Stress is a frequently heard phenomenon in society today, particularly in the work setting. The transition to the role of a university student involves many changes. Students, as a result, also describe that they feel stressed and unable to cope.
Our nursing students are experiencing considerable stress in today's academic programs. A current vision of nursing education is to help students develop into competent nurses who are prepared to be leaders in health care reform. Nurses must begin to take on a more active role in influencing health policy. In order to prepare students to assume the future extensions of their roles, it is imperative to identify sources of stress which may prevent them from functioning optimally.
An extensive literature review across health-related and other disciplines was conducted. Most literature of relevance to university student stress was found in the disciplines of medicine and nursing, as well as some from psychology, pharmacy and dentistry. There were some difficulties to overcome in reviewing the significance of this literature such as the failure of authors of unpublished PhD dissertations to have the results of their studies disseminated in journal publications; the use of different data collection instruments by researchers; the lack of replication studies to enhance the generalizability of findings; and the varying foci of each study in relation to student stress.
Despite these difficulties, researchers generally agreed that student stress is a common complaint which deserves investigation particularly because of the attrition rates in schools of nursing and the burnout rate experienced within the medical and nursing professions (Backer, 1989; Dillon, 1988; Haack, 1988; Huerta, 1990; Lindop, 1991). Some studies (Backer, 1989; Foley & Stone, 1988; Russler, 1991; Speck, 1990) were concerned primarily with testing interventions which may assist students to cope with stress, for example, by use of various stress management techniques. Few of these interventions, however, seemed to make significant differences to students' levels of stress.
Other studies focused on the identification of how student stress is manifested. Typically, findings were reported in relation to substance use (Floyd, 1991; Oleckno & Blacconiere, 1990; Young, 1987), academic achievement (Chacko & Huba, 1991; Huerta, 1990), career choices (Ortmeier, Wolfgang, & Martin, 1991; Wolfgang & Ortmeier, 1993) and symptomatology of depression (Haack, 1988; Wolf, Von Almen, Faucett, Randall, & Franklin, 1991).
Still other researchers (Beck & Srivastava, 1991; Harbin, 1989; Pagana, 1988; Toews, Lockyer, Dobson, & Brownell, 1993; Williams, 1988) were concerned with the identification of the sources of stress perceived by students which they, as educators, hoped to be able to ameliorate. The concerns of these researchers are of particular relevance to the study being presented.
This study expands on prior work by Beck & Srivastava (1991) on perceived level and sources of stress in baccalaureate nursing students which was initiated in response to persistent student complaints of feeling "stressed out" during their years at the School of Nursing. The study identified significant levels of stress in this population and the investigators emphasized the need to examine the sources of stress and initiate strategies to reduce the high levels of stress. Beck and Srivastava identified some limitations which a replication study should address: namely, small sample size and lack of comparison groups which limited the generalizability of results. With no comparison group, it was impossible to know if the stress levels perceived by nursing students were similar to, or different from other university students.
The current study was undertaken to determine if the findings of high stress levels are consistent over time, since this has implications for curriculum planning/revision, student counseling, as well as the overall quality of the beginning practitioner who emerges from a program of nursing studies. This expanded replication study included a total of four health care disciplines: nursing, medicine, pharmacy, and social work. The overall purpose of this study was:
* to investigate the perception of level and sources of stress in nursing students enrolled in a baccalaureate program at a university in Atlantic Canada (hereafter referred to as Nursing A).
* to compare these students to nursing students enrolled at a large university in Central Canada (hereafter referred to as Nursing B).
* to compare these students to those enrolled in other health-related disciplines, specifically: Medicine, Pharmacy and Social Work at the university in Atlantic Canada.
This study was descriptive correlational in design. The variables to be studied included perceived levels of stress, perceived sources of stress, general health, and selected personal characteristics of the respondents and their environments. The population consisted of full-time undergraduate university students enrolled in courses within years 2, 3 and 4 of the baccalaureate programs of the selected disciplines. The population included students from the faculties of nursing (two groups), medicine, pharmacy and social work. Quota sampling was used to obtain respondents from a potential pool of 800 students enrolled in years 2, 3 and 4 of the selected disciplines.
The sample consisted of 552 students (N=552). Absenteeism from elftes was the major reason for lack of participation. Only a handful of students actually refused to participate. Data Were gathered from years 2, 3 and 4 of the disciplined of Nursing, Pharmacy and Social Work. Data were gathered- in years 2 and 3 for Medicine. Students in year 4 of Medicine were in various clinical rotations around the city and could not be accessed within the constraints -of the sampling procedure.
Mean ages across all disciplines were close; the range was 22.3 years to 24.4 years. However, the Nursing B group (mean=24.4 years) was significantly older than Nursing A (mean=22.5 years) and Pharmacy (mean=22.3 years).
The Pharmacy and Medicine groups had almost equal numbers of male and female students (Pharmacy=55% Female/45% Male and Medicine=50% Female/50% Male). Nursing A, Nursing B and Social Work had predominantly female populations (Nursing A=90% Female/10% Male, Nursing B«90% Female/10% Male and Social Work=88% Female/12% Male).
The research proposal was reviewed and approved by the Human Subjects Review Committee at the university in Atlantic Canada. Data were gathered across all disciplines within the same week, mid-semester, a time when students seemed most stressed. Only those students who were enrolled full-time within their disciplines were asked to participate. The students were considered competent to understand the study purpose and to consent freely. Subject consent was presumed on the basis of completion of the questionnaire. To gain access to the target population of students in each discipline, a total of seven deans/directors and/or their associates as well as 25 professors were contacted. A research team member visited the participating class, at a time previously arranged with the course professor, and students were informed of the purpose of the study (verbally and in writing) and asked to voluntarily participate. Time to complete the questionnaire was about 30 minutes. Questionnaires were then distributed to the class, at which point the team member left the room. Completed questionnaires were collected by a designated student in the class who immediately returned them to the team member who was waiting in the immediate area.
Data were collected by means of a questionnaire consisting of three main instruments: the Beck-Srivastava Stress Inventory (BSSI), the General Health Questionnaire (GHQ), and a demographic profile. The Beck-Srivastava Stress Inventory (BSSI) was developed by the original investigators so that student levels of stress and specific sources of stress could be identified (Beck & Srivastava, 1991). This instrument has two parts. The first part requests the respondent to describe a stressful event which occurred in the previous month and was related to his/her role as a student. The respondent is asked to further describe the event in terms of why it was a problem, how it made the respondent feel and how he/she coped. The respondent is asked to further qualify his/her feelings about the event by scoring 15 adjectives on a 5-point Likert scale.
Part II of the BSSI consists of 44 items associated with stress in the student role. These items reflect stressors from academic, financial, interpersonal and clinical areas. The clinical items have been grouped together in a separate section so that respondents not engaged in clinical course work can omit those questions. These items are rated by the respondent using a 5-point Likert scale ranging from "not stressful" to "extremely stressful," i.e., (l)=not at all (2)=slightly (3)=quite (4)=very (5)=extremely. An overall stress score is obtained by adding the ratings of each item. Using Likert scoring, the possible range of scores for the BSSI is 44-220. Since the BSSI is a new instrument, norms have not yet been established, but mean scores above 88 are presently being considered as problematic stress levels since this would indicate that a subject is more than slightly stressed. The results for Part II will be presented in this article. The results for Part I, consisting of qualitative data will be reported at a later date.
The BSSI has excellent reliability based on the previous and current studies. For the original study, the alpha coefficient was .90; for this current study, the alpha coefficient was .82 demonstrating high consistency. Face and content validity were established for the BSSI using a panel of faculty/student experts. Factorial analysis was done on the data of the original study to assess construct validity and the instrument has been found to have a stable factorial structure. This analysis has also been completed for the present study and further support for validity has been found. The psychometric analyses of this data will be reported in detail elsewhere (Beck & Srivastava, 1995).
The General Health Questionnaire-30 item version (GHQ-30) was chosen for its brevity, ease of administration and established validity and reliability. This tool was selected so that the general level of physiological and psychological health of the respondents could be determined. Cronbach's alpha of the GHQ-30 has ranged from 0.85 to 0.93 in previously reported studies and the questionnaire possesses content validity and construct validity as demonstrated by its sensitivity to general dysphoria in subjects with non-psychotic illnesses (Goldberg & Williams, 1991). Respondents were asked to rank their present and recent experiences on a 4-point Likert scale ranging from "less so than usual" to "much more so than usual." An overall score obtained by adding the ratings of each item was used to measure the general health of the respondents. Using simple Likert scoring, the possible range of scores for the GHQ-30 is 0-90. A cutting score of 39/40 for normal versus cases (cases meaning outside of normal limits) was set by Goldberg (1972). Scores greater than 39 indicate a high probability that the individual is experiencing a level of distress that could manifest itself in either a physical and/or psychiatric illness.
The third instrument was simply a profile sheet designed to obtain personal and academic demographic information about the respondent. This information may be useful in identifying correlations among the demographics, the GHQ and BSSI.
Data were analyzed using the SPSSx Statistical Package. Statistics included analysis of variance (ANOVA and ONEWAY), frequency distribution, measures of correlation, item analysis and factor analysis. Demographic variables were analyzed using frequencies to determine the distribution of students in relation to each variable. Frequencies were used to determine the mean scores on the individual items of the BSSI and the GHQ.
Analysis of variance (ANOVA) was used to compare the mean scores among groups for the GHQ and the BSSI (Table 1). Where significance existed, a multiple comparison test, that is, the Scheffe test, was then used to determine exactly where the differences lie between groups.
Significant differences at the p<.0000 among groups were noted for total GHQ means. Mean scores ranged from [26.27 to 39.60]. All groups had fairly high GHQ scores. Nursing A is at the cutting score for caseness as determined by Goldberg, followed closely by Nursing B (Goldberg & Williams, 1991). These scores indicated a high probability that the individual is predisposed to developing either a physical or psychiatric illness.
A significant difference at the 0.05 level was found between Nursing A, Nursing B and Social Work with Medicine. In addition, Nursing A and Nursing B were significantly higher than Pharmacy.
The mean Likert BSSI scores for all groups are also displayed in Table 1. The mean scores ranged from 76.48 to 107.83. Scheffe's reveals a pattern of significance at the 0.05 level similar to that of the GHQ. Nursing A had a mean which was significantly higher than each of the other disciplines including Nursing B but Nursing B was significantly higher than only one other discipline, i.e., Pharmacy.
The means of GHQ and BSSI by each year in program were then compared across disciplines. Table 2 displays the means of the GHQ and BSSI for second year students, in all disciplines.
With respect to the GHQ scores, an analysis of variance showed significant differences at the p<. 0000 among groups. The multiple comparison test revealed a significant difference at the 0.05 level between Nursing A and all other disciplines with Nursing A having a very high GHQ score, well above the cutting score of 39/40. Social Work and Nursing B were significantly higher than Pharmacy and Medicine.
Total GHQ & BSSI All Years
With respect to the BSSI scores, significant differences occurred among groups at the p<.0000. Nursing A was significantly higher than all other disciplines at the 0.05 level. Nursing B was significantly higher than Pharmacy and Medicine.
Table 3 shows the mean GHQ scores and BSSI scores for all disciplines in year 3 of the programs. The GHQ scores for both Nursing A and Nursing B were significantly higher than Medicine at the 0.05 level. Nursing A was also significantly higher than Pharmacy. However, in year 3, the mean scores of all disciplines are below the cutting score 39/40.
The BSSI scores for both Social Work and Nursing A were significantly higher than Pharmacy. Nursing A was also significantly higher than Medicine and Nursing B.
Table 4 shows the mean GHQ scores and BSSI scores for all disciplines except Medicine at year 4. With respect to the mean GHQ scores, analysis of variance showed significant differences at the p<.0000 among groups. The multiple comparison test revealed only one score which was significantly higher than all other disciplines, that is, Nursing B. The GHQ score for Nursing B is also somewhat higher than the cutting score of 39/40.
Significantly higher BSSI means for Nursing A and Nursing B in relation to Pharmacy were noted. Nursing B was also significantly higher than Social Work.
Recall that the BSSI measures not only levels of stress but the sources ofthat stress as well. The 44 items of the BSSI which reflect stressors from academic, financial, interpersonal and clinical areas were analyzed for frequency of reporting as stressful.
Items rated 3 to 5 on the Likert scale (quite, very, extremely stressful) were considered to be stressors. Any item considered to be a stressor by at least 50% of the respondents was then included in our ranking of these stress items. We have chosen to focus our report of this ranking scheme to the top ranked items of each discipline.
Mean Scores for GHQ & BSSI- Year 2
An analysis of the ranking of stress items by discipline revealed that there were some similarities across disciplines with respect to the most frequently reported stressors by the majority of students. Table 5 illustrates those stress items most commonly reported. Each discipline, however, ranked these items somewhat differently.
In each health-related discipline, students identified the amount of class material to be learned, lack of free time, exams and grades, long hours of study, and the difficulty of work to be learned as common stressors. Financial responsibilities, and attitudes and expectations of other professionals toward their own profession were also common sources of stress to each discipline. Of the list of 11 stress items found in Table 5, the seven previously listed were the most often repeated items in the top 10 stress items identified by each discipline. The remaining items within Table 5, i.e., administrative responses to student needs, negative personal habits (such as procrastination, tardiness, perfectionism) and too much responsibility contributed to the top 10 stress items for most, but not all disciplines. Peer competition contributed to the top 10 stress items for only two disciplines.
With respect to Nursing B, 9 of the 1 1 items previously mentioned helped comprise its respondents' list of 11 stressors. These students did not identify administration's response to their needs as a source of stress. However, the majority of Nursing B respondents did identify client care responsibilities and feelings of inadequacy related to clinical performance as two additional sources of stress.
Within Pharmacy, Medicine and Social Work, the majority of each discipline's respondents identified only 10 sources of stress. The majority of respondents from Nursing B identified 11 sources of stress. The majority of respondents from Nursing A identified 19 sources of stress.
Mean Scores for GHQ & BSSI - Year 3
For Nursing A, the respondents cited 10 of the previously identified items, within its top ranked sources of stress. The respondents within this discipline, however, were unlike all others with respect to the identification of sources of stress, in that they identified an additional, extensive list of 9 stressors, 7 of which were related to the respondents' clinical experiences.
Our findings revealed that baccalaureate nursing students, regardless of year in program or university of attendance, experienced higher levels of stress and more physical and psychological symptoms than students in other health-related disciplines. Limited reports of similar findings have been previously noted (Williams, 1988) with respect to a sample of year 2 nursing students in a clinical course who scored higher stress levels than the norm established for the general student and non-student population. For some disciplines it is obvious that students seem to be handling the identified stressors while in other disciplines, particularly nursing, more stressors were reported with greater intensity. It is of particular concern that, since the very nature of the nursing profession requires that a person gives a lot of him/herself, particularly within client interactions, the clinical experiences of some baccalaureate nursing students seem to be contributing greatly to their perceived levels and sources of stress. These findings seem similar to the results of other studies which report that academic concerns related to curriculum matters or the program of studies are the major concern of nursing students (Dillon, 1988; Pagana, 1988; Zujewskyj & Davis, 1985). They also bear some similarity to the findings of Haack (1988) which suggest that clinical learning experiences may be a primary cause of the burnout experienced during a university nursing program.
The results of this study have implications for curriculum planning, evaluation and revision, as well as for student counseling in relation to stress management. If we reflect upon those stress items common to all healthrelated disciplines, it may not be particularly surprising to faculty that the nature of the field of knowledge which must be mastered by these students contributes many items to their list of perceived stressors, e.g., amount of class material, lack of free time, long hours of study, difficulty of class work, exams/grades. These findings were consistently reported in other studies (Chappie, Allcock, & Wharrad, 1993; Wolf, Faucett, Randall, & Balson, 1988). Early access to the acquisition of time management and other such valuable skills may impact on the effect of these stress items. In addition, it is necessary to look at the reasons why particular stress items are reported with greater intensity by the nursing groups, in attempt to generate plans to reduce the high stress levels.
Mean Scores for GHQ & BSSI- Year 4
With respect to the common stressors, perhaps we need to take a close and deliberate look at the nature and quantity of assignments we require of studente across courses within a given semester. Are we guilty of demanding too many submissions at the cost of students' independent, reflective time? What are the factors that cause stress levels to be higher in nursing than other health professions?
It is imperative, as faculty in schools of nursing, that we address the circumstances and factors relating to the higher stress levels than are seen in other professional schools. Our biggest challenge may be with respect to those stressors related to our students' clinical experiences, since these were the items that seemed to differentiate the nursing populations from the other disciplines. For many of our generic degree programs, students continue to gain their clinical experiences in small groups of about 10 students under the direct supervision of a faculty member. Given the nature of those stress items identified by respondents in Nursing and related to clinical, i.e., feelings of inadequacy related to clinical, client care responsibilities, too much responsibility, lack of timely feedback, atmosphere created by clinical faculty, etc., it is apparent that the teaching style of faculty may have an impact on students' perception of stress. How do we as faculty balance the need to ensure client safety by evaluating the competency of students and yet provide an atmosphere for learning which fosters in the student selfconfidence, independence and accountability for deficits with respect to cognitive, psychomotor and attitudinal competency? How can we best break that cycle of anxiety, stress, lack of self-confidence and self-esteem which not only affects the individual student and his/her relationship with faculty but which, as we glean from this study, i.e., Nursing A's stress items; clients' attitudes toward me and talking to clients about personal problems, also affects the student-client relationship? Perhaps it is important to examine the very way key people are chosen for clinical teaching assignments, i.e., the basis and rationale for clinical assignment of faculty.
Comparison of Stress Items by Ranking
These are just some of the questions which we as nursing faculty should contemplate, collectively, in our attempt to make our baccalaureate programs less foreboding passages during which our students are expected to emulate the notion of caring they have gained as a result of our mentoring. As nursing faculty, we need to examine the strategies we use as we teach in the clinical setting. In interviewing students outside the study, they shared that they often feel intimidated by faculty and fear being embarrassed in front of others if they don't always have the correct answers. When some students have levels of distress that are high enough for them to develop mental illness or even become suicidal, then there is no choice but to closely scrutinize what is happening as students are prepared for a vital role (Goldberg & Williams, 1991). Nursing students often work with vulnerable clients who are in need of strong support persons to advocate for and assist them to manage their own health, and these students must be prepared and ready to assume this role with confidence and competence.
A stronger professional identity for nursing students and the profession in general may help to overcome some of the stress being experienced. The students need to feel confident in their competencies as practitioners. As leaders in health care, nurses need to be capable of expanding their role while at the same time maintaining the integrity of the profession.
The authors plan to broaden the research baseline to include other populations of students, thus enabling more generalization of the results. Further support for validity and reliability of the Beck & Srivastava Stress Inventory is needed. It is hoped that cross-cultural reliability and validity will be established through replication studies, currently underway in other countries.
The study expanded on previous research done by two of the investigators to include a larger population and comparison groups both within and outside of nursing. The findings showed that nursing students identified more stress than their counterparts in other health-related disciplines-medicine, social work and pharmacy. Even though there were stress items common to all groups, the level of handling distress associated with these stressors was better in the health-related disciplines than in nursing. The additional sources of stress identified by nursing students were related to clinical experiences which strongly suggests that as educators, we need to take a close look at the clinical education process which encompasses the methods of teaching, the choice of clinical faculty, the attitudes toward clinical and the resultant expectations. If nursing is to survive and grow, it is imperative that we begin to be more sensitive to the stress sources that can be controlled in our education programs.
- Backer, B. A. (1989). Utilization of assessment as an intervention in nursing student stress: An exploratory project. Unpublished DSW dissertation. City University of New York, New York, New York.
- Beck, D.L. & Srivastava, R. (1991). Perceived level and sources of stress in baccalaureate nursing students. Journal of Nursing Education, 30(3), 127-133.
- Beck, D. & Srivastava, R. (1995). Development, reliability and validity of the Beck & Srivastava Stress Inventory. Submitted for publication.
- Chacko, S.B. & Huba, M.E. (1991). Academic achievement among undergraduate nursing students: The development and test of a causal model. Journal of Nursing Education, 30(6), 267273.
- Chappie, M., Allcock, N., & Wharrad, H.J. (1993). Bachelor of nursing students' experiences of learning biological sciences alongside medical students. Nurse Education Today, 13, 426-433.
- Dillon, N.B. (1988). The relationship of selected characteristics to anxiety levels of students in the first clinical course in baccalaureate nursing programs. Unpublished doctoral dissertation. University of Mississippi.
- Floyd, J. (1991). Nursing students' stress levels, attitude toward drugs, and drug use. Archives of Psychiatric Nursing, 5(1), 46-53.
- Foley, J. & Stone, G. (1988). Stress inoculation with nursing students. Western Journal of Nursing Research, 10(4), 435-448.
- Goldberg, D.P. (1972). The detection of psychiatric illness by questionnaire. London: Oxford University Press.
- Goldberg, D. & Williams, P. (1991). A user's guide to the General Health Questionnaire. Great Britain: NFER-Nelson Publishing Co. Ltd.
- Haack, M. (1988). Stress and impairment among nursing students. Research in Nursing and Health, 11(2), 125-134.
- Harbin, P.D.O. (1989). A Q-analysis of the stressors of adult female nursing students enrolled in baccalaureate schools of nursing. Unpublished DSN dissertation. University of Alabama at Birmingham.
- Huerta, CG. (1990). The relationship between life change events and academic achievement in registered nursing education students. Unpublished doctoral dissertation. Texas A & M University.
- Lindop, E. (1991). Individual stress among nurses in training: Why some leave while others stay. Nurse Education Today, 11(2), 110-120.
- Oleckno, WA. & Blacconiere, M.J. (1990). A multiple discriminant analysis of smoking status and health-related attitudes and behaviours. American Journal of Preventive Medicine, 6(6), 323-329.
- Ortmeier, B.G., Wolfgang, A.P., & Martin, B.C. (1991). Career commitment, career plans, and perceived stress: A survey of pharmacy students. American Journal of Pharmaceutical Education, 55, 138-139.
- Pagana, K.D. (1988). Stresses and threats reported by baccalaureate students in relation to an initial clinical experience. Journal of Nursing Education, 27(9), 418-424.
- Russler, M.F. (1991). Multidimensional stress management in nursing education. Journal of Nursing Education, 30(8), 341-346.
- Speck, B.J. (1990). The effect of guided imagery upon first semester nursing students performing their first injections. Journal of Nursing Education, 29(8), 346-350.
- Toews, J.A., Lockyer, J.M., Dobson, D.J.G. & Brownell, A.K. (1993). Stress and harassment: Perceptions of students and residents. Academic Medicine, 68(10), S46-S48.
- Williams, R. A. (1988). The relationship of cognitive styles and stress in nursing students. Western Journal of Nursing Research, 10(4), 449-462.
- Wolf, T.M., Faucett, J.M., Randall, H.M., & Balson, P.M. (1988). Graduating medical students' ratings of stresses, pleasures and coping strategies. Journal of Medical Education, 63, 636-642.
- Wolf, T.M., Von Almen, T.K., Faucett, J.M., Randall, H.M., & Franklin, FA. (1991). Psychosocial changes during the first year of medical school. Medical Education, 25(3), 174-181.
- Wolfgang, A.P. & Ortmeier, B.G. (1993). Career commitment, career plans, and job-related stress: A follow-up study of pharmacy students as pharmacists. American Journal of Pharmaceutical Education, 57, 25-28.
- Young, E.H. (1987). Relationship of residents' emotional problems, coping behaviors, and gender. Journal of Medical Education, 62, 642-650.
- Zujewskyj.T. & Davis, L. (1985). Sources and effects of anxiety in videotape learning experience. Nursing Papers, 17(3), 75-85.
Total GHQ & BSSI All Years
Mean Scores for GHQ & BSSI- Year 2
Mean Scores for GHQ & BSSI - Year 3
Mean Scores for GHQ & BSSI- Year 4
Comparison of Stress Items by Ranking