SELECTED CORRELATES OF EXAMINATION ANXIETY AND ACADEMIC PERFORMANCE OF STUDENTS IN PUBLIC SECONDARY SCHOOLS IN KHWISERO SUB- COUNTY, KAKAMEGA COUNTY, KENYA BY MUKOLWE ASAKHULU NEWTON E83/13043/2009 DEPARTMENT OF EDUCATIONAL PSYCHOLOGY A THESIS SUBMITTED IN PARTIAL FULFILMENT FOR THE REQUIREMENTS FOR THE AWARD OF THE DEGREE OF DOCTOR OF PHILOSOPHY IN EDUCATIONAL PSYCHOLOGY IN THE SCHOOL OF EDUCATION OF KENYATTA UNIVERSITY OCTOBER, 2015 ii DECLARATION I confirm that this thesis is my original work and has not been presented in any other university. The thesis has been complemented by referenced works duly acknowledged. Where text, data, graphics, pictures or tables have been borrowed from other works, including the internet, the sources are specifically accredited through referencing in accordance with anti-plagiarism regulations. __________________________ Date: ___________________ Mukolwe, Asakhulu Newton E83/13043/2009 We confirm that the work reported in this thesis was carried out by the candidate under our supervision as University supervisors. _________________________ Date: ___________________ Prof. Edward Oyugi Department of Educational Psychology Kenyatta University __________________________ Date: ___________________ Dr. Sammy Tumuti Department of Educational Psychology Kenyatta University iii DEDICATION I dedicate this thesis to my beloved parents, Mr. Wilfred Joash Asakhulu Mukolwe and Mrs. Beatrice Maikuba Andayi Asakhulu. Their patience, support, encouragement, prayers and intrinsic motive to educate me resulted in this accomplishment. iv ACKNOWLEGDEMENT My successful accomplishment of this work resulted from God’s blessings, health and His wonderful gift of life. I am also grateful for the assistance, encouragement, inspiration and moral and professional support I received from my supervisors; Prof. E. Oyugi and Dr. S. Tumuti. To my beloved wife Ms. Nancy Matendechere Imbusi, a teacher at Ngenia Boys High School, Limuru, and a PhD student in Mathematics at Kenyatta University, and my lovely children Valerie Shitandi, Mablyne Omulanya, Ainsley Asakhulu, Brayden Imbusi and Hansel Asakhulu, I recognise their love, support and joy; and to my parents and siblings for family support. I also thank schools’ administrations and students, including all other people who were involved in facilitating my research work. I am also grateful to Ms. Praxides O. Nakhungu of Maasai Mara University, and Ms. Enyslim N. Wekesa of Gertrude Children’s Hospital – Muthaiga, Nairobi. I also recognise support from lecturers, fellow students, and friends from Kenyatta University, particularly, Prof. F. Okatcha, Dr. P. Ndambuki, Dr. T. Kinai, Dr. J. Dinga, Dr. E. Kigen, Dr. C. Wawire, Mr. J. Lunalo, Ms. M. Mung’ala, Dr. S. Mutueleli, E. Mwenda, E. Mwaniki, A. Ireri, G. Soi and R. Marangu; and Mzee Peter Omondi, Uncle D. Lilondo, Mr. W. Mutiso, Mr. C. Lumbasi, Ms. H. Chimoyi and Ms. D. Mutai; including Prof. H. Gatumu, University of Nairobi and Dr. J. Kwena, Pwani University. I wish you all God’s blessings. Finally, I am grateful to the Management and the Council of Kenyatta University for offering me a partial scholarship that enabled me to begin my PhD studies. v TABLE OF CONTENTS Title Page ……………………………………….………………………… (i) Declaration ……………………………………………………………….. (ii) Dedication …………………………………………………………..……. (iii) Acknowledgement ………………………………………………………... (iv) Table of Contents ……………………………………………………..….. (v) List of Tables ……………………………………………………………... (vii) List of Figures …………………………………………………………….. (viii) Abbreviations and Acronyms ………………………………………..……. (ix) Abstract ………………………………………………………………….... (x) CHAPTER ONE: INTRODUCTION AND CONTEXT OF THE STUDY ……………………………………………………………. ……. 1 1.1 Introduction ………………………………………………………….. 1 1.2 Background to the Study …………………………………………….. 1 1.3 Statement of the Problem ……………………………………………. 8 1.4 Purpose of the Study ………………………………………………… 10 1.5 Objectives of the Study ……………..................................................... 11 1.6 Research Hypotheses ………………………………………………… 11 1.7 Assumptions of the Study ……………………………………………. 12 1.8 Limitations of the Study ……………………………………………… 13 1.9 Delimitations of the Study …………………………………………… 13 1.10 Significance of the Study ……............................................................ 14 1.11 Theoretical Framework……………………………………………… 15 1.11.1 The Interference Model…………………………………………… 17 1.11.2 The Learning-Deficit Model………………………………………. 18 1.11.3 The Conservation of Resources (COR) Theory…………………… 20 1.12 Conceptual Framework ……………………………………………… 22 1.13 Operational Definition of Terms ……………………………………. 23 CHAPTER TWO: REVIEW OF RELATED LITERATURE ….......... 26 2.1 Introduction…………………………………………………………… 26 2.2 Examination Anxiety and Academic Performance …………………… 26 2.3 Academic Procrastination, Examination Anxiety and Academic Performance …………………………………………………………... 31 2.4 Locus of Control, Examination Anxiety and Academic Performance …………………………………………………………... 34 2.5 Academic Resilience, Examination Anxiety and Academic Performance …………………………………………………………... 37 2.6 Sex Differences in Examination Anxiety, Academic Procrastination, Locus of Control and Academic Resilience ………… 40 2.6.1 Sex Differences and Examination Anxiety ………………………. 40 2.6.2 Sex Differences and Academic Procrastination ………………….. 41 2.6.3 Sex Differences and Locus of Control …………………………… 42 2.6.4 Sex Differences and Academic Resilience ……………………….. 42 2.7 Summary and Gap Identification …………………………………….. 43 vi CHAPTER THREE: RESEARCH DESIGN AND METHODOLOGY ……………………………………………………… 45 3.1 Introduction………………………………………………………….. 45 3.2 Research Design and Locale ………………………………………… 45 3.3 Population …………………………………………………………… 46 3.4 Sampling Techniques and Sample Size Determination ……………... 47 3.5 Research Instruments ……………………………………………....... 50 3.6 Data Collection ………………………………………………………. 56 3.7 Data Analysis ………………………………………………………... 58 CHAPTER FOUR: FINDINGS, INTERPRETATIONS AND DISCUSSIONS ………………………………………………………….. 61 4.1 Introduction …………………………………………………………. 61 4.2 General and Demographic Information ……………………………... 62 4.2.1 General Information ……………………………………………… 62 4.2.2 Demographic Information ………………………………………... 63 4.3 Examination Anxiety and Students’ Performance in KCSE Exams ………………………………………………………………... 73 4.4 Academic Procrastination, Examination Anxiety and Students’ Performance in KCSE Exams …………………………….. 76 4.5 Locus of Control, Examination Anxiety and Students’ Performance in KCSE Exams ……………………………………….. 79 4.6 Academic Resilience, Examination Anxiety and Students’ Performance in KCSE Exams ……………………………………….. 81 4.7 Sex Differences in Examination Anxiety, Academic Procrastination, Locus of Control and Academic Resilience ………... 84 CHAPTER FIVE: SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ……………………………………………….. 89 5.1 Introduction …………………………………………………………. 89 5.2 Summary …………………………………………………………….. 89 5.3 Conclusions ………………………………………………………….. 90 5.4 Recommendations …………………………………………………… 92 5.4.1 Policy Recommendations ………………………………………… 92 5.4.2 Recommendations for Further Research …………………………. 94 References ……………………………………………………………….. 96 Appendices: Appendix A: Research Instruments …………………….. 115 Appendix B: Time Frame ………………………………. 125 Appendix C: Budget ……………………………………. 126 Appendix D: Research Authorization Documents ……… 127 Appendix E: Map of Khwisero Sub-County ……………. 131 Appendix F: Khwisero Sub-County Secondary Schools’ List .……...................................................... 132 Appendix I: Khwisero Sub-County Secondary Schools’ KCSE Performance Trends …………….…. 135 vii LIST OF TABLES Table 3.1 Schools’ Residential Status in Khwisero Sub-county …… 46 Table 3.2 Sample Distribution ……………………………………… 50 Table 3.3 Reliability Coefficients for each Sub-scale ………………. 55 Table 4.1 Pearson Product Moment Correlation Test for the Relationship between Exams Anxiety Scores and KCSE Exams Scores …………………………………………….. 74 Table 4.2 Pearson Product Moment Correlation Test for the Relationship between Academic Procrastination, Exams Anxiety Scores and KCSE Exams Scores ……………….. 76 Table 4.3 Pearson Product Moment Correlation Test for the Relationship between Locus of Control, Exams Anxiety Scores and KCSE Exams Scores ………………………… 79 Table 4.4 Pearson Product Moment Correlation Test for the Relationship between Academic Resilience, Exams Anxiety Scores and KCSE Exams Scores ………………………… 82 Table 4.5 Independent Samples t-Test for Sex Differences in Examination Anxiety, Academic Procrastination, Locus of Control and Academic Resilience ……………………….. 85 viii LIST OF FIGURES Figure 1.1 Relationships between the Variables of the Study ……….. 22 Figure 4.1 Distribution of Students by Sex and Schools’ Residential Status …………………………………………………….. 64 Figure 4.2 Distribution of Students by Sex and Age ………………… 65 Figure 4.3 Distribution of Respondents by Sex, Age and KCSE grade Performance ……………………………………………… 66 Figure 4.4 Distribution of Respondents by Schools’ Residential status and KCSE Grade Performance …………………………… 67 Figure 4.5 Statistical Characteristics of Respondents by Sex and Exams Anxiety, Academic Procrastination, LOC, Academic Resilience and KCSE Points ………………….. 68 Figure 4.6 Distribution of Respondents by Age and Means of Exams Anxiety, Academic Procrastination, LOC, Academic Resilience and KCSE Points ……………………………… 70 Figure 4.7 Distribution of Respondents by Schools’ Residential Status and Means of Exams Anxiety, Academic Procrastination, LOC, Academic Resilience and KCSE Points …………… 72 ix ABBREVIATIONS AND ACRONYMS COR Conservation of Resources CPE Certificate of Primary Education CTAS Cognitive Test Anxiety Scale DEO District Education Office EAACE East African Advanced Certificate of Education EACE East African Certificate of Education EACPE East African Certificate of Primary Education GAS General Adaptation Syndrome GPA Grade Point Average KACE Kenya Advanced Certificate of Education KCE Kenya Certificate of Education KCPE Kenya Certificate of Primary Education KCSE Kenya Certificate of Secondary Education KICD Kenya Institute of Curriculum Development KNEC Kenya National Examination Council LOC Locus of Control MDGs Millennium Development Goals NCST National Council for Science and Technology SPSS Statistical Package for Social Sciences TAI Test Anxiety Inventory x ABSTRACT The purpose of this study was to analyze the relationship between exams anxiety, academic performance and other selected correlates, which were, academic procrastination, Locus of control and academic resilience. Indeed, the phenomenon of exams anxiety has been widely studied in relation to academic performance. However, exams anxiety, as a psychological construct, also needs to be understood in terms of other related factors that might precipitate it. In addition, the study also investigated sex differences in relation to the above variables. The study was conducted on form four students in public secondary schools who attempted the 2012 KCSE exams in Khwisero Sub-county, Kakamega County, Kenya. Exams anxiety and such related correlates are among factors that influence the levels of academic performance in students in Khwisero Sub-county, which are relatively low as compared to other Sub-counties in Kakamega County. The study was guided by the Conservation of Resources (COR) theory, which holds that one needs resources to cope with and overcome stressful situations, such as exams related anxieties in students. The interplay of the correlates mentioned above could constitute such resources. Purposive sampling, stratified random sampling and simple random sampling designs were used accordingly to select 359 form four students. These sampling procedures yielded 203 boys and 156 girls for the study. A questionnaire containing sub- scales on test anxiety, academic procrastination, Locus of control and academic resilience was used for data collection. This questionnaire was initially piloted to assert its validity and reliability. The study adopted a correlational research design. The hypotheses on the relationship between exams anxiety, academic performance and the selected correlates were tested using Pearson Product Moment Correlation. On the other hand, the hypothesis on sex differences in these variables was tested by Independent Samples t-Test. All the tests were carried out at 0.05 level of significance. The findings indicated that majority of students experienced exams anxiety; girls were high on academic resilience and were more of internalizers, while boys were high in academic procrastination and were more of externalizers. Further, the findings showed that exams anxiety was significant and negatively related to academic performance; academic procrastination positively and significantly correlated with exams anxiety, but negatively and insignificantly correlated with academic performance; Locus of control negatively and insignificantly correlated with exams anxiety, but positively and insignificantly correlated with academic performance; and academic resilience negatively and insignificantly correlated with exams anxiety, but positively and insignificantly correlated with academic performance. The study concluded that exams anxiety was a real phenomenon that affected students’ academic performance in exams, and that academic procrastination, locus of control and academic resilience are important factors in relation to exams anxiety and academic performance. The study recommended that the issue of exams anxiety should be involved in guidance and counseling policy strategies; and that further research should also consider other exams anxiety correlates such as testwiseness, subject choice, teaching pedagogy and intelligence. 1 CHAPTER ONE INTRODUCTION AND CONTEXT OF THE STUDY 1.1 Introduction This chapter presents among other items, the background to the problem investigated, the purpose, the objectives and the significance of this study. Theoretical and conceptual frameworks, as the main guiding principles to the study, shall also be presented. Lastly, operational definitions of the key terms in the study will be highlighted. 1.2 Background to the Study The Kenya vision 2030 recognizes education and training, within the social pillar, as one of the platform that will transform Kenya into a newly industrialized middle income country by the year 2030 (Republic of Kenya, 2007). The education goals of vision 2030 emphasize on the provision of globally competitive quality education, training and research for development. Alongside the Kenya vision 2030, are the Millennium Development Goals (MDGs) whose main aim is to encourage development by improving social and economic conditions in the world’s poorest countries by 21. The MDGs focuses on three areas of human development; these are, bolstering human capital, improving infrastructure, and increasing social, economic and political rights. Education is the key within the human capital. 2 In Education, learners’ educational outcomes and achievements are evaluated and graded using examinations (Chinta, 2005). In fact, testing is common in everyday life, from school content-specific tests (that is, class tests and national examinations) to tests taken to move up in jobs status, thus adding a great deal of pressure to test achievement and grades. Hence, in most cases, this leads many people to become anxious when presented with examinations (Huberty, 2010; Supon, 2004; Collins, 1999). This form of anxiety is known as test or examination anxiety. Spielberger and Sarason (1989) define test anxiety as a situation-specific trait that refers to the anxiety states and worry conditions experienced during examinations. The level of anxiety can fluctuate over time in response to both internal and external stimulation. Observable behaviors of anxiety can be noticed during the completion process of a quiz, a test or an examination. Some of those behaviors might include perspiration, excessive movement and questioning of instructions. A disruption or disorganization of effective problem-solving and cognitive control, including difficulty in thinking clearly, can also lead to test anxiety (Freidman & Bendas -Jacob, 1997). These anxiety related behaviors during testing conditions may in one way or another affect academic performance of students. Indeed, a minimal amount of anxiety is needed to mobilize human beings to respond rapidly and efficiently; but when in excess, it may foster poor response and even inhibit response (Simpson, 3 Parker & Harrison, 1995). Hence, too much anxiety during examination may interfere with students’ concentration on the test, thus lowering their performance in examination (Cassady & Johnson, 2002) . The history of examination in educational set ups in Kenya has existed in both pre-colonial and post-colonial Kenya, and this has seen learners being occasionally exposed to examination. Indeed, after independence in 1963, Kenya, together with Uganda and Tanzania, formed the East African Community in 1967. The three countries adopted a single system of education (similar to the British system of education), the 7-4-2-3. This system consisted of seven years of primary education, four years of secondary education, two years of high school, and three years of university education, hence 7-4-2-3. To make transitions in this system, learners sat for regional examinations known as the East African Certificate of Primary Education (EACPE), and then the East African Certificate of Education examination (EACE). Finally, they sat for the East African Advanced Certificate of Education (EAACE), after which they joined university education. With the collapse of the East African Community in 1977, Kenya continued with the same system of education but changed the examination names from their regional identity to the national identity. The East African Certificate of Primary Education became the Certificate of Primary Education (CPE), the East African Certificate of Education became the Kenya Certificate of Education (KCE) and 4 the East African Advanced Certificate of Education became the Kenya Advanced Certificate of Education (KACE). In 1980, the Kenya National Examination Council (KNEC), was established by the Government through an Act of Parliament (CAP 225A) as a non–profit making institution. Its responsibilities include conducting school and post-school national examinations, except university examinations, and award certificates to successful candidates. KNEC examinations are based on the syllabus prepared by the Kenya Institute of Curriculum Development (KICD – formally known as Kenya Institute of Education, KIE) for all public and private schools and colleges preparing for the KNEC examinations. In January 1985, the 8-4-4 system of education was introduced following the Mackay Report of 1982 (Republic of Kenya, 1988)., which adopted eight years of primary education, four years of secondary education and four years of university education, hence 8-4-4. With the introduction of the 8-4-4 system, CPE became KCPE (Kenya Certificate of Primary Education) while KCE became the Kenya Certificate of Secondary Education (KCSE). The main limitation of all the aforementioned regional and national examinations is that the scores obtained from these examinations have been regarded as the sole objective means used to admit students into secondary schools, colleges and university courses, and job placements (Wambua, 2003). With such significance accrued on the exam scores, students are bound to develop test anxiety symptoms as they approach examination periods (as noted by 5 Huberty, 2010; Supon, 2004; Collins, 1999). This is well exemplified by a host of events associated with the preparation and the release of results for final examinations in many schools around the country. These include; holiday tuitions (Kimani, Standard Newspaper, August 7, 2011), class repetition for learners who did not perform well, highly publicized prayers for the candidates (Owayo, Standard Newspaper, October 15, 2011), strikes associated with examination issues (Imprint, 2008), examination leakages and cheating cases, and examination results cancellation, like in the case of North Eastern Kenya schools in 2011 KCSE results (KNEC, 2012). The list also includes suicidal cases related to poor performance in national examinations. Majority of research work done on the effects of test anxiety on students’ academic performance mainly indicates a negative relationship between test anxiety and performance (Huberty, 2010; Stowell & Bennett, 2010; Flannery, 2008; Parks-Stamm, Gollwitzer & Oettingen , 2008; Jing, 2007; Chapell, Blanding, Silverstein, Takahashi, Newman, Gubi and McCann, 2005; Cassady, 2004; Cheek, 2002; McDonald, 2001; Ondondi, 1983). The above research findings imply that increase in test anxiety lowers test performance of students. However, the findings in studies carried out by Ndirangu, Muola, Kithuka and Nassiuma (2009) and Vogel and Collins (2008) found no relationship between test anxiety and academic performance of students. 6 The above conflicting research findings call for more empirical studies to validate the relationship between examination anxiety and academic achievement of students, especially in Kenya, where studies reviewed (that is, Ndirangu, et al. 2009 and Ondondi, 1983) indicate contrasting results. In addition, many studies carried out on the relationship between test anxiety and performance, have rarely addressed different factors that are associated with the development of test anxiety hence the effects on academic performance in learners. In the current study, the following factors (herein referred to as correlates) were considered: academic procrastination, Locus of control and academic resilience. Academic procrastination involves the habits of studying lessons in the late night hours and at the last minute (Hess, Sherman & Goodman, 2000 ). Sirois and Pychyl (2002) study revealed that students who procrastinate had high stress, poor coping strategies and also avoidance behaviors. Those students who suffered from such avoidance coping styles resisted completing assignments and addressing other deadlines that eventually evoked tension and anxiety. Thus, such students are bound to experience high examination anxiety and low academic performance. Locus of control is another correlate of examination anxiety and academic performance. Locus of control (or location of control) refers to people's very general, cross-situational beliefs about what determines whether or not they get reinforced in life (Rotter, 1989). 7 The determinants may be internal, hence internal Locus of control, or external, hence external Locus of control. Thus, internals believe that success or failure is due to their own efforts, whereas externals believe that the reinforcers in life are controlled by luck, chance, or powerful others (Rotter, 1989). Consequently, internals considers success in an exam as a result of their hard work (for example, good study habits), while, externals will consider failure on an exam to be the result of an unfair test. In Carden, Courtney and Rebekah (2004) study, it was found that internals showed significantly lower academic procrastination, low test anxiety and reported higher academic achievement than externals. These findings indicate the importance of Locus of control in the relationship between examination anxiety and academic performance in learners. Another important correlate of examination anxiety and academic performance considered in this study was academic resilience. Resilience usually refers to a measure of stress coping ability (Masten, 2001). It involves personal competence, trust in one’s instincts, positive acceptance of change, control and spiritual influences (Connor & Davidson, 2003). Research evidence has indicated that personal resources like mastery, adaptability and emotional stability combine to produce resilience that would buffer negative effects of test anxiety on academic performance (Prince-Embury, 2005). Hence, learners with high academic resilience are expected to have low test anxiety and attain high standards of academic achievement. 8 The relationship between the above correlates with examination anxiety and academic performance could be among the factors that might influence the levels of academic performance in Khwisero Sub-County. Because Khwisero Sub- County was inaugurated in 2009 (Republic of Kenya, 2009), as an independent Sub-County in Kakamega County, its academic standards can now be clearly seen. The trend since its inception shows that the performance of students in national examinations has been quite low (Kabaka, 2011). Thus, a study that looks into the factors associated with exam performance in the Sub-County was vital at this point in time. Therefore, the present study was designed to investigate the above factors as they relate to examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County, Kakamega County, Kenya. 1.3 Statement of the Problem In the foregoing discussion, empirical evidence has given mixed findings on the relationship between exam anxiety and academic performance in students. Nevertheless, high expectations set by both parents and teachers over improvements in the mean scores in national examinations (Choti, 2009), contributes to increasing pressure, hence anxiety in students. Such anxiety episodes in students are normally high prior to taking major examinations and even afterwards as they anticipate for exam results to be announced. 9 The performance of students in national examinations in Khwisero Sub-County has been comparatively low in relation to other Sub-Counties in the larger Western region. For example, in the 2010 KCSE examinations the Sub-County was ranked number 22 among the 28 Sub-Counties in the Western region (Kabaka, 2011). The examination results data from the Sub-County Education Office for the years 2009, 2010 and 2011 (See Appendix VII), exemplifies this by showing that very few students acquire the minimum university entrance grade (C+). Further, of these, very few join and complete university education due to high poverty levels in the Sub-County as the local economy is mostly driven by subsistence agriculture (Kabaka, 2011). Indeed, with such poverty levels, the teaching-learning resources in many public schools are scarce, leading to inadequate curriculum coverage (Ingubu & Wambua, 2011; UNICEF, 2005; KIPPRA, 2001), and less examination preparation among students, hence this might heighten test anxiety in students. It is out of such state of affairs that the researcher, apart from investigating the relationship between exam anxiety and academic performance, was also interested in studying the influence of the aforementioned correlates on exam anxiety hence academic performance of students in Khwisero Sub-County. In essence, as review of research literature has shown in the background section of this study, these correlates when positively utilized should have a buffering effect against exam anxiety and a boosting effect on academic performance of 10 students. Hence, the earlier the students in Khwisero Sub-County prepare for exam by avoiding academic procrastination, the better for them to reduce exam anxiety and increase their levels of academic performance. Similarly, the more they are directed by internal locus of control and the more they become academically resilient, the less they experience exam anxiety and the higher they increase their mean performance scores in exam. Thus, it was important to investigate the relationship between exam anxiety and academic performance of students in Khwisero Sub-County by also considering the relationship between exam anxiety and academic performance, and the above correlates (that is, academic procrastination, Locus of control and academic resilience). In addition, due to scarcity of research literature on these variables locally, it was important to investigate the relationship between these variables to add more research literature on the hypothesis relating to the widely publicized influence of exam anxiety on academic performance of students. 1.4 Purpose of the Study The purpose of this study was to find out the relationship in the selected correlates of examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County. In addition, the study attempted to identify sex differences in relation to the selected correlates of examination anxiety and academic performance of students in public secondary schools in Khwisero Sub- County. 11 1.5 Objectives of the S tudy The study specifically aimed at finding out the; i) degree of relationship between examination anxiety and academic performance of students in public secondary schools in Khwisero Sub- county ii) relationship between academic procrastination, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-county iii) relationship between Locus of control, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub- county iv) relationship between academic resilience, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-county v) sex differences among students in public secondary schools in Khwisero Sub-county in relation to their levels of examination anxiety, academic procrastination, Locus of control and academic resilience. 1.6 Research Hypotheses The following research hypotheses were tested in this study: Ha1 There is a negative significant relationship between examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County 12 Ha2 There are positive significant relationships between academic procrastination, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County Ha3 There are positive significant relationships between Locus of control, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County Ha4 There are positive significant relationships between academic resilience, examination anxiety and academic performance of students in public secondary schools in Khwisero Sub-County Ho5 There are significant sex differences among students in public secondary schools in Khwisero Sub-County in relation to examination anxiety, academic procrastination, Locus of control and academic resilience. 1.7 Assumptions of the Study This study was guided by the following assumptions: i) The scales used for data collection yielded valid and reliable information for testing of the hypotheses under the study ii) That in any case where exam irregularities occurred undetected, they had insignificant effect in the analysis of data and on the overall results and findings of this study 13 1.8 Limitations of the Study Standardized scales to measure the variables of the study are not locally available. As such, the researcher adapted items from standardized scales previously used in other research jurisdiction. Items to measure exam anxiety were adapted from Cassady and Johnson (2004) and Sarason (1984); the items used to measure academic procrastination were drawn from Steel (2010), Diaz-Morales, Ferari, Diaz and Argumedo (2006), Tuckman (1991) and Lay (1986). Rotter’s (1) scale was adapted and used to measure Locus of control. Similarly, Conner-Davidson (2003) and Wagnild and Young (1993, 1987) studies provided items that were used to measure academic resilience. The items adapted from the scales used in these previous studies were subjected to a pilot study to ascertain their validity and reliability. Lastly, the generalization of the findings from this study can only be done to the extent that students in other regions are similar in characteristics and conditions in relation to those of Khwisero Sub-County. 1.9 Delimitations of the Study For the purpose of this study, form four students in public secondary schools who had registered for the 2012 Kenya Certificate of Secondary Examinations (KCSE) in Khwisero Sub-County, Kakamega County, were involved in the study. Form four students were chosen because they were going to sit for the final national 14 examinations, hence an appropriate sample to demonstrate the influence of exam anxiety on academic performance. Furthermore, since these national examinations are standardized, they yielded the most valid and reliable data for the analysis of hypotheses in the study. The study concentrated on exam related anxiety as measured by the examination anxiety scale whose validity and reliability were ascertained. In addition, the study confined itself to the following variables as related to examination anxiety; academic procrastination, Locus of control and academic resilience. 1.10 Significance of the Study The findings of this study may have significant implications to students, teachers, parents, counselors and other stakeholders in the Education sector in Kenya. These include appreciating examination anxiety as a real phenomenon, its potential effects on academic performance, factors that may precipitate it, management and strategic programs in dealing with it. The findings of this study may also contribute to research literature for local, international and cross-cultural comparisons among scholars, researchers and policy makers in Education and other Social Sciences. Within the same framework, it is the hope of the researcher that by contributing to research literature, this would stimulate even further research work, both locally and 15 internationally, in appreciation of examination anxiety as a universal phenomenon that cuts across time and space. 1.11 Theoretical Framework This section highlights some of the prominent theories and or models related to anxiety and or stress. The discussion assumes an evolutionary approach to these theories beginning with earliest models to the more recent model (the COR theory) that was finally adopted by the study. In the earliest theoretical framework, the effects of test anxiety on performance variables, such as competition, were depicted as an inverted U curve by Yerkes and Dodson (1908), precisely known as the Yerkes-Dodson curve. The description of the curve indicates that up to the summit of the curve (the optimal level), anxiety could be adaptive because it motivates students, and improves their functioning. Beyond the optimal level, anxiety is considered maladaptive as it causes distress and impairs functioning, hence productivity (Yerkes & Dodson, 1908). Indeed, this implies that some amount of anxiety is necessary to spark off functioning in human beings, however increasingly excessive anxiety leads to poor performance in human activities. Early works conceived test anxiety as a singular attribute that could be measured with unidimensional scales (Sarason, 1961). However, even with such early studies employing unidimensional measures of test anxiety, there was a 16 perception that there were at least two manifestations of test anxiety. Sarason (1961) proposed that interfering anxieties encountered in evaluative situations were likely a combination of heightened physiological activity and self- deprecating ruminations. Thus, since the early 1970s, there has been wide acceptance of the view that test anxiety is composed of two dimensions (traditionally referred to as emotionality and worry). Emotionality referred to autonomic reactions which tend to occur under examination stress, while worry was conceptually identified as cognitive expression of concern about one's own performance (Liebert & Morris, 1967). These two factors were further independently expounded theoretically by Selye (1976) and Lazarus (1966), respectively. In Selye’s (17) General Adaptation Syndrome (GAS) theory, the response pattern to stress proceeds in three stages, the alarm reaction, resistance, and exhaustion stages. In the alarm reaction stage, the body’s physiological mechanisms prepare to handle the shock through autonomic excitation and increased adrenalin discharge. In the resistance stage, the organism indicates some adaptation to the stressor. As aversive stimulation persists, resistance gives way to the stage of exhaustion leading to no response or death. This is purely a physiological explanation to stress, and exam anxiety can cause panic attacks in students, thus following the same physiological reaction pattern. 17 On the other hand, Lazarus (1966) offered a Cognitive Appraisal Theory that focused on an individual’s expectations of a specific encounter, leading to the perception of a given encounter as harm, threat or a challenge. Perceiving an event as harm or threat would incapacitate performance, while perceiving it as a challenge made an individual to feel confident in resolving the danger posed by the encounter. Since much significance is accrued on exam scores in making professional and life decisions, students are bound to perceive exam as a harm or a threat and or a challenge, hence incapacitating their performance in exam, and or developing confidence in overcoming the dangers posed by exam anxiety, respectively. Subsequently, most research in test anxiety was directed towards creating better measures of test anxiety (Spielberger, Gonzalez, Taylor, Algase & Anton, 1978; Liebert & Morris, 1967) and eventually determining the differential impacts of the two factors (Hembree, 1988). It is out of such early studies that two models were developed to explain the negative effects of test anxiety on academic performance; these are; the Interference Model (Sarason, 1986) and the Learning- Deficit Model (Naveh-Benjamin, McKeachie, & Lin, 1987). 1.11.1 The Interference Model The Interference Model seemingly developed from the early ideas of Attention Theory by Wine (1971), which stated that anxious persons divide their attention between task relevant activities and preoccupations with worry, self-critism and 18 somatic concerns, thus depressing their performance. Nevertheless, the Interference Model also posits that anxious persons are distracted due to worrying and task-irrelevant cognitions (Sarason, 1986). This model was quite tenable hence vast amount of research was done on it (Cassady, 2001; Ferando, Verea & Lorenzo, 1999; Bandalos, Yates, & Thorndike-Christ, 1995; Williams, 1991; Hembree, 1988; Sarason, 1978). However, some researchers (for example, Birenbaum & Nasser, 1994 ) emerged to claim that Interference Model by no means explained the only preponderance of criterion variables during the test-taking situation. Consequently other variables needed to be taken into consideration. These opponents proposed that low performance of highly test-anxious students would exhibit deficit knowledge of inadequate mastery, which had nothing to do with any form of interference; hence, the advent of the Learning-Deficit Model. 1.11.2 The Learning-Deficit Model The Learning-Deficit model proposes that the anxiety itself is a result of some inadequate preparation, for example, in study habits or in test-taking skills, thus, leading to anxiety and less performance on the test (Naveh-Benjamin, McKeachie, & Lin, 1987). However, the model is unsatisfactory since it suggests that it is the lack of skills which result in poorer examination results, rather than the anxiety, which is merely a side issue (Tobias, 1990). 19 Nevertheless, some researchers conceptualized the Interference and Learning- Deficit models as being mutually exclusive (Naveh-Benjamin, 1991; Benjamin, Mckeachine, Lin & Holinger, 1981). However, Tobias (1985) rejected the belief that the deficit and Interference Models were alternative explanations, but conceptualized them as being complementary. According to him optimal performance is achieved by those students who have good domain-specific skills and low test anxiety, because such students have the greatest proportion of their cognitive capacity available to cope with task demands. Such views were also supported by Zohar (1 998). Certainly, many factors impact on examination stress and performance. Chapell, Takahashi, Silverstein, Newman, McCann, Blanding and Gubi (2005) assert that any reasonable model of school achievement needs to consider a wide array of factors such as scholastic abilities, study habits, school attitudes, self-perceptions and self-efficacy, students’ health, classroom environment, opportunities for enrichment, among others. Hence, test anxiety could be considered as just one of many variables surrounding examination performance. It is out of such conceptualizations that Hobfoll (1998) developed the Conservation of Resources (COR) Theory in response to the need to incorporate more fully both the objective and perceived environmental factors into the process of coping with anxiety/stress. COR theory (Hobfoll, 1998) explains the impact of 20 assessment on emotions and performance by focusing on the resources of individuals and or groups (Buchwald, 2003). 1.11.3 The Conservation of Resources (COR) Theory According to COR Theory, the primary motivation in human beings is to build, protect and foster their resource pools for self and social bonds defenses (Hobfoll, 1998). The theory provides a model for preventing resource loss, maintaining existing resources, and gaining resources necessary for engaging in appropriate behavior. COR Theory argues that resources are the key components to determining individuals' appraisals of events as stressful , and they define how individuals are able to cope with stressful situations. COR Theory proposes that those individuals lacking resources are more vulnerable to experience loss spirals and those with plenty of resources have more opportunity for resource gain, hence they will be more resilient (Hobfoll, 1998). COR Theory can be applied to education and assessment by conceptualizing test anxiety as a loss of resources (loss of self-worth, motivation, and of productive cognitive processing) affecting students’ learning and performance. Within the framework of COR Theory, tests and exam can be understood as stressors (Buchwald, 2010). When individuals generally strive to obtain, retain, and protect what they value (that is, attaining high academic achievement and performance in the case of 21 students), testing situations increase the probability for a loss of these valued resources. Stress in testing situations can be so excessive that it hinders a person's ability to prepare properly and face testing situations, effectively. For students, the increasing pressure to perform well in tests, combined with cognitive interference, a lack of self-efficacy and motivation as well as negative life conditions (for example, inadequate resources and opportunities to learn), can lead to test anxiety and feelings of hopelessness. This, in turn, can make students unable to improve their performance, as test anxiety is associated with poor performance reducing self-worth and prospects for personal growth (Buchwald, 2010). Research work conducted on COR’s Theory has validated these assertions (Buchwald & Schwarzer, 2010; Black & William, 1998). With regard to the present study, the variables that constitute resources as alleged by the COR theory are less academic procrastination, internal locus of control and high academic resilience. These factors could buffer the negative effects of test anxiety thereby reducing its negative influence on academic performance. At the same time, they could have a positive influence on students’ academic achievement. The relationship between these factors (also called correlates) with test anxiety and academic performance, including sex differences across the same variables, is explained in the following conceptual framework. 22 1.12 Conceptual Framework In the foregoing description of the problem investigated in this study, and the theoretical guidelines to the study, the following conceptual framework, as shown in Figure 1.1 was developed by the researcher to aid in the conceived relationships between independent and dependent variables of the study. Key:  Independent Variable s: Examination Anxiety, Academic, Procrastination, Locus of control and Academic Resilience  Dependent Variable : Academic Achievement  Categorical Variable: Sex differences Figure 1.1: Relationships between the Variab les of the Study Source: Researcher, (2012)  Internal Locus of control  Less academic procrastination  High level of academic resilience  External Locus of control  High academic procrastination  Low level of academic resilience Exam ination Anxiety Lowers Increases High Low Sex differences Academic Performance 23 In Figure 1.1, academic procrastination, Locus of control and academic resilience are independent variables that amount to resources in the light of the COR Theory. It then follows that less academic procrastination, internal Locus of control and high level of resilience are resources which would lower the threat of examination anxiety (also and independent variable) and increase students’ academic performance as a dependent variable. Conversely, more academic procrastination, external Locus of control and low level of resilience would amount to lack of resources and eventually leading to increased threat of examination anxiety and low academic performance in students. Hence, the relationship between exam anxiety and students’ academic achievement would be negative. At the same time, there are sex differences (as a categorical variable) in these correlates, including examination anxiety and academic performance. All the above possible relationships and sex differences in the variables of the study were statistically tested. 1.13 Operational Definition of Terms For the purpose of this study the following terms were defined as follows: i) Academic performance – refers to the points or grades scored by the students in the KCSE exam. Other terms used interchangeably with academic performance were academic outcomes, learning outcomes, school achievement and academic achievement. ii) Academic procrastination – refers to the tendency to repetitively delay or postpone completion of school and academic tasks until the 24 last minute, such as homework, assignment, projects and exam revision. iii) Academic resilience – refers to high levels of achievement motivation and performance despite the presence of stressful events and conditions that place individuals at risk of doing poorly in school and ultimately dropping out of school. It may involve academic and exam stress, pressures and academic performance expectations, and poverty. iv) Correlates – refers to factors or things that have a mutual relationship or connection, in which one thing affects or depends on another. In this study, academic procrastination, Locus of control and academic resilience were considered as correlates of exam anxiety v) Examination anxiety – refers to states of worry and apprehension that occurs in students in the face of examinations or test conditions. The terms test/ exam anxiety and cognitive test anxiety were used interchangeably in this study. vi) Locus of control – refers to ones perception about the underlying main causes of events in his/her life. That is, a belief about whether the outcomes of ones actions are contingent on what one does (i.e. based on personal decisions and efforts – internal control orientation, or on events outside one’s personal control, i.e. fate, luck – external control orientation). vii) Secondary school – in Kenya, it refers to the second level of the formal education process that one joins after successfully completing 25 the basic or primary school education level. It begins from form one to form two, to form three and finally to form four. viii) Sex differences – refers to anatomical differences between males and females in terms of masculine and feminine, respectively. 26 CHAPTER TWO REVIEW OF RELATED LITERATURE 2.1 Introduction This chapter is divided into six sections. The first section provides a review of related literature on the relationship between examination anxiety and academic performance. The second, third and fourth sections provide a review of literature on academic procrastination, Locus of control and academic resilience in relation to examination anxiety and academic performance. This is followed by a review of related literature on sex differences in examination anxiety, academic procrastination, Locus of control and academic resilience. The last section presents summary of the review of related literature. 2.2 Examination Anxiety and Academic Performance Research on test anxiety has a long and fruitful history. First studies relating to test anxiety were conducted as early as 1914 (Folin, Demis & Smillie, 1914). Thereafter, exam anxiety received a considerable attention in the 1930s (for example; Brown, 1938a, 1938b; Neumann, 1933; Luria, 1932). Eventually, the concept began to be investigated under its own name in 1952, when Mandler and Sarason published a series of studies on test anxiety and how it relates to performance, as well as developed an instrument to assess individual differences in test anxiety in adults, the Test Anxiety Questionnaire (Sarason & Mandler, 1952). All through the early decades to 1990s, empirical studies on the 27 relationship between exam anxiety and exam performance found a negative relationship; that is, higher test anxiety positively correlated to lower exam performance (for example, Sarason, 1958a, 1965; Speilberger, 1966; Sarason, 1975; Humbree, 1988; Kurosawa & Harackiewicz, 1995; Zohar, 1998). Later studies also support this contention. In one such study by Rana and Mahmood (2010) on the relationship between test anxiety and academic achievement of students at the post graduate level, a sample of 414 students was randomly selected from seven different science departments in a public sector University in Lahore, Pakistan, was used. Data was collected using the Test Anxiety Inventory (TAI). Pearson correlation, multivariate statistics and regression analyses were run for data analysis. It was found that a significant negative relationship existed between test anxiety scores and students’ achievement scores. The results further showed that a cognitive factor (worry) contributed more in test anxiety than affective factors (emotional). In another study by Yousefi, Talib, Mansor, Juhari and Redzuan (2010) the purpose was to determine the relationship between test-anxiety and academic achievement among 15-19 years old high school adolescents in Sanandaj, Iran. The respondents comprised of 400 students (200 boys and 200 girls). A self administered questionnaire was used for data collection which included a Test- Anxiety Inventory (TAI), Grade Point Average (GPA) score and personal 28 information. The result showed a significant negative correlation (r = -0.23, p=.000) between test anxiety and academic achievement among adolescents. In Onyeizugbo (2010) study he examined self-efficacy and test anxiety as correlates of academic performance among 249 undergraduate students of a university in Eastern Nigeria. General Self-efficacy Scale and Westside Test Anxiety Scale were used to assess self-efficacy and test anxiety, respectively. Average scores of students in two psychology degree courses were used to assess their academic performance. The results on test anxiety showed a significant negative correlation between test anxiety and academic performance (r = -0.43, p< .001). In addition, regression analysis showed a significant model emerged, whereby test anxiety proved to be a significant predictor of the variability in academic performance, ß = -.390, p < .001. Still in Nigeria, Faleye (2010) conducted a study to investigate cognitive test anxiety and learning outcomes of 113 undergraduate education students in Obafeni Awolowo University. The sample consisted of 77 males and 36 females with an average of 24.07 years. They responded to 27 items on the Cognitive Test Anxiety Scale (CTAS) developed by Cassady and Johnson (2002). The findings showed that there was a negative relationship between cognitive test anxiety and students’ academic performance. 29 An array of studies showing a negative relationship between test anxiety and academic achievement are highlighted as follows: In Rezazadeh (2009) a statistically significant negative correlation was observed between test anxiety and academic achievement. A study conducted by Nicholson (2009) to explore the effects of test anxiety on student achievement of grade 11 students, revealed that anxiety and achievement are related to each other. Khalid and Hasan (2009) conducted a study on a purposively selected sample of 187 undergraduate students to explore the relationship between test anxiety and academic achievement and found that students with academic achievement had low test anxiety scores and vice versa. Chapell, Blanding, Takahashi, Silverstein, Newman, Gubi, and McCann (2005) conducted a research study to explore the relationship between test anxiety and academic performance. They collected data from a large sample of graduate and undergraduate students and found a significant and negative relationship between test anxiety and academic achievement. Hancock (21) investigated the effects of students’ test anxiety and teacher’s evaluation practices on students’ achievement and motivation at post the secondary level. He found statistically significant results which revealed that all students, especially students with high anxiety level, performed poorly and were less motivated to learn. Thus he concluded that that when students who are particularly test-anxious are exposed to a highly evaluative assessment environment in their educational institution, they perform poorly and are less 30 motivated to perform (Hancock, 2001). A research study conducted by Cassady and Johnson (22) to investigate the effect of cognitive test anxiety on students’ academic performance and found that cognitive test anxiety exerts a significant stable and negative impact on academic performance measures. Albero, Brown, Eliason and Wind (1997), on the basis of their research study, concluded that students having high test anxiety had significantly lower scores. Oludipe (2009) conducted a study to explore how test anxiety affects students’ performance levels in the sciences, especially in Physics, and concluded that low test-anxious students performed better than high test-anxious students on both numerical and non-numerical tasks in Physics. Other studies showing negative relationship between test anxiety and academic performance include Cassady (2004) and Stober (2004). However, other studies have shown lack of relationship between test anxiety and academic performance. In Vogel and Collins (2008) study, one group of students was given pop quizzes and another one planned quizzes. The results showed no difference in anxiety and quiz grades between the two groups. Therefore, academic performance was not found to be related to test anxiety. Furthermore, in a local study by Ndirangu, et al (2009), the results showed no significant relationship between test anxiety and academic performance (r = 0.06). However, there was a statistically significant difference (P < 0.01, t = -3.736) between test anxiety levels before and after examinations. 31 The above reviewed studies concentrated more on the direct relationship between examination anxiety and academic achievement. Nevertheless, test anxiety is a psychological construct that need to be widely studied in terms of the factors that may precipitate it. Thus, the current study, apart from investigating the relationship between examination anxiety and academic achievement, it further examined the contributions brought to examination anxiety by academic procrastination, Locus of control and academic resilience. 2.3 Academic Procrastination, Examination Anxiety and Academic Performance Procrastination has increasingly become a topic of interest across multiple fields, including education, where it is referred to as academic procrastination (Hess, Sherman & Goodman, 2000). Procrastination behavior in general is described as the difficulties that an individual has in performing daily tasks due to incapability to organize time management effectively (Ferrari, Johnson & McCown, 1995). In academic sense, it may involve doing homework, preparing for exam or doing the term papers assigned at the end of the term at the last minute. Studies have shown that the habits of studying lessons in the late hours and at the last minute are related to procrastination behavior (Hess, Sherman & Goodman, 2000; Ferrari, Harriott, Evans, Lecik-Michna, & Wenger, 1997 ). Reasons for procrastination behaviors are detailed in the research literature, and they include individual’s inefficiency in time management, difficulty of concentration or the 32 feeling of weak responsibility, anxiety and fear of being unsuccessful in one’s actions owing to negative perceptions, setting unrealistic expectations for academic performance, improper cognitive ascription and the tendency to become faultless (Ferrari, 1992; McCown, Petzel & Rupert, 1987). Procrastination in the academic realm holds many negative consequences including lost time, increased stress, lower grades, poor health, decreased long- term learning and lower self-esteem (Hoover, 2005). Procrastination is also associated with low self-efficacy, self-denigration, lower level of resourcefulness, higher levels of self-consciousness, self-handicapping and depression (Flet, Blankestein & Martin , 1995). Even though the outcomes produced by procrastinating are overwhelmingly negative, students increasingly engage in it. Indeed, research indicates that procrastination behavior is prevailing considerably in learners, especially among university students (Hoover, 2005; Haycock, 1993). Within the context of Conservation of Resources Theory (Hobfoll, 1998), most students who experience test anxiety also have a problem with procrastination. This pattern of avoidance leads to a loss cycle: procrastination leads to last minute cramming, which leads to self doubts, then excessive anxiety during testing situations and eventually poor academic performance. Hence, according to the COR Theory, it could be hypothesized that test anxiety results from avoidance and procrastination behaviors, that exists in an environment where there is limited 33 renewal of resources (Buchwald & Hobfoll, 2004) , that is, less time for exam preparation. To support this, a study by Sirois and Pychyl (2002) on 374 undergraduate students at Carleton University in Ottawa, found that procrastination is related to not only higher stress and poor coping strategies, but also avoidance behaviors. It was revealed from the study that students who suffered from these avoidance coping styles resisted completing assignments and addressing other deadlines that evoked tension and anxiety. In another study by Farran (2004) on a sample of 186 undergraduate students at Fordham University, it was found that academic procrastination was significantly and positively associated with both depression and anxiety. In other words, participants who reported higher academic procrastination also reported higher depression and anxiety. Milgram and Toubiana (1999) study also support this finding. In addition, vast research evidence shows that procrastination is also associated with such variables as poor academic performance (%alkÕs Duru, 2 Cassady, 2 dakÕcÕ, 23 Fritzsche,