Professional Documents
Culture Documents
Development of Instrument of Bullying Behaviors in Nursing Education Based On Structured Equation Modeling
Development of Instrument of Bullying Behaviors in Nursing Education Based On Structured Equation Modeling
Development of Instrument of Bullying Behaviors in Nursing Education Based On Structured Equation Modeling
Research Article
a r t i c l e i n f o s u m m a r y
Article history: Purpose: Bullying behaviors experienced during nursing education negatively affect students, educators,
Received 14 September 2017 quality of training, and patient care. The purpose of this study was to develop a valid, reliable, short, and
Received in revised form comprehensive scale to measure the bullying behaviors of nursing students in the education
16 July 2018
environment.
Accepted 23 July 2018
Methods: The Bullying Behaviors in Nursing Education (BBNE) tool was developed by adapting the
Workplace Psychological Violence Behaviors scale. The data were collected from 442 nursing students
Keywords:
from April to May 2017. The BBNE was tested using structural equation modeling, and validity and
bullying
nursing
reliability results were obtained. Moreover, complementary statistics were determined, and the violence
prevalence behaviors experienced by nursing students were reported.
reproducibility of results Results: The full BBNE scale, with 30 items and four factors, was not verified (c2/df ¼ 4.31); 12 items
students were excluded, and the modified structure with 18 items and four factors was verified (c2/df ¼ 2.60;
root-mean-square error of approximation ¼ .06). The scale's Cronbach's a coefficient is .88, and the
structure reliability is .92. Twenty-five percent of the 442 students scored 1 or higher in the total scale,
showing that they were subject to bullying behaviors.
Conclusion: The BBNE scale can be used to measure the bullying behaviors of nursing students in the
education environment. When such behaviors are identified, students can struggle on the personal and
organization level. Providing a safer and comfortable education environment for nursing students who
are the guarantee of the future of health care will positively affect the quality of education and patient
care in parallel.
© 2018 Korean Society of Nursing Science, Published by Elsevier Korea LLC. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
https://doi.org/10.1016/j.anr.2018.07.002
p1976-1317 e2093-7482/© 2018 Korean Society of Nursing Science, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
246 K. Cerit et al. / Asian Nursing Research 12 (2018) 245e250
A focus group interview was conducted with six nursing stu- of the nursing students was 21.18 (±1.61), grade point average was
dents who were experiencing or had experienced bullying at 2.91 (±0.35) of 4, and 380 (86.0%) were women while 62 (14.0%)
nursing school for preparing of the items of the BBNE scale. The were men (Table 1).
interview of the students was analyzed descriptively and system-
atically. According to the interview analysis, new 17 items were Confirmatory factor structure of the scale
added to the factors. After then, a plot study was conducted among
20 nursing students for determination of the items and under- A CFA was used to analyze if the structure with 30 items and four
standing of the statements. By recommendations of the students, factors (Model I) was not verified for nursing students. Considering
two items were removed from the item pool, and three items were the fit indexes used to determine construct validity, the Model I c2/
rewritten as statements. df value was ¼ 1720.13/399 (4.31) and RMSEA ¼ .09 (Table 2). Ac-
Fifteen items of the WPVB scale were included without change cording to these results, it was determined that Model I did not
in the BBNE scale. The BBNE scale adapted for nursing students have good fitness.
comprised 30 items in total, seven items in Factor 1 (ISO), seven The items for which error variances were related to each other
items in Factor 2 (AA), nine items in Factor 3 (PER), and seven items according to CFA correction indexes were excluded from the mea-
in Factor 4 (NEG). A six-point Likert scale was used to define the surement tool. When the errors of the items are correlated, the
frequency of behaviors (0 ¼ never experienced, 1 ¼ experience for a model's theoretical structure is impaired because the assumption
few times a year, 2 ¼ experience for a few times a month, that the variables of structural equation modeling are independent
3 ¼ experience for a few times a week, 4 ¼ experience once a day, from each other is violated.
and 5 ¼ experience a few times a day). The Cronbach's a coefficient In conclusion, 12 items were excluded from the measurement
of the original scale is .93. model, and a measurement model with 18 items and four factors
was obtained. Three items were excluded from each factor. More-
Data analysis over, the error variances of two items in Factor 1 were adjusted
according to modification recommendations. In the end, Model I
LISREL 8.80 (Scientific Software International Inc., Lincolnwood, was modified and the new structure (Model II) was created.
IL, USA) and SPSS 20.0 (IBM Inc., Armonk, NY, USA). software were The fit indexes of Model II were examined and showed that this
used for data analysis. The number, percentage, means, and stan- model demonstrated good fitness (Table 2). Table 3 provides
dard deviations were calculated to determine the grade, gender, average points for items and factors as well as standardized loads, t
and ages of the participants. A confirmatory factor analysis (CFA) values, R2 values, and average points for the items in Model II.
was run with LISREL to provide proof of the scale's construct val- The fit indexes of Model II were calculated. c2/df [330.62/127]
idity. The parameter estimations of the measurement tool were with a value of 2.60 (good fit), RMSEA ¼ .06 (good fit), SRMR ¼ .05
obtained with CFA, and the statistically suitable measurement (GFI ¼ .92, and AGFI ¼ .90) as was perfect fitness (NFI ¼ .95,
model was determined. At this point, items which were not NNFI ¼ .96, and CFI ¼ .97). The critical N value calculated for the
considered statistically significant were excluded from the model, model is 222.92. The entire scale's extracted variance is 51.61%, and
recommended corrections were made, and goodness of fit indexes the average variance extracted of all the factors in the scale is
(GFIs) of the measurement tool were obtained. provided in Table 4.
In assessment of model fit, researchers considered Chi-square
(c2)/degree of freedom (df), adjusted GFI (AGFI), GFI, normed fit Findings about reliability of the scale
index (NFI), non-NFI (NNFI), comparative fit index (CFI), root-
mean-square error of approximation (RMSEA), and standardized The Cronbach's a internal consistency coefficient for the BBNE
root mean square residual (SRMR) values. For reliability of the scale was calculated to be .88. The Cronbach's a for Factors 1e4
measurement tool, Cronbach's a coefficients were obtained using were .73, .77, .77, and .70, respectively. The CR of the scale is .92, and
SPSS 20.0 software, and the construct reliability (CR) and the the CR of factors varies between .70 and .78 (see Table 4).
average variance extracted values were calculated. Independent Bullying behaviors experienced by nursing students
samples t-test, analysis of variance, and Pearson correlation ana- (prevalence).
lyses were conducted for analysis of demographic data. Table 3 presents the average scores of the students on the BBNE
scale. The average score of the students on the total scale was
0.70 ± 0.61.
Results
Distribution of the students who experience bullying a few times
(1 point) and more violent behaviors a year is provided in Figure 1.
The study population was divided almost exactly in half be-
Twenty-five percent of the students scored 1 or more points on the
tween University A (48.0%) and University B (52.0%). The mean age
total scale, and it was determined that they were subject to bullying
behaviors in the education environment. Evaluating the students
Table 1 Characteristics of Nursing Students (N ¼ 442). who scored 1 or more points on the factors, it was determined that
Table 3 Items, Standardized Loads, t Values and R2 of Bullying Behaviors in the Nursing Education Scale.
the nursing students experienced attack on academic achievement students from the education environment” factor was excluded
the most (43.4%), then attack on personality (42.1%), isolation of from the model owing to a low R2 value (.017) and a high error
students from the education Environment (16.5%), and direct variance (.90). In addition to that, two item from “isolation of stu-
negative behaviors (10.9%). dents from the education environment”, three items from “attack
Considering the behaviors most experienced by students, it was on academic achievement”, two items from “attack on personality”,
determined that 79.4% experienced ‘not making eye contact while and three items from “direct negative behaviors” were excluded
talking’, 62.4% experienced ‘limiting self-expression’, and 58.3% without making any correction because error variances of these
experienced ‘not being trusted in the competence related to lec- were found to be related to other items' error variances. One item
tures’. Students reported that they were exposed to these behaviors from “attack on personality” factor was excluded from the model
by classmates, instructors, clinical nurses, patient's relatives, and owing to a high error variance (.84). As a result, after 12 items were
doctors. excluded, the four-factor model composed of 18 items was
No statistically significant difference was obtained between the validated.
exposure of the nursing students in the study to bullying behaviors Model fit was evaluated using fit indices as recommended by
by the school (t ¼ 0.13; p ¼ .895), grade (F ¼ 1.39; p ¼ .244), and Schreiber et al [27]. To measure GFI for the recommended theo-
gender (t ¼ 1.82; p ¼ .073) (p > .05). No statistically significant retical model, c2 (with associated degrees of freedom and p value),
difference was obtained with the correlation analysis between relative c2 (c2/df < 5.00), RMSEA < .05, GFI > .90, AGFI > .90,
mean age (r ¼ .02; p ¼ .738), grade point average (r ¼ .05; CFI > .90, NFI > .95, and SRMR < .08 are used [28]. Study results
p ¼ .338), and being exposed to bullying (p > .05). show that the BBNE scale c2/df value is 2.60; RMSEA value is .06,
and the data exhibit good fitness with the new model. Considering
other fitness indexes related to the model, it was determined that
Discussion the model has good fitness.
It can be concluded that with a higher Cronbach's a coefficient in
Evaluation of validity and reliability of the scale a new model, the items in this scale are consistent with each other
to the same extent and question the components of the same achievement are more preferred by the attacker as demonstrated
property. The Cronbach's a coefficient of the BBNE scale is .88 and of by the results of this study shown in Figure 1 [8,12,15,17].
the factors is .70e.77. Accordingly, it was determined that the scale Moreover, passive-aggressive behaviors are manipulative be-
was reliable. haviors in which the attacker is not at the forefront but exhibits
these behaviors under his or her control. Therefore, they are not
Evaluation of the items and factors in the scale easily distinguished, and the attacker frequently displays kind be-
haviors to cover up the bad behaviors [31]. Thus, it becomes harder
University years are the years of transition from youth to for the victims to understand that they are victims of bullying, to
adulthood. In this period, some students leave their families, enter immediately distinguish the attacker, and to take timely measures.
a new environment, try to make friends, are left alone, and cope In conclusion, aggressive behaviors are exhibited to isolate the
with many challenges, including financial problems. The education victim from the education environment and against his or her
environment is an arena in which the student tries to survive. The personality and student status, and thus, the victim's public dignity,
image of a student in the education environment depends on how self-respect, self-value, and self-confidence are damaged. In this
educators (academics, assistants, nurses, and so on) and friends study, the behaviors most experienced by the students were attack
treat them in public. In the literature although more than half of on academic achievement student status and manipulative be-
the students are reported to be exposed to violent behavior haviors against their personality. These behaviors are very hard to
[5,12,15], this ratio appears to be lower in this study. That may be prove as they do not leave concrete marks such as physical violence.
explained by the fact that nursing students cannot accurately
describe behaviors due to the lack of sufficient information about Study limitations
bullying and as a result of the institution's normalization of the
bullying, students may regard these behaviors as part of their A vast majority of the nursing students enrolled in the univer-
education. It is also reported that bullying behaviors in organiza- sities where the study was conducted were not willing to be
tions are related to the mentality and culture of the institutional involved in the study for various reasons. The study reached a total
management [1,22]. With the differences in the literature, it is of 23% of the students in both universities. Low participation is a
seen that the students reported the source of bullying behaviors as limitation of the study. The purpose of this study, however, is not to
their classmates [4,15], instructors [4,8,12,15], clinical nurses measure the amount of the bullying experienced by the students.
[2,8,11,12], patient relatives [2,8,11,12], and doctors [2,11,15]. The But low or high reporting of behaviors by the students is not sig-
results of the study on source of the bullying may be due to dif- nificant for the analysis of the validity, reliability, and conceptual
ferences in the definition of bullying behaviors and in the sampled construct of the scale [17,30]. In this study, the critical N value is
groups. calculated as 222. The fact that 442 students took part in the study
Bullying behaviors experienced by students in a university ed- demonstrates that a sufficient sample was obtained for data anal-
ucation environment prevent them from developing healthy re- ysis in the study.
lations and is a continuous and persistent attack on their self- Data were collected from nursing students enrolled in two state
confidence and self-respect. In this process, the students' ability universities in the capital city of Turkey, Ankara. Therefore, results
of self-expression is limited; they are continuously interrupted, are limited only to nursing students. Moreover, the study results
ignored, and excluded, and their identity is redefined accordingly. have country-specific cultural characteristics, and it is recom-
In the literature, the reason for these behaviors is to rule, exclude mended to test the results in different cultures.
from the environment, or destroy [29]. Attackers who exhibit
bullying behaviors use various strategies. Some of these strategies Conclusion
are passive, and some of them are direct negative, aggressive be-
haviors [17,30]. Direct negative behaviors can range from practical It is very important to ensure a peaceful and safe education
jokes and mild violence to physical violence. As direct negative environment to train nurses to whom we will entrust the health
behaviors are easy to distinguish and enable the victim to imme- service in the future. Bullying, however, negatively affects the
diately take measures for self-defense, it is hard for the attacker to learning process of the individual. Bullying behaviors in the edu-
reach his or her target. Therefore, these behaviors are less preferred cation environment are not desirable as they negatively affect
by the attacker as demonstrated by the results of this study shown students, educators, and education quality. Faculty, instructors,
in Figure 1 [8,15,17]. On the other hand, exposure of the victim to nurse managers, and clinical nurses must implement strategies that
direct negative behaviors indicates that the victim also is experi- will end the bullying behaviors and promote a healthy work envi-
encing intensive bullying behaviors [31]. ronment. The most efficient way to reduce the impact of bullying is
When passive-aggressive behaviors are considered separately, to prevent it. For this reason, it is necessary to identify bullying
they can be accepted as ordinary interactions of daily life. However, behaviors experienced by nursing students in the education envi-
the meaning and content of these behaviors continuously and ronment. There may also be strategies that could be implemented
frequently repeated in the education environment change in time to reduce the effect of bullying, including a supportive university
and turn into consciously organized dangerous social relations. environment and teaching assertiveness. By doing these, it might
Limiting the ability of self-expression; not being trusted in be possible to protect nursing students from bullying behaviors and
competence related to lectures; being left alone during breaks; not to prevent damages experienced in the education process. In the
making eye contact while talking; not being included in the school, study, the 18-item BBNE scale was considered sufficient, compre-
internship, and friend group; and continuously being assigned hensive, valid, and reliable in identifying the bullying behaviors
tasks over their capacity are examples of passive-aggressive be- experienced by nursing students in the education environment.
haviors. According to the findings of the study, students who were
exposed to bullying behavior were mostly subjected to the forms of Conflicts of interest
passive aggressive behavior such as, “not making eye contact while
talking” [8,12], “limiting self-expression”, and “not being trusted in The author(s) declared no potential conflicts of interest with
the competence related to lectures” [12,15], which are consistent respect to the research, authorship and/or publication of this
with literature. These attacks on personality and academic article.
250 K. Cerit et al. / Asian Nursing Research 12 (2018) 245e250
Acknowledgments 17. Einarsen S, Hoel H, Notelears G. Measuring exposure to bullying and harras-
ment at work: validity, factor structure and psychometric properties of the
negative acts questionnaire-revised. Work Stress. 2009;23(1):24e44.
The authors would like to express their sincere appreciation to https://doi.org/10.1080/02678370902815673
the health professionals who participated in this study and experts 18. Kassem AH, Elsayed RS, Elsayed WA. Bullying behaviors and self efficacy
for their opinions. among nursing students at clinical settings: comparative study. J Educ Pract.
2015;6(35):25e36.
19. Yıldırım A, Yıldırım D. Mobbing in the workplace by peers and managers:
References mobbing experienced by nurses working in healthcare facilities in Turkey and
its effect on nurses. J Clin Nurs. 2007;16(8):1444e53.
1. An Y, Kang J. Relationship between organizational culture and workplace https://doi.org/10.1111/j.1365-2702.2006.01814.x
bullying among Korean nurses. Asian Nurs Res. 2016;10(3):234e9. 20. Yıldırım D, Yıldırım A. Development and psychometric evaluation of work
https://doi.org/10.1016/j.anr.2016.06.004 place psychologically violent behaviours instrument. J Clin Nurs. 2008;17(10):
2. Magnavita N, Heponiemi T. Workplace violence against nursing students and 1361e70. https://doi.org/10.1111/j.1365-2702.2007.02262.x
nurses: an Italian experience. J Nurs Scholarsh. 2011;43(2):203e10. 21. Bardakçı E, Partlak Günüşen N. Influence of workplace bullying on
https://doi.org/10.1111/j.1547-5069.2011.01392.x Turkish nurses' psychological distress and nurses' reactions to bullying.
3. Fowler MD, Davis AJ. Ethical issues occurring within nursing education. Nurs J Transcult Nurs. 2016;27(2):166e71.
Ethics. 2013;20(2):126e41. https://doi.org/10.1177/0969733012474290 https://doi.org/10.1177/1043659614549073
4. Cooper JR, Walker J, Askew R, Robinson JC, McNair M. Students' perceptions of 22. Terzioglu F, Temel S, Uslu Sahan F. Factors affecting performance and pro-
bullying behaviors by nursing faculty. Issues Educ Res. 2011;21(1):1e21. ductivity of nurses: professional attitude, organisational justice, organisational
5. Curtis J, Bowen I, Reid A. You have no credibility: nursing students' experiences culture and mobbing. J Nurs Manag. 2016;24(6):735e44.
of horizontal violence. Nurse Educ Pract. 2007;7(3):156e63. https://doi.org/10.1111/jonm.12377
https://doi.org/10.1016/j.nepr.2006.06.002 23. Yıldırım D. Bullying among nurses and its effects. Int Nurs Rev. 2009;56(4):
6. Geller NF. Examining bullying, harrassment, and horizontal violence 504e11. https://doi.org/10.1111/j.1466-7657.2009.00745.x
(BHHV) in student nurses [dissertation]. New York: Columbia University; 24. Ekici D, Beder A. The effects of workplace bullying on physicians and nurses.
2013. p. 45e51. https://doi.org/10.7916/D82R4016 Aust J Adv Nurs. 2014;31(4):24e33.
7. Randle J. Bullying in the nursing profession. J Adv Nurs. 2003;43(4):395e401. 25. Kızıloglu E, Bedük A. The effect of civil servants' mobbing perception on
https://doi.org/10.1046/j.1365-2648.2003.02728.x organizational commitment: a practice in special provincial administration of
8. Allari R. Nursing students' experiences of work place violence during clinical Erzurum. Mediterr J Soc Sci. 2015;6(4):144e9.
placements: a qualitative study. Glob Adv Res J Med Med Sci. 2016;5(8): https://doi.org/10.5901/mjss.2015.v6n4p144
228e36. 26. Yıldırım D, Yıldırım A, Timuçin A. Mobbing behaviors encountered by nurse
9. Schaeffer A. The effects of incivility on nursing education. Open J Nurs. teaching staff. Nurs Ethics. 2007;14(4):447e63.
2013;3(2):178e81. https://doi.org/10.4236/ojn.2013.32023 https://doi.org/10.1177/0969733007077879
10. Thomas SP, Burk R. Junior nursing students' experience of vertical violence 27. Schreiber JB, Nora A, Stage FK, Barlow EA, King J. Reporting structural equation
during clinical rotations. Nurs Outlook. 2009;57(4):226e31. modeling and confirmatory factor analysis results: a review. J Educ Res.
https://doi.org/10.1016/j.outlook.2008.08.004 2006;99(6):323e38. https://doi.org/10.3200/JOER.99.6.323-338
11. Ferns T, Meerabeau L. Verbal abuse experienced by nursing students. J Adv 28. Bagozzi R, Yi Y. On the evaluation of structural equation models. J Acad Market
Nurs. 2008;61(4):436e44. https://doi.org/10.1111/j.1365-2648.2007.04504.x Sci. 1988;16(1):74e94. https://doi.org/10.1007/BF02723327
12. Clarke CM, Kane DJ, Rajacich DL, Lafreniere KD. Bullying in undergraduate 29. Field T. Bully in sight: how to predict, resist, challenge and combat workplace
clinical education. J Nurs Educ. 2012;51(5):269e76. bullying. Overcoming the silence and denial by which abuse thrives. Leed, UK:
https://doi.org/10.3928/01484834-20120409-01 Success Unlimited; 1996, 358p.
13. Ekici D, Cerit K, Gür E, Mert T, Türkmen S. Analysis of managerial problems in 30. Waschgler K, Ruiz-Hern andez JA, Llor-Esteban B, Jimenez-Barbero JA. Vertical
health services. 1st ed. Ankara: Sim Press; 2016. p. 49e464. and lateral workplace bullying in nursing: development of the hospital
14. Alzahrani HA. Bullying among medical students in a saudi medical school. BMC aggressive behavior scale. J Interpers Violence. 2013;28(12):2389e412.
Res Notes. 2012;5:335. https://doi.org/10.1186/1756-0500-5-335 https://doi.org/10.1177/0886260513479027
31. Escartín J, Rodríguez-Carballeira A, Gomez-Benito J, Zapf D. Development and
15. Çelik SS, Bayraktar N. A study of nursing student abuse in Turkey. J Nurs Educ.
2004;43(7):330e6. https://doi.org/10.3928/01484834-20040701-02 validation of the workplace bullying scale EAPA-T. Int J Clin Health Psychol.
16. Levett-Jones T, Lathlean J. Belongingness: a prerequisite for nursing students' 2010;10(3):519e39.
clinical learning. Nurse Educ Pract. 2008;8(2):103e11.
https://doi.org/10.1016/j.nepr.2007.04.003