Unravelling the complexities of nursing students' feedback on the clinical learning environment: a mixed methods approach

Background Clinical placement is an essential part of nursing education, and students' experiences on clinical placement can affect the quality of their learning. Understanding nursing students' positive and negative perceptions of clinical placement experience is therefore important. Objectives To describe nursing students' satisfaction with their clinical placement experiences and identify any variations in satisfaction based on demographic characteristics. Design Mixed methods online survey with qualitative items. Setting Four universities in Australia. Participants Students (n = 213) enrolled in an undergraduate nursing degree. Methods Between 2010 and 2012, students completed online surveys following their clinical placement experiences. The surveys included demographic questions and the Clinical Learning Environment Inventory (CLEI-19), a 19-item tool measuring students' satisfaction with clinical placement. The surveys included two open-ended questions asking students to share their most satisfying and challenging experiences whilst on placement. Descriptive statistics and thematic analyses were undertaken. Results Of the 213 participants, those in health-related employment and those with English as an additional language (EAL) were less satisfied with the clinical facility and with clinical facilitator support respectively, as indicated by the CLEI-19 subscale scores. Qualitative findings showed students were positive about the opportunity to make a difference and be involved in nursing, and negative about clinical facilitator support. Nevertheless, those who were most critical in their written comments about their placement were those who only spoke English at home. Conclusions Although the study found overall satisfaction with clinical placement, the lower satisfaction reported by students in health-related employment, and the mixed findings regarding language spoken and satisfaction, warrant further attention.


Introduction
Clinical experience is an essential learning activity and integral for the professional development of all undergraduate nursing students. As the recruitment of nursing students into higher education continues to escalate to meet the looming global mass exit of retiring nurses from the workforce (Aiken et al., 2009), the capacity for clinical placements to meet this growing student demand is increasingly being challenged. Other factors also contributing to this demand-supply strain include the decreased numbers of hospital beds, a reluctance to accept more students due to the time and resources required to support them in an already overstretched nursing workforce environment, and a lack of qualified nurse preceptors (Barnett et al., 2008;Leners et al., 2006). It is therefore vital that the nursing education sector and undergraduate students maximise learning opportunities during clinical placements.
Two factors identified as key determinants of student satisfaction of clinical learning experience are quality clinical facilitator support and the available range of clinical learning opportunities (Courtney-Pratt et al., 2012;Lewin, 2007). Both of these dimensions are measured in a recently published abbreviated Clinical Learning Environment Inventory  that assessed students' satisfaction with both clinical facilitators' support of learning and the clinical facility (Salamonson et al., 2011). Nevertheless, one of the limitations of a standardised scale is the inability to explicate explanations for participants' ratings. The addition of open-ended questions is likely to provide richer information, and perhaps insights into the rationales for students' ratings, as well as elaboration on the type of support (or lack thereof) students received from clinical facilitators and clinical staff, and information on student views of the quality of clinical placements (Agamy and Alhakim, 2013;Grebennikov and Shah, 2013).

Background
Over the last decade, the widening participation agenda in higher education has been at the core of education policy in the United Kingdom, the United States and Australia, in an attempt to redress the educational inequality between social classes and under-represented minority groups (James, 2007;Jones and Thomas, 2005;Kettley, 2007). In Australian universities, the increasing number of nursing students brings an expanding diversity, including those for whom English is an additional language, as well as those who are spending a substantial amount of time participating in paid work whilst undertaking their nursing studies (Rochford et al., 2009;Salamonson et al., 2012).
Although a number of studies have explored the contribution of the clinical facility and supervisor to student satisfaction and quality of clinical placement (Courtney-Pratt et al., 2012;Henderson and Tyler, 2011;Lewin, 2007;Salamonson et al., 2011), few studies have explored how students' demographic characteristics may affect the perceived quality of clinical placements. This study is timely given the increasing diversity of students and the impact of differing demographics on learning styles, communication skills and interpersonal relationships. For instance, students from culturally and linguistically diverse backgrounds have been reported to have different learning styles to Australian-born, English speaking students (Chan, 2003). These students are more likely to employ didactic learning techniques, showing less appreciation for problem-based and participative learning; they are also less likely to question teachers or make appointments with them due to concepts of respect and maintaining face (Jeong et al., 2011). Clinical placements provide a unique and complex learning environment that is very different to the university classroom setting. This complexity challenges students to continue to learn whilst being in an unfamiliar environment. Therefore it is important to determine how diverse student groups perceive the quality of the complex clinical learning environment.
This study sought to explore students' perceptions of their clinical learning environment by combining quantitative survey results with students' comments to open-ended survey questions. In particular, the study sought to address the following research questions: 1.
How satisfied are nursing students with their clinical placement, and what aspects of their clinical placement do they find most satisfying, and most challenging?

2.
Are there any socio-demographic group differences in nursing students' feedback of their clinical facilitators and the clinical facility?

Methods
Data presented in this paper are part of a larger study. Elsewhere, we have published findings in relation to nursing students' experiences of adversity and negative workplace cultures, and tested the psychometric properties of a revised instrument used to assess students' perception of their clinical learning environment Salamonson et al., 2011). This paper reports a later phase of this mixed method longitudinal study of the

Validity, reliability and rigour
To enhance rigour in the qualitative data analysis, responses to the open-ended questions were studied independently by two researchers (JM & KP) and the key clusters of positive and negative comments of the two dimensions of the CLEI-19 identified. Differences in the coding and classification of key themes were discussed and resolved by consensus.

Factorial validity and reliability of CLEI-19
The Kaiser-Meyer-Olkin Measure of sampling adequacy was 0.91 suggesting the correlation matrix of the CLEI-19 items was suitable for factor analysis. The scree plot indicated that the optimal number of factors to be extracted was two (Eigenvalues of 8.6 and 2.9 respectively), accounting for 60.53% of total item variance. Using exploratory factor analysis procedure, principal component analysis with Varimax rotation yielded the same two-factor solution as that previously reported (Salamonson et al., 2011

Ethical considerations
By submitting the survey responses, students accepted that they had read the study information sheet and consented to participate in the study. Students were clearly informed that their participation in the survey was both anonymous and voluntary. The conduct of the study was approved by each of the relevant University Human Research Ethics Committees.

Data analysis
Survey data was downloaded from the Qualtrics® online platform. Quantitative data were analysed using SPSS Version 21 and qualitative data were imported into QSR NVivo Version 10. The CLEI-19 was analysed using descriptive statistics. Principal component analysis (PCA) was used to determine the dimensionality, and Cronbach's alpha coefficient was used to determine the internal consistency of the tool. Pearson's chi-square, Mann-Whitney and Kruskal Wallis tests were used to test for group significances. Logistic regression analysis was used to determine socio-demographic predictors of high CLEI-19 scores. A value of P < 0.05 was considered to be statistically significant.
Qualitative data were analysed and categorised into positive and negative comments using were engaged in term-time health-related employment (Table 1).

Socio-demographic differences and level of nursing education
We compared socio-demographic differences of participants with year of enrolment in their nursing studies. Approximately 52% (n =27) of first year participants, 60% (n = 48) of second year participants and 75% (n = 61) of third year participants were engaging in term-time paid work (p = 0.016). No other statistically significant group differences were detected.

Predictors of high CLEI-19 scores
Using backward conditional method, logistic regression analysis yielded only one variable-

Qualitative findings
The length of responses to the two open-ended questions varied from 1-600 words.
Qualitative data were categorised into positive and negative comments using the two Kappa values indicate substantial to near perfect agreement.

Data integration
Interestingly, some quantitative results and qualitative data appeared to portray conflicting messages. Although the quantitative data indicated that students with EAL were less satisfied with the clinical facilitator (Table 2), this finding was not borne out in the qualitative data. For instance, seven of the nine comments related to 'the lack of engagement of the clinical facilitator' category were made by native English-speaking students (Table 4). Similarly, although the quantitative data suggested that those who were engaged in health-related work were less satisfied with the clinical placement (Table 2), the frequencies of negative comments made by those in health-related work were no more than those not participating in paid work or those in non-health-related work.

Discussion
Students with EAL were less satisfied with the clinical learning environment as indicated by the CLEI-19 scores; nevertheless, the qualitative data showed that native English speakers provided more negative responses regarding their placement. An explanation for these seemingly conflicting findings could be that students with EAL lacked sufficient confidence to provide written data in response to the open-ended comments. Previous research findings among culturally diverse students suggest that learning and teaching can be challenged by cultural and language barriers (Pitkajarvi et al., 2013).
It could also be that EAL students were less satisfied with the clinical learning environment, as reflected by the lower mean CLEI-19 score, because clinical facilitators may be less prepared to accommodate the specific learning needs of this group. In Australia, the changing demographics of the student population, with increasing numbers of students from non-English speaking backgrounds, create particular challenges for nurse educators and health care institutions (Parker and McMillan, 2007).
Clinical facilitators often have inadequate preparation for the role (Omansky, 2010) or experience workload pressures that limit their capacity for effective clinical teaching and can impact on the depth of learning opportunities for students (McCarthy and Murphy, 2010;Omansky, 2010). Additionally, much learning in the clinical setting is dialogic and informal, with a rich exchange of experiential knowledge between the student and experienced clinician (Finnerty and Collington, 2013). It is feasible that clinical facilitators may be less skilled in meeting the situated learning in practice needs of EAL students, which may help explain why these students were less satisfied than students who were native English speakers.
Another trend that was made evident through further analysis of the data was the significantly lower satisfaction of working students when compared to non-working students, particularly with those working in a health-related field. Previous research may offer several explanations for this finding. For example, it has been shown that the longer hours a student works in a week, the more negative the impact on their overall university experience (Rochford et al., 2009). Studies also report that there is a distinct conflict of interest as students try to balance time for work and study, which could contribute to higher stress levels and an overall lower perceived quality of university experiences (Nicholl and Timmins, 2005;Watts and Pickering, 2000). During clinical placement, financial stress on students can increase, particularly in rural placements, which can further exacerbate this conflict of interest (Cuthbertson et al., 2004;Schofield et al., 2009). It is possible that these were also factors in our study; however, collecting data that could explain the lower levels of satisfaction among working students would need to be further explored in future work.
In the interpretation of the results of this study, several limitations need to be considered.
Out of a population of over 4000 nursing students, only 213 (5.3%) completed the survey.
Due to this limited sample size, care needs to be taken in the interpretation of these data.

Nevertheless, the mixed methods design of this study that combined participants' responses
to standardised scales and responses to open-ended questions provided greater insight into students' clinical experiences.

Conclusion
This study set out to explore how satisfied nursing students were with the quality of their clinical placement, and which aspects they found most -or least -satisfying. Overall, One of the more interesting findings to emerge from this study is that students who were employed in health-related work were actually less satisfied with their clinical placement than those in non-health related employment, or those not in paid work. More research is needed to better understand the reasons for this, particularly given the commonly held belief that nursing students employed in health settings benefit from skills acquisition and are better prepared for transition to graduate practice.

Highlights
Although native English-speaking students were more critical in their written comments, nursing students with English as an additional language were less satisfied with clinical facilitator support, indicated by the Clinical Learning Environment Inventory scores Nursing students who were in health-related paid employment were less satisfied with their clinical learning experience than students employed in non-health settings, or those not in paid employment The use of a mixed-method approach enabled researchers to gain a more meaningful explanation of nursing students' satisfaction with their clinical learning experience *Research ighlights