THE IMPACT OF EXAMS ON ANXIETY LEVELS AMONG UNIVERSITY STUDENTS

flagyl 400mg click azithromycin

Objective: To explore the anxiety levels in exam situations among female college students at a female nursing college.

Methods: The participants were 50 first year, 46-second year, 44 third year and 47 fourth year students.  The Arabic version of the Test Anxiety Inventory was used as a measure of the students’ anxiety proneness to evaluative situations such as tests or exams.  One-Way ANOVA and the Scheffe post hoc test were used to analyze the results of the study.

Results: Significant differences were found between the scores of the 1st year and 4th year students with the latter displaying higher test anxiety as measured by the Test Anxiety Inventory.

Conclusion: Frequent college exams over the years do not have an effective role in reducing test anxiety levels among students.  Psychological intervention and strategies are required in order to lessen the high level of test anxiety for these students.

Key words: Test anxiety, Test anxiety inventory, Academic performance, Trait anxiety, State anxiety.

JRMS Dec 2006; 13(2): 14-19

 
Introduction

Test anxiety is a universal phenomenon, as old as the concept of personal evaluation, but the concentration of research activity in this field is of rather recent origin (1). Test anxiety seems like a benign problem to some people, but it can be potentially serious when it leads to high levels of distress (2).

  The fear of test is prevalent at all educational levels.  This is not surprising, since a paper and pencil tests is generally used many times each year to determine educational placement, eventually occupational acceptance and ultimately life-style (3).  Acceptable level of anxiety stimulates and improves the educational achievement, while increasing anxiety level may inversely affect attention, concentration and performance (4).

Spielberger points out to the presence of two types of anxiety:  State and trait (5).  State anxiety is conceptualized as a transitory emotional state or condition of the human organism that varies in intensity and fluctuates over time, whereas trait anxiety refers to a relatively stable personality characteristic that predisposes an individual to react to threatening situations with sometimes debilitating psychological, physiological and behavioral responses (6).

 Test anxiety is considered to be a form of state anxiety and its level can be predicted by the extent of trait anxiety in certain threatening situations, particularly those that endanger the individual’s self esteem, such as test taking (5,6).  Examinations usually arouse higher levels of state anxiety among individuals with high levels of trait anxiety (5). 

In the past four decades research has identified two main sources of test anxiety.  Firstly, evidence indicates that some highly test-anxious students have deficits in the organisational stage of test preparation due to a deficiency in learning or study skills (7-9).  Such students develop high levels of anxiety going into a testing situation because they are unable to process or organize information effectively for recall.  Therefore, the principal problem for this type of student is not the test but the preparation for the test.

  The second source of test anxiety arises from habitual, negative, irrelevant thoughts that some students have during a testing situation (10) which distract students from the task of taking the test and causes them to focus on their fears, inadequacies and past failures, thus establishing a conditioned fear.  These students may have adequate study skills but they become anxious and distracted during the test, causing poor performance.

In effect, studies have shown students with test anxiety stemming from either one of the two identified sources, i.e. inadequate preparation skills or distraction during  the  test  procedure,  develop  learned helplessness (11-13).  Learned helplessness was originally proposed as an animal model of human depression, which reflects the learned expectation of inescapable punishment. A social application of this concept involves students who have learned that they fail no matter what they do, therefore 'learned helplessness' in relation to test anxiety is due to their focusing on their previous fears, inadequacies and past failures (12). 

Additionally students have a tendency to feel higher test  anxiety  in  proportion  to  the  difficulty  of  the subjects (14,15).  Additional support to this is provided in a study conducted on 130 Libyan secondary school students attending the first and second years of the scientific stream.  The findings of this study indicated the presence of significantly higher levels of test anxiety in the second year students (16).  A South African study on university students revealed that there is no significant correlation between test anxiety and academic performance when the test concerned was reported  to  be  relatively  easy  by  the  majority  of students (17). 

There is an extensive amount of empirical evidence of the negative effects of test anxiety on academic performance. Hembree in a meta-analysis of 562 studies found that test anxiety causes poor academic achievement (18).  He concluded that test anxiety is a learned condition, which is small to non-existent in the early grades but firmly in place and negatively related to performance by grade 5 (18). 

In a cross-cultural study, El-Zahhar and Hocever found evidence for test anxiety in the USA and Brazil, and particularly in the Arab world where typically extreme consequences to performance on examinations in high school are attached (19).  It therefore seems evident that social and cultural evaluative factors have an important bearing on test anxiety.  

Test anxiety researchers have identified several personality variables which are believed to be systematically related to test anxiety (20) and which are also empirically related to academic performance: self-efficacy (21) and locus of control (18).  Students with low self-efficacy tend to be higher in test anxiety and vice-versa, and those who have an internal locus of control are more likely to have lower level of test anxiety (22). 

These personality traits are potentially important in establishing appropriate remedial interventions.  Controlled studies have previously concluded that test anxiety levels tend to fall, and examination performance improves when self-efficacy levels are made to increase (20).  More recent studies focus on the effectiveness of psychological strategies that help students to reduce the features of test anxiety, through behavioural (systematic desensitisation based on the principles of reciprocal inhibition and utilizing relaxation), cognitive-behavioural, and hypnotic techniques (2,3). 

The literature reviewed for this study suggests that test anxiety levels rise with increased difficulty of the subjects being studied.  The following study aims to provide empirical evidence to support or refute these assumptions by exploring the rates of anxiety levels as measured by the Test Anxiety Inventory (TAI) in a group of female university students in Jordan and by comparing anxiety levels between the 1st, 2nd, 3rd and 4th year students.


Methods

The study was carried out in a female nursing college.  The participants were 50 first year, 46-second year, 44 third year and 47 fourth year students.  The mean age for first year students was 18.35, for second year students 19.22, for third year students 20.23 and for the fourth year students 21.3.

The Test Anxiety Inventory (TAI) developed by Spielberger (5) and translated to Arabic language in Egypt by Abdul-Hameed (4) was used as a measure of the students’ anxiety proneness to evaluative situations such as tests or exams.  The TAI was administered to all nursing college students during the week of final examinations.

The TAI is a self-report inventory designed to measure test anxiety as a situation-specific personality trait.  The TAI consists of 20 statements, and the respondents indicate on a four point Likert-type scale how often they experience the feeling described in each statement, producing total scores of 20-80 (4).  For the purposes of this study scores of 60 to 80 were considered to reflect 'high' test anxiety, scores of 46 to 59 were considered to reflect 'moderate' test anxiety, scores of 31 to 45 were considered to reflect 'mild' test anxiety and scores of 20 to 30 were considered to reflect 'low' test anxiety.  This arbitrary division was made by calculating the mean and the standard deviation of the data which were 44.9 and 13.61 respectively.

 The cut-off point of the data was set at 44.9 and the 'moderate' and 'mild' levels of test anxiety were considered to be those who were included between plus or minus one standard deviation.  Anything above one standard deviation from the mean was considered to be 'high' test anxiety and similarly scores below one standard deviation from the mean were considered to be 'low' test anxiety.  The decimal numbers obtained following these calculations were rounded to whole numbers.

The TAI provides a measure of total test anxiety as well as measures of two test anxiety components: worry and emotionality.  Eight of the items measure the emotionality component and eight items measure the worry component.  The remaining four items contribute to the total test anxiety score but are not scored on either the worry or emotionality subscales.  The raw scores obtained from TAI are prepared for analysis by calculating their percentile ranks.

The TAI has consistently displayed sound psychometric properties in use with college student populations. Test-retest reliability coefficients range from 0.80 at 2 weeks to .62 at 6 months (5).

 Similar test-retest reliability coefficients are reported for the Arabic version of the TAI.  A study conducted on a sample of Egyptian university students report a reliability coefficient of 0.89 at two weeks (4) for and another study conducted in Libya on high school students report a reliability coefficient of 0.74 at three weeks (16). 

High internal-consistency reliability has also been reported with the Cronbach alpha reliability coefficient being 0.93 (23) and 0.67 for the worry component and 0.87 for the emotionality component (4).

 Evidence of concurrent validity has also been provided with correlation coefficients of 0.82 and 0.83 for males and females, respectively, by comparisons made between the TAI and Sarason's (6).  Test Anxiety Scale.  The Arabic version of the TAI when compare with scores obtained from the Arabic version of Sarason's indicate lower but good concurrent validity with a correlation coefficient of 0.67 (16). 

Construct validity of the arabic version has been provided by data collection in normal conditions and test conditions.  Evidence indicated that both male and female scores were significantly higher in the emotionality and worry scales when the test was performed in a test situation (4). 

Data analysis was carried out using SPSS software.  Descriptive statistics, one-way ANOVA and the Scheffe post hoc procedure were utilised to analyze the data.


Results

Overall,  the  majority  of the student nurses (41.7%,
n=78) had ‘mild’ test anxiety and the number of students who had 'low' anxiety (15%, n=28) and those who had 'high' anxiety (19.9%, n=26) were similar.  Within each year group there was a greater percentage of first year students who experienced either 'low' (32%, n=16) or 'mild' (32%, n=16) test anxiety and a greater percentage of second (47.8%, n=22), third (50%, n=22), and fourth (38.3%, n=18) year students who experienced 'mild' test anxiety.  There were considerably more fourth year students who experienced high test anxiety (25.5%, n=12) compared to the other year groups (See Table I.).

The percentile ranks of the raw data for the total score on the TAI and for the emotionality and worry dimensions was calculated and one-way ANOVA applied.  This test indicated the presence of significant differences between the different grade students on the levels of test anxiety experienced (F = 3.505, 3 and 183 df, p = 0.017).  Similarly, analysis with one-way ANOVA for the worry (F = 2.726, 3 and 183 df, p = 0.046) and emotionality component (F = 3.177, 3 and 183 df, p = 0.025) revealed significant differences between the different grade students.  These differences have been depicted in the following box-plots using the raw scores (See Figures 1,2 and 3).   

 
Fig. 1: Box-plot depicting the differences between the four student groups’ scores on the TAI.
 

Fig. 2: Box-plot depicting the differences between the four student groups’ scores on the worry component of the TAI.
 

Fig. 3:  Box-plot depicting the differences between the four student groups’ scores on the emotionality component of the TAI.

There were clear differences in the test anxiety score and in the emotionality and worry components of the test anxiety inventory.  First year students collectively scored the lowest and the fourth year students collectively scored highest in all dimension of the TAI (See Figures 1, 2 and 3).  

One-way ANOVA carried out on the first, second, third and fourth year students' mean scores on the TAI effectively established the presence of significant difference in the scoring patterns of the students (see Table II) but did not determine exactly which groups differed from each other.  Since many patterns of difference were possible secondary analysis was required to pinpoint the source of significant difference.  In order to determine where the significant differences lied, further analysis was undertaken by applying the Scheffe post hoc test. 

Post hoc analysis of the ANOVA findings revealed significant differences only between the fist year and fourth year student in all dimension of the scale.  


Discussion

The results of the present study revealed highly significant differences on test anxiety levels between 1st year and 4th year students only (p=0.02) in which 1st year students demonstrated the least amount of test anxiety.  Furthermore, a clear pattern of progressively higher test anxiety levels is evident for students at each consecutive higher class (see Figure 1.). 

This result is consistent with a Canadian study which found increasing  levels  of  anxiety with increased exposure to   tests   among  students  who  were  registered  in  a baccalaureate nursing programme (24).  Other studies on elementary students from grade 2 to grade 5 also indicate the presence of progressive increase in test anxiety (18). 

Such results suggest test anxiety levels increase over the years, because of the stress of exam situation, heavy tasks and different responsibilities that increased year by year.  High test anxiety is especially related to tasks that are difficult or complex and administered under achievement orienting conditions that emphasise the evaluation of performance (25).

  This situation  is  very  similar  to  the  position  of  the  4th year  students   who  are  expected  to  succeed  with  high  grades,  given  their  future  career  and  placement  is  highly  correlated  with their  academic  performance.

This evidence should drive researchers to ask if frequent exams experience during college levels have no positive effect on decreasing the test anxiety degrees which is expected in seniors' level.  On the contrary it seems to be increasing the test anxiety, which was evident in current study.  Various strategies of psychological intervention and counselling are helpful to settle down this phenomenon.

 There is a need to increase awareness about this problem among teachers and students to counteract any negative impacts.  Frequent exposure to examination settings (mock exams) do not seem to be effective in reducing anxiety in students who score high on test anxiety.  Evidence indicates that psychological methods that pay  attention  to the cognitive factors like self-efficacy (21) and internal locus of control (18) are probably more effective.

 These cognitive variables have been shown to be negatively related to test anxiety and positively related to academic performance (18) but the assessment of these variables was beyond the scope of this study.  Nonetheless intervention strategies focusing on cognitive factors like self-efficacy and internal locus appear to be valuable methods for helping students to reduce their test anxiety and to improve their exam performance by developing a greater sense of control over their own examination outcome, including control over their own cognitions. 

Finally there is a need for further studies to be conducted on test anxiety in the context of Arabic culture (Jordanian culture) with the specific aim of exploring the relationship between test anxiety, academic performance and personality traits such as self-efficacy and locus of control.

Table I:  Number of student nurses for scoring within each anxiety level and each grade on the TAI.

		Grade
		1st year	2nd year	3rd year	4th year	Total
Anxiety Level	Low	16 (32.0%)	5 (10.9%)	4 (9.1%)	3 (6.4%)	28 (15%)
	Mild	16 (32.0%)	22 (47.8%)	22 (50.0%)	18 (38.3%)	78 (41.7%)
	Moderate	13 (26.0%)	13 (28.3%)	15 (34.1%)	14 (29.8%)	55 (29.4%)
	High	5 (10.0%)	6 (13.0%)	3 (6.8%)	12 (25.5%)	26 (13.9%)
Total		50 (100%)	46 (100%)	44 (100%)	47 (100%)	187 (100%)

Table II.  Results of one way ANOVA for the ranked scores of the students on the TAI and on the worry and emotionality dimensions.

Rank of Scores	df	F	Sig.
Total TAI	3	3.505	.017
Worry	3	2.726	.046
Emotionality	3	3.177	.025

Table III: Presentation of the significant results of the Scheffe test.          

			Mean Difference (I-J)	Std. Error	Sig.
Dependent Variable	(I) Grade	(J) Grade
Emotion	1st year	4th year	-31.88745	10.792824	.036
Worry	1st year	4th year	-30.96043	10.827543	.046
Total score	1st year	4th year	-34.26809	10.776613	.020

 
References

1.     Ahlawat KS. Structure of Test Anxiety and its influence on achievement in different school subjects in Jordan. Abhath Al-Yarmouk: Humanities and Social Sciences Series 1989; 5(2): 79-100.

2.     Wachelka D, Katz RC. Reducing Test Anxiety and improving academic self-esteem in high school and college students with learning disabilities. Journal of Behaviour Therapy and Experimental Psychiatry 1999; 30(3):191-198.

3.     Klauber RW. Hypnosis in education and school psychology. In: Wester WC, Smith AH (Eds). Clinical Hypnosis: a Multidisciplinary Approach J.B. Lippincott Company. New York 1984; 591-622.

4.    Abdul-Hameed L. Test Anxiety Inventory. Egypt, Al-Nahda Al-Masreya 1984.

5.    Spielberger CD,Vagg RR. Test anxiety: Theory, Assessment and Treatment, Washington DC. Taylor and Francis. 1995.

6.      Sarason IG. The Test Anxiety Scale: Concept and research. In C. D. Spielberger & Sarason IG (Eds.), Stress and anxiety, New York: Wiley. 1975; 5: 193-216.

7.      Culler RE & Holahan CJ. Test anxiety and academic performance: the effects of study-related behaviours.  Journal of Educational Psychology. 1980; 72(1): 16-20.

8.   Wittmaier BC. Test anxiety, mood and performance.  Journal of Perspectives in Social Psychology 1974; 29(5): 664-669.

9.      Wine JD. Test anxiety and direction of attention.  Psychological Bulletin 1971; 76: 92-104.

10.     McKeachie  WJ,  Pollie  D,  Speisman  J. Relieving
anxiety in classroom examinations. Journal of abnormal psychology 1985; 50(1): 93-98.

11.    Paulman RG, Kennelly KJ. Test anxiety and ineffective test taking: Different names, same construct? Journal of Educational Psychology 1984; 76(2): 279-288.

12.    Ganz B, Ganz M. Overcoming the problem of learned helplessness. College Teaching 1988; 36(2): 82-84.

13.   Tobias S. Test anxiety: Interference, defective skills and cognitive capacity.  Educational Psychologyst 1985; 20: 135-142.

14.    Calvo MG, Carreiras M. Selective influence of test anxiety on reading processes. British Journal of Psychology, 1993; 84: 375-388.

15.    Blankstein KR, Flett GL, Watson MS. Coping and academic problem solving ability in Test Anxiety. Psychological Reports 1991; 69: 1128-1130.

16.    Zarrad FM. Diagnostic Study for cases of Test Anxiety Aelm Al-Nafs Al-Tarbawi: Majallet Al-Thakafeh Al-Nefsiyye 1995; 1: 57-75.

17.    Mwamwenda TS. Test anxiety and academic achievement among South African university students Psychological-Reports. 1994; 75(3): 1593-1594.

18.    Hembree R. Correlates, Causes, Effects and Treatment of Test Anxiety.  Review of Educational Research 1988; 58(1): 47-77.

19.   El-Zahhar NE, Hocever D. Cultural and sexual differences in test anxiety, trait anxiety and arousability.  Journal of Cross Cultural Psychology 1991; 22(2): 238-249.

20.    Zeidner M. Test anxiety: The state of the art.  New York & London: Plenum 1998.

21.    Parajes F. Self-efficacy beliefs in academic settings.  Review of Educational Research 1996; 66(7): 543-578.

22.    McIlroy D, Bunting B, Adamson G. An evaluation of the factor structure and predictive utility of a test anxiety scale with reference to students' past performance and personality indices. British Journal of Educational Psychology 2000; 70: 17-32.

23.   Taylor J. Development of a short form of the Test Anxiety Inventory Journal of General Psychology 2002; 8: 72-81.

24.    Rukholm EE, Viverais GA. A Multifactorial study of Test Anxiety and coping responses during a challenge examination. Nurse Education Today 1993; 13(2): 91-99.

25.    Sarason ID, Stoops R. Test anxiety and the passage of time.  Journal of Consulting and Clinical Psychology 1978; 46(1): 102-109.