The Psychological Construct of Exam Anxiety - Research Paper Example

Design a Questionnaire Total Number of Words: 2,000 Aims/Purpose: This report aims to measure the accuracy, reliability and validity of the research questionnaire designed to examine the psychological construct of exam anxiety. Objectives: To be able to discuss the following: (1) the importance of measuring the internal consistency; (2) the relevance of mean, standard deviation, standard error, skewness, and kurtosis when conducting factor analysis; and (3) the significance of Cronbach’s alpha values and regression analysis. Research Method: SPSS was used in the computation of statistics. A total of 50 research participants were invited to complete the research questionnaire presented in appendix I. Research Findings: Internal consistency of the Physical Factors behind exam anxiety was as high as 0.88. Out of the four (4) sections in the research questionnaire design, only factors related to physical signs of exam anxiety were accepted in this study. Other factors such as cognitive factors, personality factors, and factors related to the nature of exam was not included because of their relative low or negative Cronbach’s alpha values. Conclusion: The usefulness of research findings is highly based on the accuracy and validity of the research survey questionnaire. Therefore, it is always best to test not only the internal consistency of the research questions but also its relevance by computing for the Cronbach’s alpha. Table of Contents Abstract ………………………………………………………………………..………… 2 Table of Contents ………………………………………..……………………………… 3 1. Introduction ……………………………………………………….……………. 4 2. Research Method …………………………………….…………………………. 4 3. Result …………………………………………………………………………… 7 3.1 Internal Consistency ……………………………………………….. 7 3.2 Factor Analysis …………………………………………………….. 7 3.3 Cronbach’s Alpha ………………………………………………….. 8 3.4 Regression Analysis ………………………………………………... 9 4. Conclusion ……………………………………………………………………… 10 References …………………………………………………………………………….. 11 Appendix I – Research Questionnaire ………………………………………………….. 12 Appendix II – Raw Data Used in the Computations …………………………………… 15 Appendix III – Summary of Factor Analysis ………………………………………….. 16 Chapter 1 – Introduction Also known as test anxiety, exam anxiety is commonly felt by students before or during the actual exam. Specifically the term exam anxiety can be defined as a negative feeling or emotions that can serve as either behavioural or psychological related when taking either a formal or informal examination (Dusek, 1980). As a form of negative motivation, the levels of students’ exam anxiety could somehow affect the students’ overall performance in school (Hill & Wigfield, 1984). Since exam anxiety could somehow affect the students’ overall performance in school (Hill & Wigfield, 1984), it is best to learn more about all factors that can trigger the development of such negative feeling (i.e. physical factors, cognitive factors, and personality factors). Therefore, the main purpose of this report is to measure the accuracy, reliability and validity of the research questionnaire designed to examine the psychological construct of exam anxiety. As such, a research questionnaire presented on appendix I on page 12 was designed for this purpose. Eventually, the designed research questionnaire will be tested using the factor analysis and reliability test analysis. Chapter 2 – Research Method The research questionnaire was designed using a five-point likert scale questions. As such, the score of “1” means that the statement is not very true to the research participant whereas a score of “5” means that the statement given is very true to the research participants (i.e. 1 = not very true of me; 2 = a little true of me; 3 = somewhat true; 4 = often true of me; and 5 = very true of me). Out of the 30 research questions presented in appendix I on page 12, the first seven (7) questions are referring to the physical factors related to exam anxiety. Research questions number 8 to 15 refers to the cognitive factors related to exam anxiety. Research questions number 16 to 23 refers to the personality factors related to exam anxiety. Research questions number 24 to 30 refers to factors relating to the nature of the exam. (See Figure I – Summary of the Research Questionnaire Design below) Figure I – Summary of the Research Questionnaire Design Regardless of age, gender, year level, or course taken, a total of 50 students were invited to participate in this study. In relation to research ethics, each potential research participants were told about the research topic and were given the right and opportunity to accept the invitation to participate in this study based on their own free will (Whiteford & Trotter II, 2008). Aside from respecting the autonomy of each study and the practice of informed consent, the names and contact number of the research participants were not included in this report in respect to the law on confidentiality (Loue, 2000). As part of the data collection, the research questionnaire presented on appendix I on page 12 was photocopied and was officially distributed to at least 50 research respondents who voluntarily agreed to participate in this study. In general, a good questionnaire is one that has been tested for its validity and reliability. In line with this, reliability aims to test the consistency of the instrument used in the research survey study whereas the validity aims to test its accuracy in terms of measuring its objectives (Guerrero & Piasecki, 2008). Therefore, to test for reliability, the value of Cronbach’s alpha should be between 0.7 to 0.8 before accepting the research questionnaire (Cheung, 2013; Guerrero & Piasecki, 2008). Often times, Cronbach’s alpha with very low value means that the research questionnaire is not reliable. (See Table I – Summary of Cronbach’s Alpha Values below) Table I – Summary of Cronbach’s Alpha Values Values Remarks 1 Perfect 0.9 Excellent 0.7 to 0.8 Satisfactory / Good Below 0.7 Less Reliability Source: Cheung, 2013; Guerrero & Piasecki, 2008; To analyse research questions 1 to 30, the gathered data was encoded in SPSS to determine the Cronbach’s alpha value and the factor analysis. As such, the correlations between the dependent (i.e. exam anxiety) and independent variables (i.e. sweaty palms, lose appetite, fast heartbeat, nausea, sleeping problem, drinking more water, tapping fingers) were tested in each item. Using regression method, the close relationship between the dependent and independent variables will be determined by looking at multiple regression, R Square (R2), t stat, and P-values. Based on the factor analysis, it is possible to assess the quality of the research questionnaire design. The value of multiple regression ranges between 0 to 1 and that value close to 1 means having a very good regression model. In analysing the t stat value, those with negative values should be rejected. On the other hand, P-value lower than 0.10 should be accepted. Often times, P-value measures potential error between the dependent and independent variables. Therefore, a very high P-value close to 1.0 means that the error is high. Chapter 3 – Results 3.1 Internal Consistency Regardless of age, gender, nationality, nor course taken, a total of 50 research participants have completed the research questionnaire. Basically, the research questionnaire was modified directly from the Anxiety Control Questionnaire – Child version (ACQ-C) (Brown et al., 2004). This clearly explains why the internal consistency of the Physical Factors behind exam anxiety was as high as 0.88. 3.2 Factor Analysis Factor analysis is actually referring to the correlational method that is commonly applied each time there is a need to describe the underlying factors within a set of large number of variables (Dawg, 2013). Often times, it tries to identify the correlations between variables that bind them together as a single underlying factor that drives their values. In general, skewness purposely measures the symmetry of a given distribution whereas the kurtosis pertains to the peakedness or flatness of the given distribution (Norman & Streiner, 2008, p. 25). Each time the distribution is symmetric, the value of skewness should have the value of “0”. In case the value of skewness is negative, it means that the value of median is more than the value of mean which makes the distribution skew to the left. In other words, given the fact that the value of skewness is positive, it means that the value of median is less than the value of mean which makes the distribution skew to the right (Norman & Streiner, 2008, p. 25). When performing factor analysis, it is possible to closely examine not only the range and mean values of each independent variable presented in the proposed research questionnaire design but also the standard deviation, standard error, skewness of the distribution as well as the kurtosis. For example, in variable number 1 as shown in appendix 3 on page 16, the skewness of the distribution is -0.4591 whereas the kurtosis is -0.4067. It means that this particular model is far from having a normal distribution which should normally be represented by skewness equals to 0 and a kurtosis equals to 3. In this report, normal distribution will be classified as one that has skewness value not more than 2.0. Often times, negative skewness close to “0” value can already be considered as one that is positively skewed (Baguley, 2012, p. 57). Therefore, distribution of each of the 30 variables tested in this study can be considered as a normal one. The standard error simply pertains to all possible error found within the sample mean values. Often times, the use of bigger sample size can result to smaller values in standard error. In the study of medicine, the values of standard deviation and standard error should be within the range of positive or negative 1.0 (Merritt, 2007, p. 160). Based on the expected value set on the standard deviation, it is possible to remove questions number 1, 2, 3, 5, 6, 7, 8, 9, 13, 16, 19, and 22 because all these variables have standard deviation values of more than positive 1.0. (See Appendix III – Summary of Factor Analysis on page 16) 3.3 Cronbach’s Alpha In testing for the reliability of the research questionnaire design, the value of Cronbach’s alpha for the entire 30 question item was 0.2 which is relatively way below what constitute a good research survey questionnaire. To learn more about which section in this particular research questionnaire design fall short of its objective to know more about the factors that causes exam anxiety, I have decided to compute for the Cronbach’s alpha of each of the four (4) sections of the research survey questionnaire. Specifically the section on Personality Factors and Factors on the Nature of the Exam scored negative whereas the Cronbach’s alpha value of the Physical Factors and Cognitive Factors were both positive. Unfortunately, the Cronbach’s alpha value of Cognitive Factors was close to almost zero at 0.1. Therefore, the only section in the proposed research questionnaire design should include only the Physical Factors behind exam anxiety since it has a Cronbach’s alpha value at 0.5359. For this reason, only the relationship between the physical factors and exam anxiety has been computed using regression analysis. (See Table II – Summary of the Cronbach’s Alpha Value of the Research Questionnaire Design below). Table II – Summary of the Cronbach’s Alpha Value of the Research Questionnaire Design Entire Questionnaire Only Physical Factors Only Cognitive Factors Only Personality Factors Only Factors on Nature of the Exam Cronbach’s Alpha Value 0.2 0.5359 0.1 -0.18 -3.31 K 30 7 8 9 6 Ɛvar 24.11 9.49 6.80 7.82 3.43 var 29.89 17.55 7.48 6.77 0.91 3.4 Regression Analysis Based on multiple regression result, the relationship between the physical factors and exam anxiety is represented by multiple R at 0.7105 and R2 of 0.5049. Therefore, one can easily argue that the model being tested is fairly good. It means that the seven (7) independent variables being tested in this model explains 50.49% of all physical factors that represents exam anxiety. Looking at each of the seven (7) independent variables, only sweaty hands and tapping fingers have negative t-stat value of -1.6711 and -1.6739 respectively. Therefore, this particular independent variable should be rejected automatically. On the other hand, the P-value of drinking more water at 0.8857 is close to 1. Therefore, this particular independent variables should also be rejected. It means that only lose appetite, fast heartbeat, nausea, and sleeping problem clearly represents the physical factors of exam anxiety. (See Table III – Summary of T-Stat and P-Value on page 10) Table III – Summary of T-Stat and P-Value Independent Variable T-Stat P-Value Accept/Reject sweaty palms -1.6711 0.1023 Reject lose appetite 1.9146 0.0625 Accept fast heartbeat 2.5032 0.0163 Accept nausea 2.3404 0.0242 Accept sleeping problem 2.7206 0.0095 Accept drinking more water 0.1445 0.8857 Reject tapping fingers -1.6739 0.1017 Reject Chapter 4 – Conclusion The usefulness of research findings is highly based on the accuracy and validity of the research survey questionnaire. Therefore, it is always best to test not only the internal consistency of the research questions but also its relevance by computing for the Cronbach’s alpha. Often times, the process of generating multiple research survey questions is very important. In the process of conducting factor analysis, it is possible to decide which among the previously prepared research questions can be removed from the list. Regression analysis should be conducted to determine the close relationship between the dependent and independent variables being tested. As such, it is best to look at the values of multiple regression, R Square (R2), t stat, and P-values when making important decision. Being able to design a research questionnaire that has a high reliability score may contribute a lot to the study of psychology. Using a highly reliable research questionnaire on exam anxiety, psychologists could learn more about the factors that may trigger the students’ anxiety level when having an exam. By learning more about the root cause of such negative feelings, psychologists could gain better understanding on how to help the students control or avoid high levels of exam anxiety each time they are in school. References Baguley, T. (2012). Serious Stats: A Guide to Advanced Statistics for the Behavioral Sciences. London: Palgrave Macmillan. p. 57. Brown, T., White, K., Forsyth, J., & Barlow, D. (2004). The structure of perceived emotional control: Psychometric properties of a revised Anxiety Control Questionnaire. Behavior Therapy , 35: 75-99. Cheung, Y. (2013). Statistical Analysis of Human Growth and Development. Boca Raton, FL: CRC Press. p. 254. Dawg, T. (2013, March 23). Factor Analysis - An Introduction. Retrieved March 3, 2015, from https://www.youtube.com/watch?v=MB-5WB3eZI8 Dusek, J. (1980). The development of test anxiety in children. In Sarason, I.G. (ed) "Test anxiety: theory, research and applications. Hillsdale, NJ: Erlbaum. p. 88. Guerrero, A., & Piasecki, M. (2008). Problem-based Behavioral Science and Psychiatry. New York: Springer Science. p. 204. Hill, K., & Wigfield, A. (1984). Test Anxiety: A Major Educational Problem and What Can Be Done about It. The Elementary School Journal , 85(1): 105-126. Loue, S. (2000). Textbook of Research Ethics: Theory and Practice. New York: Kluwer Academic. p. 147. Merritt, J. (2007). Epidemiology, Biostatistics, and Preventive Medicine. 3rd Edition. Philadelphia, PA: Sauders Elsevier. p. 160. Norman, G., & Streiner, D. (2008). Biostatistics: The Bare Essentials. Ontario: BC Decker Inc. p. 25. Whiteford, L., & Trotter II, R. (2008). Ethics for Anthropological Research and Practice. Long Grove, Illinois: Waveland Press Inc. p. 69. Appendix I – Research Questionnaire Thank you for your interest in this study of exam anxiety. Participation involves completing a questionnaire with 30 questions. All of the information you provide will be treated confidentially, and results will be presented anonymously. There is no need to include any identification data. Instructions: Following are some statements about how people experience anxiety during exam period. Please think about past exams and circle the number that best represents what you experience. 1 – Not very true of me 2 – A little true of me 3 – Somewhat true 4 – Often true of me 5 – Very true of me 1. I have sweaty palms during an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 2. I lose my appetite prior to having an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 3. My heart tends to beat faster before an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 4. I feel nauseated before an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 5. I don’t have trouble sleeping the night before the exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 6. I drink more water than usual during an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 7. I find my fingers tapping or my legs jiggling while taking a test. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 8. I tend to forget some knowledge that I’ve revised during an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 9. I make unnecessary mistakes to questions I know the answer to. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 10. My mind goes blank during an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 11. I can settle into an exam environment in a short time. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 12. I don’t think I’ll pass the exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 13. I usually have negative thoughts about the result of my exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 14. I tend to accumulate negative thoughts about myself in general during an exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 15. As soon as an exam is over, I find that I can stop worrying about how I did. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 16. I am confident after having done enough revision for your exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 17. I enjoy being tested on my knowledge. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 18. I often feel anxious when performing a task. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 19. I enter the exam feeling confident. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 20. I feel that I am not smart enough to do well. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 21. Getting a good grade on a test doesn’t increase my confidence about the next exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 22. I am very concerned about how others will view me if I do poorly on a test. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 23. My emotion feelings affect my performance. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 24. I feel more confident about an exam on the topic which I enjoy. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 25. I feel more anxious when the exam takes place in the morning. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 26. If an exam takes place in the evening, I feel more calm before the exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 27. I feel more calm if I have done enough revision for the exam. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 28. I expect to do poorly on a test no matter how much or how hard I study. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 29. I can do better in an exam which is hosted on a computer than writing on a paper. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me 30. I feel more confident about an exam on the topic which I find it easy. Not very true me --- 1 --- 2 --- 3 --- 4 --- 5 --- Very true of me Appendix II – Raw Data Used in the Computations Appendix III – Summary of Factor Analysis Variable Range Mean Stdev Std Err Skew Kurt 1 4 3.4879 1.1387 0.1626 -0.4591 -0.4067 2 4 3.3061 1.1938 0.1705 -1.1093 -0.0517 3 4 2.8775 1.2012 0.1716 -0.6664 0.3951 4 3 3.8775 0.9271 0.1324 -0.6718 -0.4019 5 4 3.4081 1.2569 0.1795 -1.0664 -0.1771 6 4 3.4081 1.3222 0.1888 -1.2920 0.2604 7 4 3.1836 1.1119 0.1588 -0.8414 0.0947 8 4 3.6734 1.0285 0.1469 0.2430 -0.6099 9 4 3.6938 1.0449 0.1492 0.2335 -0.7132 10 2 4.3061 0.7130 0.1018 -0.8553 -0.5283 11 3 2.4897 0.7107 0.1015 -0.1593 -0.3256 12 3 4.1020 0.8227 0.1175 -0.0072 -0.6629 13 4 3.734 1.0562 0.1508 0.2577 -0.7642 14 4 2.285 1 0.1428 -0.2310 0.4281 15 4 3.8571 0.9789 0.1398 0.4343 -0.8136 16 4 3.1020 1.1409 0.1629 -0.7582 0.1435 17 3 2.9387 0.8516 0.1216 -0.6791 -0.3021 18 3 3.8367 0.8978 0.1282 -0.5305 -0.3851 19 4 2.5714 1.0408 0.1486 -0.7174 0.1485 20 4 4 0.8660 0.1237 1.8394 -1.003 21 4 3.8775 0.9044 0.1292 1.001 -0.8069 22 4 2.6122 1.1146 0.1592 -0.7266 0.1760 23 3 4.2857 0.7637 0.1091 0.2408 -0.8357 24 2 4.3265 0.6578 0.0939 -0.6685 -0.4620 25 3 3.7346 0.9741 0.1391 -1.0124 -0.1369 26 3 1.7755 0.6540 0.0934 1.5895 0.7283 27 2 4.3469 0.5969 0.0852 -0.6140 -0.2906 28 3 1.7346 0.6700 0.0957 1.4174 0.7990 29 3 3.9183 0.8620 0.1231 -0.3653 -0.4487 30 2 4.2448 0.7781 0.1111 -1.1855 -0.4643 Read More