This cross-sectional study adopted students of an Institute of Technology in 2010 as the study population. The class as a unit, stratified cluster random sampling based on the number of classes was used to recruit fourth and fifth year students. A structured questionnaire was used to collect data on the factors affecting sleep quality and learning satisfaction in nursing students. Power analysis was used to calculate the appropriate sample size. This study referred to Polit and Beck’s [20] study, which suggested adopting an effect size of 0.2, power of 0.8, and α as 0.05 for nursing-related studies. The calculated sample size was 197. Questionnaires were sent to samples be self-administered. A total of 207 students responded, yielding a response rate 96%.

Demographic Data Questionnaire: The demographic questionnaire contained information on the year of study, participation in extracurricular activities, holding of part-time jobs, motivation to study nursing, semester scores, and interest in nursing of the students.

Chinese version of the Pittsburgh Sleep Quality Index (C-PSQI): We used the translated (Chinese) version of the Pittsburgh Sleep Quality Index of Chiang [21] , originally developed by Buysse, Reynolds III, Monk et al. [22] . This is a self-rated questionnaire which assesses sleep quality and disturbances over a 1-month time interval. Nineteen individual items generated seven “component” scores: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficient, sleep disturbances, use of sleeping medication, and daytime dysfunction. The seven components were scored from 0 to 3 points, with each component weighted equally. A score of “0” indicated no difficulty, while a score of “3” represented severe difficulty. The possible total scores of the seven components ranged from 0 to 21. A higher score signified worse sleep quality. “Poor sleep” was defined as a PSQI score of greater than 5 [22] . The Cronbach’s α value for this study was 0.83.

Learning Satisfaction Scale: The 22-item scale, which was developed by Wang and Lin [1] study, which focused on the relationship between perceptions toward clinical nursing instructors’ caring behavior and learning satisfaction among nursing students. The scale of learning satisfaction contains affective, cognitive, and psychomotor domains, and a 4-point Likert scale was applied. A score of 4 points represents “highly satisfied”, a score of 3 points represents “satisfied”, a score of 2 points suggests “dissatisfied”, and a score of 1 point indicates “highly dissatisfied”. The Cronbach’s α value was 0.88, implying the internal consistency of the original scale, and the Cronbach’s α value for this study was 0.96.

This study was approved by the Institutional Review Board of the medical center (IRB approval number: 100- 2571D) and the administrative departments of the school before the authors began to recruit participants. To protect the rights of the study participants, the authors first explained the study purpose and the data collection procedure to participants in the class meeting. After signing a consent form, participants would receive the study questionnaire. If they felt any discomfort or inappropriateness during the study, participants could freely withdraw from the study at any time. The questionnaires were anonymous, and the collected data were coded for further statistical analysis. To ensure participants’ privacy, this study would not disclose the results of any participants.

SPSS for Windows (Version 17.0; SPSS, Inc., Chicago, IL, USA) was used for data analysis. To describe the distribution of demographic data, sleep quality, and learning satisfaction, we used descriptive statistics expressed in terms of frequency, percentage, mean, and standard deviation. We tested the differences learning satisfaction and demographic data using a t test or a one-way ANOVA and the Scheffe’s method. We tested correlation among sleep quality, and learning satisfaction using Pearson’s product-moment correlation. Stepwise multiple regression was used to identify predictors of learning satisfaction, and significance level p value was set at 0.05.

Most of the study participants studying at the 5-year junior college were in their fifth year of study (52.5%), and a large proportion of the participants did not participate in any extracurricular activities (N = 133, 66.5%). Most of the students did not have a part-time job (83%), and half of the students studied nursing because their parents suggested them to do so. A total of 50% of students reported “neutral” interest in nursing (Table 1). Regarding learning satisfaction, the mean score of overall learning satisfaction among participants was 3.05 ± 0.37 points, suggesting a satisfaction level between “highly satisfied” and “satisfied”. As shown in Table 2, participants demonstrated the highest learning satisfaction toward the cognitive domain, followed by the psychomotor domain, and finally the affective domain. With respect to sleep quality, the mean total score of C-PSQI among the

study participants was 6.17 ± 2.92, representing that more than half of participants (53%) had poor sleep quality. The authors further analyzed the scores of each sub-dimension of sleep quality and found that 35.5% of participants had poor subjective sleep quality. Regarding sleep latency, more than half of participants could fall asleep within 30 min. The mean sleep duration among students was 6.19 ± 0.98 hours/day, and 19.5% of students slept fewer than 5 hours a day. Regarding habitual sleep efficiency, if ≥85% is taken as the standard for good sleep efficiency, 20.5% of students experienced poor sleep efficiency. About 34.5% of the study participants had experienced sleep disturbances more than once. With respect to use of sleep medication, only 4% of the study participants occasionally took sleeping pills, and 32.5% of participants reported that their daytime functions were not affected (Table 2).

“Year of study”, “participation in extracurricular activities”, “holding of a part-time job”, “motivation to study nursing”, and “semester grades” were not found to be significantly related to overall learning satisfaction, while “interest in nursing” was significantly correlated. The authors further conducted a Scheffe’s post hoc comparison and found that compared to students who answered “not interested at all” for “interest in nursing”, those who answered “highly interested”, “interested”, “neutral”, or “not interested” had greater overall learning satisfaction. The study examined the relationships between demographic characteristics and the affective, cognitive, and psychomotor domains. The results show that interest in nursing was significantly associated with the affective and psychomotor domains. An additional Scheffe’s post hoc comparison suggested that compared to students who answered “not interested at all” in terms of their interest in nursing, those who answered “highly interested”, “interested”, “neutral”, or “not interested” had greater learning satisfaction in the affective domain, and those who answered “interested” had greater learning satisfaction in the psychomotor domain (Table 3).

The study results showed no significant relationship between overall learning satisfaction and C-PSQI score. However, a significantly negative relationship was observed between the affective domain of learning satisfaction and C-PSQI score (r = −0.141, p < 0.05), representing that a higher score in the affective domain of learning satisfaction was associated with a lower C-PSQI, which indicated better sleep quality. Regarding the analyses between overall learning satisfaction and different dimensions of sleep quality, overall learning satisfaction was found to be significantly and negatively associated with “subjective sleep quality”, “use of sleep medication”, and “daytime dysfunction”. These findings indicated that lower scores in these three constructs were linked to greater overall learning satisfaction (Table 4).

This study further adopted learning satisfaction as the dependent variable, and incorporated “demographic characteristics” and “sleep quality” as independent variables, which were significantly correlated with overall learning satisfaction. The variable “interest in nursing” was first converted into a dummy variable, and then included in a stepwise multiple regression model for identifying its predictive power on learning satisfaction. The results show that learning satisfaction was significantly correlated with “interest in nursing” (F = 23.59, p < 0.01). That is, 10.2% of the variance in learning satisfaction could be explained by interest in nursing (Table 5).

The mean C-PSQI score among the study participants was 6.17 ± 2.92 points, and 53% of participants had poor sleep quality. This study showed a higher prevalence of poor sleep quality than found by Tsai and Li [9] . The reason may be different department in which participants were general college students. However, the participants recruited in this study were in their fourth and fifth years of study at a junior college nursing school, the internship pressure may affect the sleep. Though the mean C-PSQI score indicated poor sleep quality, according to the scores of subjective sleep quality, approximately 65% (n = 129) of students answered “good” and “very good” for their self-perceived sleep quality. This finding suggests that when paying attention to the individual feelings of subjective sleep is an important factor to solve sleep problems.

The results of this study showed that mean score of overall learning satisfaction level between “highly satis-

^**p < 0.01.

^*p < 0.05, ^**p < 0.01.

^**p < 0.01.

fied” and “satisfied”. This finding is consistent with that of the Wang and Lin [1] study, which investigated perceptions toward the caring behavior of clinical nursing instructors and learning satisfaction among 296 nursing students of a junior college of healthcare management, and found that overall learning satisfaction among the nursing students was between “highly satisfied” and “satisfied”. In this study, students had the highest learning satisfaction in the cognitive domain and the lowest satisfaction in the affective domain. Such may reflect the general nursing students had sufficient professional knowledge, at present, the. Nevertheless, this finding was contradictory with a study by Lofmark and Wikblad [23] , which reported that the general nursing students had insufficient theoretical knowledge. Leigler [24] indicated that the goal of nursing education is to establish the knowledge and skills of students for providing clinical care.

This study found the lowest satisfaction toward the affective domain, which is consistent with the finding of the Feng et al. [25] study, which examined the learning effectiveness of 264 students adopting different admission channels. That study showed that students of the Department of Healthcare Management had fair performance, with a mean score in the affective domain of learning effectiveness of 3.21 points. The literature had suggested that the nurturing of moral character and the establishment of values lie in affective domain learning, which also affects students’ professional spirit and attitudes in the future work [3] [4] . The nursing specialty serves humans, and thus human-centered consideration is a core value of nursing. The cultivation of humanistic literacy helps nurses hold a human-centered philosophy when confronted with the impact of changes in the healthcare structure and values [21] [26] . However, given current management practices, which value technology and cost-efficiency, nurses face several challenges. As a result, in addition to instructing professional knowledge and skills, nursing instructors could also emphasize students’ learning in the affective domain, which includes the nurturing of moral character and the establishment of values to strengthen students’ humanistic literacy.

Additionally, Cronbach’s alpha, based on a set of items’ mean interitem correlation, is the most popular statistic testing internal consistency [21] , the Cronbach’s α value of Learning Satisfaction Scale for this study was 0.96, higher than the original scale [1] . Dennick and Tavakol [27] indicated that reliability is the consistency to measure the degree of error, respondents, situational factors can be caused by the error. Participants in the original scale [1] , including the different school systems and fill out the questionnaire on end of their internships, however, this study was not. Perhaps resulting in Cronbach’s α of the two study differences. A high value of alpha (>0.90) may suggest redundancies [21] . Therefore, future research should pay attention to the appropriateness of the research tools and the test length should be shortened.

A significant relationship was found between affective domain of learning satisfaction and C-PSQI score, means that students with better sleep quality had a higher score in the affective domain of learning satisfaction, a finding similar to Feng et al. [25] . Such may reflect that nursing education is highly professional, and nursing practices serve human individuals, abilities such as interpersonal interaction and communication and crisis management are essential to nurses. As a result, teaching in the affective domain, which emphasizes the nurturing of moral character and values, is necessary in nursing education. The authors therefore recommend that nursing students receive sleep health education to enhance learning effectiveness in the affective domain, which establishes students’ attitudes and values.

The results of this study showed that lower scores of subjective sleep quality, use of sleep medication, or daytime dysfunction brought higher overall learning satisfaction, a finding consistent with prior studies [15] [28] that demonstrated that subjective sleep quality is a key predicting factor of sleep quality, and long-term sleep disorder not only affects physical and mental conditions, but indirectly influences learning efficiency.

There was a significant difference between interest in nursing (demographic characteristics) and learning satisfaction in this study. Students who were more interested in nursing had greater learning satisfaction in the affective and psychomotor domains. This finding was consistent with Wang and Lin [1] found that students possessing higher interest in nursing practices had higher learning satisfaction.

The major predictor of learning satisfaction was “interest in nursing”, accounting for 10.2% of learning satisfaction variance in this study. This finding indicates that individual with more concerned about their personal interests, sense of achievement, self-development, social interaction, and health and leisure activities, and therefore value their jobs in terms of its achievement, development, and material rewards [29] . Therefore, under these circumstances, cultivating the interest of nursing students toward nursing practices not only enhances their learning satisfaction, but benefits their career planning.