildung in dieser Leseprobe nicht enthalten

Gender, Socio-Economic Status and English belong to Factor one and it can be conclude that this factor represent social status of the students. Factor two contain only Z-Score variable. Hence it can be concluded that factor two represent cognitive capacities of the students. Both Number of attempt and Agri/Physics variables belong to factor three and it can be identify as the preference of the students to the Agriculture subject is the last factor.

Subjects Classification

Advanced Programme subjects of Advanced Crop Production Technology and Plantation Management and Forestry modules were categorized using Cluster analysis.

Advanced Crop Production Technology

Out of the 12 subjects in Advanced Crop Production Technology five clusters were retained with their interaction (Figure 4.7). Cluster A contains Fruit and Vegetable Production (C3207), Organic Crop Production Systems (CS3210), Scientific Research and Communication in Crop Science (CS4104). Cluster B contains Protected Culture (CS4109) and Research project (CS4200) subjects and it show the highest performance and lowest variance. The students obtained the highest result in the Research project.

Cluster C includes Floriculture and landscape Horticulture (CS4110), Rice Production Technology (CS4111), Advanced Field Crop Production (CS4112). Performance in this group is moderate compared to the other four groups. Cluster D contains Design and Analysis of Experiments (CS3201) and it is clearly separate from other clusters. This subject requires mathematical and logical thinking ability to perform well. Cluster E includes Tissue culture (CS3206), Crop Physiology (CS3208), Statistical Methods II (CS4103). This group requires cognitive and logical ability more than other groups.

Performances of the students for subjects of Cluster A and B were somewhat equal and high compared to the subjects of Cluster C, D and E. The variance of subjects of Cluster A and B are low compared to the Cluster C, D and E. Variance may depend on the cognitive ability of the students. Hence higher GPA students had performed for the subjects of the Cluster C, D and E better than those who were with low GPA.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4.7: Dendrogram of the subjects of Advanced Crop Production Technology

Plantation Management and Forestry

The subjects of Plantation Management and forestry were also categorized using Cluster analysis (Figure 4.8). Cluster A contains Management of Rubber, Coconut & Export Agricultural Crops (CS3202), Tea Plantation management (CS4101), Forest Management (CS4102) and Tree Diversity and Improvement (CS3204). Cluster B contains Design and Analysis of Experiments (CS3201), Scientific Research and Communication in Crop Science (CS4104), and Agroforestry Systems (CS3203). Cluster C includes Statistical Methods ii (CS4103), Research project (CS4200) subjects.

The subjects in Cluster A require practical knowledge and its application. Because, these four subjects include field visits and practical. Therefore, students have to synthesize and apply those knowledge and practical experience at the exam. But, subjects in Cluster B require cognitive and logical thinking abilities to perform well. Cluster C includes two subjects having the lowest and highest average GPA.

Figure 4.8: Dendrogram of the subjects of Plantation Management and Forestry

CONCLUSION

The performance (measured by GPA) was different across departments. The performance of students of the Crop Science department was lower than other departments. However, this is mainly due to large number of students in the Crop Science modules. Due to this the average performance gets pulled down. When only the top 15 students were considered the performance of students in the Crop Science modules was very high. A significant difference exists between the two Crop Science modules in terms of GPA and the better performance was observed in Advanced Crop Production Technology Students. A/L Z-Score, Gender, District, Socio-Economic status, number of A/L attempts, A/L English grade and Agriculture/Physics as an A/L subject were all significantly related to academic performance of the Crop Science students while Vacancy Filling or 1st Selection, Ethnicity, General Knowledge at A/L do not affect the performance significantly. Knowledge of English was the highest related variable to performance followed by Z-score and Gender. Basically, social status of the students, cognitive capacities of the students and the preference of the students to the Agriculture subject are the common factors relating to the performance of the students in Crop Science Majoring module.

REFERENCES

- Anonymous (2006), prospectus 2006-2010. Faculty of agriculture, University of Peradeniya, Peradeniya, Sri Lanka.

- Chatfield, C. and Collins, A.J. (1980). Introduction to Multivariate Analysis. Chapman and Hall, London, UK.

- Dilani, D.G.A. (2007). Relationship between Academic Performance and Student Selection Criteria. (In) Proceedings of' the Annual Research Symposium Faculty of' Graduate Studies, University of Kelaniya. pp 152.

- Ediriweera, N.A., Weerakkody, W.A.S. (2008). Influence of Gender on Academic Performance: A Comparative Study between Management and Commerce Undergraduates in the University of Kelaniya, Sri Lanka. (In) Proceedings of the Annual Research Symposium, Faculty of Graduate Studies, University of Kelaniya. pp 201

- Fernando, G.Y.L., Weerahandi, S. (1979). A statistical analysis of Admission and Performance of Science students at Vidyodaya University. Vidyodaya J. Arts. Sci., Lett., 7;111-122

- Fidell, L.S. and Tabachnick, B.G. (1983). Using Multivariate Statistics. Harper & Row, New York, USA.

- Goonathilak, M.D.R.P., Gunarathne, N.P., Hapuarachchi, C.T., Jayawardena, A. (2008). Factors affecting performance at the IBSS examination. Student Medical Journal 1: 11-16

- Gunawardhana, R. 1993. Academic performance of, university students admitted under different categories. (In) Proceedings of the Sri Lanka Association for the Advancement of Science, Part 1(Abstract), 49th Annual Session, Colombo, Sri Lanka. pp. 257.

- Hewage, S.N., Salgado, L.S.S., Fernando, G.M.O., Liyanage, P.L.C.K., Pathmeswaran, A., de Silva, N.R. (2011). Selection of medical students in Sri Lanka: time to re-think criteria. Ceylon Medical Journal 56: 22-28

- Johnson, R.A. and Wichern, D.W. (2007). Applied Multivariate Statistical Analysis. Pearson Education, Inc., New Jersey, USA.

- Kottahachchi, D. 1992. Student (inter-linguistic) mobility. Proceedings of the Sri Lanka Association for the Advancement of Science, Part I (Abstract), 48th Annual Session, Colombo, Sri Lanka. pp 137.

- Nithlavarnan, N. and Sinnathamby, M. (2012). Effects of Socioeconomic Status of Parents on Educational Attainment of the Undergraduates of the University of Jaffna: A Comparative Study on Arts and Law Students. (In) Proceedings of the Abstracts of Jaffna University International Research Conference, PP 74 University of Jaffna, Sri Lanka.

- Ranasinghe, P., Ellawela A., Gunatilake, S.B. (2012). Non-cognitive characteristics predicting academic success among medical students in Sri Lanka. [online] Available at: < http://europepmc.org/articles/PMC3547768> [Accessed 12 October 2013]

- Ranjani R.P.C., Karunarathne W.V.A.D., and Weligamage. S. (2007). Determinants of accounting undergraduates academic performance. 11th International Conference on Sri Lankan Studies, 2nd-4th November 2007, University of Portsmouth, UK

- Ranjani R.P.C., Karunarathne W.V.A.D., and Weligamage. S. (2013). Determinants of Management undergraduates’ academic performance in Sri Lanka. The Association of Southeast Asian Institutions of Higher Learning Conference, 30th April - 3rd May, 2013, Surabaya, Indonesia.

- Sandika, A.L., Atapattu, N.S.B.M., and Weerasinghe, W.M.C.B. (2012). Factors Affecting the Academic Performance of Agriculture Undergraduates: a Case in Faculty of Agriculture, University of Ruhuna. SAARC Journal Of Educational Research 9;1-13

- Thattil R.O. and Nalaka G.D.A. (2010). Development of an empirical subject classification system. (In) Proceedings of the International Conference on Statistical Concepts and Methods for the Modern World. Peiris et al., (Eds). pp. 100. Applied Statistical Association of Sri Lanka.

- The National Education Commission of Sri Lanka. (2009). National policy framework on higher education and technical and vocational education. Colombo, Sri Lanka: author