Skip to main content
NCBI home page
Nursing Open logoLink to Nursing Open
. 2025 Jan 10;12(1):e70137. doi: 10.1002/nop2.70137

Influencing Factors of Social Anxiety of Undergraduate Nursing Students Based on Random Forest Model: A Cross‐Sectional Study

Yuling Jia 1, Yuexue Yue 1,✉
PMCID: PMC11723782  PMID: 39794003

ABSTRACT

Aim

Social anxiety is common among students, yet research specifically on undergraduate nursing students is scarce. Previous studies have identified various influencing factors but lack a clear ranking of their importance, making it challenging to implement effective intervention strategies with limited resources. This study aims to assess the level of social anxiety in undergraduate nursing students and rank the importance of its influencing factors, providing essential insights for developing efficient and targeted interventions.

Design

A cross‐sectional study was conducted with 546 undergraduate nursing students in China.

Methods

Data were collected via a web‐based questionnaire, which included the fear of positive and negative evaluation scale, the self‐efficacy scale, and the social anxiety scale. The random forest model was employed to analyse the data and rank the importance of the influencing factors.

Results

The total social anxiety score among undergraduate nursing students was 11.20 ± 4.71. Social anxiety had a positive correlation with both fear of positive evaluation (r = 0.521, p < 0.001) and fear of negative evaluation (r = 0.648, p < 0.001) and a negative correlation with self‐efficacy (r = −0.273, p < 0.001). According to the random forest model, the top five predictors of social anxiety in this study were fear of negative evaluation, fear of positive evaluation, personality traits, self‐efficacy, and academic performance.

Conclusion

Attention to social anxiety in undergraduate nursing students is necessary, particularly for those with a higher fear of evaluations, introverted personalities, lower self‐efficacy, and poor academic performance. Nursing educators should consider recognising social anxiety and developing personalised interventions tailored to the specific characteristics of these students.

Patient or Public Contribution

All participants contributed to the conducting of this study by completing self‐reported questionnaires.

Keywords: influencing factors, random forest model, social anxiety, undergraduate nursing students

1. Introduction

Social anxiety refers to a situation in which individuals have some unreasonable, excessive fear of interpersonal interactions and their performance in certain social conditions. It is accompanied by individuals' intense emotional experience (i.e., nervousness or distress) and a tendency to avoid social interactions (Regier, Kuhl, and Kupfer 2013). Several studies have focused on college students' social anxiety (Kraft et al. 2021; Lyvers, Hanigan, and Thorberg 2018; Meng et al. 2021; Yu et al. 2020), demonstrating that social anxiety is linked to decreased subjective well‐being (Spinhoven et al. 2021) and poorer quality of interpersonal relationships (Azad‐Marzabadi and Amiri 2017). Moreover, social anxiety contributes to various negative emotional experiences, including low self‐esteem, frustration, fear, and loneliness (Oren‐Yagoda, Melamud‐Ganani, and Aderka 2021). Social anxiety has also been associated with more severe consequences, such as addictive behaviours, including mobile phone and internet addiction (Annoni et al. 2021), aggressive behaviours, depression (Ye et al. 2017), and potentially suicidal tendencies (Chung et al. 2022).

However, most existing studies have focused on general college students rather than nursing students (Manning et al. 2017). Nursing is a profession that places a high value on social interaction and relationships in clinical settings, where close communication and collaboration are essential among nurses, patients, caregivers, physicians and medical technicians from various disciplines (Lau and Wang 2013). Social anxiety in nursing students is particularly concerning because it may impede their ability to perform essential clinical duties, establish effective patient relationships, and function as part of a healthcare team. Despite this, there are few international studies on nursing students' social anxiety (Turan et al. 2019; Jun and Lee 2017). Previous studies in China have addressed this issue (Ye et al. 2017). However, they primarily focused on nursing students in higher vocational colleges and nursing interns (Xie, Gao, and Yang 2021; Shan et al. 2018). There remains a significant gap in the literature regarding research specifically targeting undergraduate nursing students (Ye et al. 2017).

Given the global development of higher nursing education, undergraduate nursing students are increasingly becoming the backbone of the future nursing workforce (Su et al. 2018). Thus, it is critical to investigate social anxiety within this group, as it may directly influence their educational experiences and future professional performance. Identifying the factors contributing to social anxiety among undergraduate nursing students is essential for developing targeted interventions to reduce its prevalence, thereby ensuring the cultivation of high‐quality nursing professionals (Perez‐Martin et al. 2022). Prior research has shown that social anxiety is prevalent among nursing students and is influenced by various personal, familial, and societal factors (Xu and Yang 2019). However, the specific contributing factors and their relative impact on social anxiety remain a topic of debate. For instance, a study by Shan et al. (2018) identified factors such as interpersonal relationships, leadership roles within the class, and academic performance as contributors to social anxiety. Nevertheless, the relationship between being an only child and social anxiety is still uncertain. Similarly, Ding et al. (2008) reported that 60.41% of undergraduate nursing students experienced social anxiety, particularly those who were the only child and from rural areas, but the relationship between student leadership and social anxiety was inconclusive.

Moreover, previous studies, such as those by Shan et al. (2018) and Ding et al. (2008), predominantly relied on traditional statistical techniques like chi‐squared or multiple regression analysis. While these methods identified several associated variables, they were limited in pinpointing specific factors that require further investigation, thereby restricting the development of potential intervention strategies. In contrast, the random forest model, a machine learning algorithm, offers a robust alternative. It can process large‐scale, complex and irregular data, allowing researchers to rank influencing factors based on their contribution to outcome variables (Couronne, Probst, and Boulesteix 2018; Raita et al. 2019). Unlike classical statistical models, random forest is not constrained by assumptions of independence or normality (Song et al. 2015) and has proven effective in ranking the importance of influencing factors (Li et al. 2021; Liu et al. 2021). Therefore, this study aims to assess the current status and associated factors of social anxiety among undergraduate nursing students through a cross‐sectional survey and to prioritise these factors according to their significance. By doing so, we aim to enhance nursing educators' awareness and understanding of social anxiety, ultimately informing the development of more effective educational interventions.

2. Methods

2.1. Study Setting and Design

A cross‐sectional study was conducted in October and November 2021 within the nursing program at a medical university in Southwest China. The nursing program spans 4 years of study and includes two majors: Nursing and Midwifery. Over the past 3 years, the program has enrolled more than 500 undergraduate nursing students annually.

2.2. Participants and Sampling

Participants were selected using a convenience sampling method. The inclusion criteria were: (1) registered full‐time undergraduate students in the Nursing and Midwifery programs during the survey period; (2) students in their first, second, or third year of study. The exclusion criteria were: (1) diagnosed with depression during the survey period (to eliminate potential confounding variables, as depression and social anxiety often co‐occur and may influence each other); (2) physically disabled; (3) absent from school on leave during the survey period.

According to M. Kendall's sample content estimation method (Fang 2007), the sample size was 5–10 times the number of variables or predictors in our analysis. In the present study, with a maximum of 12 predictor variables, the recommended sample size should fall within the range of 75–150, with an additional 20% allowance for invalid questionnaires. Based on the specific research circumstances, the final sample size for this study was 546, meeting the necessary criteria for research validity.

2.3. Ethical Considerations

Ethical considerations were addressed in this study by obtaining approval from the Guizhou Medical University Institutional Review Board (Approval Number: 2024‐10). Informed consent was obtained from all participants by placing the informed consent form at the outset of the online scale, ensuring that the survey would commence only after participants confirmed their informed consent.

2.4. Measurements and Data Collection

2.4.1. Measurements

2.4.1.1. The General Information Questionnaire

The general information questionnaire was designed according to the purpose of the study and the literature review. It included gender, place of birth, the only child, left‐behind experience, student leaders, practice experience, personality, academic performance, promotion‐seeking, etc. Among them, the only child was a phenomenon with only one child in the family, arising under China's family planning policy (Feng and Wang 2016). Left‐behind experience refers to the situation where children or adolescents under the age of 16 are left alone in their hometown or entrusted to the care of others (such as grandparents or other relatives) for more than 6 months. This occurs when one or both of their parents have left their place of origin to earn a living (Liu, Zhou et al. 2021). Student leaders were students who held leadership positions in student councils, classes, or societies. Practice experience refers to students in student societies or working part‐time for more than 1 month. The following rules grouped academic performance: academic achievement in the top 25% of the class was recorded as upstream, in the 25%–75% of the class was midstream, and in the bottom 25% of the class was downstream. Promotion‐seeking was the situation in which students plan to pursue a master's degree in nursing after their undergraduate studies.

2.4.1.2. The Brief Fear of Negative Evaluation Scale‐Straightforward Items (BFNES‐S)

The BFNES‐S was developed and subsequently refined by Rodebaugh et al. (2004) and Weeks et al. (2005) and has been widely used in evaluating the fear of negative. It retained only eight positive scoring questions and had better predictive validity than the original version. The scale used a 5‐point scale (0–4); its scores ranged from 0 to 32. The higher the score, the more fearful of negative evaluations. This scale has demonstrated good content, structural and discriminant validity. It has been widely adopted in analysing personality traits among Chinese students, as evidenced by a robust Cronbach's α coefficient of 0.856 (Wu et al. 2021). In this study, we obtained an even higher Cronbach's α coefficient of 0.926, indicating a high level of internal consistency.

2.4.1.3. Fear of Positive Evaluation Scale (FPES)

FPES, developed by Weeks et al. (2008), comprises 10 items, including two reverse‐scored items (which are not included in the total score), and is rated on a 10‐point scale ranging from 0 to 9, with total scores varying between 0 and 72. Higher scores reflect more significant levels of fear of positive evaluation. The scale exhibits good content, structural, and discriminant validity, has been culturally adapted, and is widely used in China (Wu et al. 2021). In this study, the Cronbach's α coefficient reached 0.866, indicating high internal consistency.

2.4.1.4. General Self‐Efficacy Scale (GSES)

GSES was a one‐dimensional scale developed by (Luszczynska et al. 2005). After several revisions, the scale consisted of 10 items and was scored on a 4‐point scale from ‘not at all true’ to ‘completely true’ on a scale of 1–4. The total scores ranged from 1 to 4. Higher scores indicated higher levels of self‐efficacy. The scale exhibits good reliability and validity (Wang, Hu, and Liu 2001) and is widely recognised as a standard instrument for evaluating self‐efficacy. The Cronbach's α coefficient in this study was 0.908, indicating acceptable reliability.

2.4.1.5. Social Anxiety Subscale of Self‐Consciousness Scale (SASS‐CS)

Developed by Fenigstein and revised by Scheier and Buss (1975), SASS‐CS was a six‐item measure of subjective anxiety and verbal and behavioural difficulties experienced by individuals when socialising. Responses were scored on a five‐point scale (from 0, extremely incompatible, to 4, extremely compatible). Question 4 was reverse‐scored. Its total scores range from 0 to 24, and a higher score indicates a higher tendency to experience subjective social anxiety. This scale demonstrates satisfactory content and structural validity, as well as acceptable reliability (Wang, Wang, and Ma 1999). In this study, the Cronbach's α was 0.834.

2.4.2. Data Collection Methods

The survey questionnaire and instructions were imported into the Questionnaire Star platform (https://www.wjx.cn/). The researcher pre‐tested the survey to improve how the items were presented, identify any errors or contradictions, and record the time taken to complete it. After making the final modifications, we contacted five tutors for the nursing classes. We requested their assistance in posting the survey instructions, the QR code, and the link to the questionnaire in their class QQ groups. This way, we could invite students to complete the online questionnaire after class. Nursing students participated in the survey by voluntarily clicking on the link or scanning the QR code, and an informed consent form was set up on the first page of the questionnaire. A logic check was established in the general information section of the questionnaire, ensuring that if an answer is incorrect or irregular, a prompt automatically appears to remind the respondent to verify their response. All entries in the questionnaire were compulsory, and the creator set the questionnaire to be completed only once by one IP. After the questionnaires were collected, they were checked and screened by two researchers to eliminate questionnaires that took less than 2 min to complete, had confusing logic, and had a high rate of overlapping answers to ensure the quality of the questionnaires. Five hundred sixty‐eight questionnaires were collected. Twenty‐two invalid questionnaires were excluded, resulting in 546 valid questionnaires, yielding a reasonable return rate of 96.13%.

2.5. Statistical Methods

SPSS 24.0 software was used for statistical analysis. The test levels were all two‐sided probabilities, α = 0.05, and p < 0.05 indicated statistical significance. Quantitative variables, including fear of negative evaluation, fear of positive evaluation, and self‐efficacy, were described using M ± SD. Qualitative variables, including general information such as gender, the only child, etc., were described using frequency (n) and percentage (%). One‐way analysis of variance was performed using independent samples t‐test, ANOVA, chi‐squared test, or Fisher's exact probability method. LSD post hoc tests were used for multiple comparisons of academic performance and personality. Pearson or Spearman correlation analysis was used to correlate social anxiety, general self‐efficacy, and fear of positive and negative evaluation. The study was modelled using the ‘random Forest’ package of R 4.1.2 software to rank the importance of the 12 variables (the only child, gender, place of birth, practice experience, promotion‐seeking, academic performance, personality, student leader, left‐behind experience, fear of negative evaluation, fear of positive evaluation, self‐efficacy) initially included in the study. Variable importance scores were used to evaluate the impact of variables on the occurrence of outcomes. In this research, the percentage increase in mean square error (%IncMSE) was employed to assess the significance of variables. This methodology entails the random allocation of values to each predictor variable, with the model's prediction error increasing when a more critical predictor variable has its value substituted randomly. A higher %IncMSE value indicates greater importance of the variable in question. We also set the seed number and outputted the result via the importance () and varImpPlot () functions.

3. Results

3.1. Social‐Demographic Characteristics of Participants

A total of 546 undergraduate nursing students completed valid questionnaires, of which 106 (19.4%) were male, and 440 (80.6%) were female; 10.6% of students were the only child; 81.0% of students came from rural areas; 50.9% of students had left‐behind experience.

Nursing students' social anxiety scores were statistically different (p < 0.05) by grade, personality, academic performance and promotion‐seeking. By post hoc LSD test, the score of social anxiety of students with ‘downstream’ academic performance was significantly higher than that of midstream and upstream; the score of social anxiety of introverted students was considerably higher than that of extroverted and neutral, as shown in Table 1.

TABLE 1.

Social‐demographic characteristics of participants (N = 546).

Variables Frequency (N, %) Social anxiety (M ± SD) Statistics (t/F) p
Gender −0.739 a 0.460
Male 106 (19.4%) 10.90 ± 4.95
Female 440 (80.6%) 11.27 ± 4.65
The only child 0.307 a 0.759
Yes 58 (10.6%) 11.38 ± 4.34
No 488 (89.4%) 11.18 ± 4.75
Place of birth 0.179 b 0.836
Rural 442 (81.0%) 11.23 ± 4.81
Urban 29 (5.3%) 10.69 ± 3.98
Towns 75 (13.7%) 11.24 ± 4.39
Left‐behind experience 1.465 a 0.143
Yes 278 (50.9%) 11.49 ± 4.62
No 268 (49.1%) 10.90 ± 4.79
Student leaders −1.804 a 0.072
Yes 94 (17.2%) 10.40 ± 4.80
No 452 (82.8%) 11.37 ± 4.68
Practice experience −1.548 a 0.122
Yes 375 (68.7%) 10.99 ± 4.67
No 171 (31.3%) 11.66 ± 4.77
Personality 47.503 b < 0.001
Introverted 121 (22.2%) 14.23 ± 4.16
Extrovert 69 (12.6%) 8.22 ± 4.39
Neutral 356 (65.2%) 10.75 ± 4.41
Academic performance 5.542 b 0.004
Upstream 114 (20.9%) 10.04 ± 4.96
Midstream 350 (64.1%) 11.35 ± 4.59
Downstream 82 (15.0%) 12.18 ± 4.60
Promotion‐seeking −2.840 a 0.005
Yes 395 (72.3%) 10.85 ± 4.69
No 151 (27.7%) 12.12 ± 4.64
Open in a new tab

Note: Bold values represent p‐values < 0.05.

Post hoc LSD test: Academic performance: midstream = downstream>upstream; personality: introverted>extroverted; introverted > neutral; neutral > extroverted. The only child: the only child in the family, no siblings, a phenomenon that has arisen under China's family planning policy. Left‐behind experience: children or adolescents under the age of 16 who are left alone in their hometown or entrusted to others relatives for more than 6 months. Because one or both parents have left their place of origin to earn their living. Student leaders: students hold leadership positions in student councils or classes or clubs. Academic performance: achievement in the top 25% is recorded as upstream, in the 25%–75% as midstream, and the bottom 25% as downstream. Promotion‐seeking: students plan to pursue a master's degree in nursing after their undergraduate studies.

a

t‐test for two independent samples.

b

ANOVA.

3.2. Scores and Correlation Analysis of Interesting Variables

In this study, the social anxiety score among undergraduate nursing students was found to be 11.20 ± 4.71. The average score for fear of negative evaluation was 15.90 ± 7.18, while the score for fear of positive evaluation was 26.02 ± 13.10. The self‐efficacy score was measured at 2.38 ± 0.53.

The results indicated a positive correlation between social anxiety and both types of evaluation fear: fear of positive evaluation (r = 0.521, p < 0.001) and fear of negative evaluation (r = 0.648, p < 0.001). Additionally, social anxiety was found to be negatively associated with self‐efficacy (r = −0.273, p < 0.001). Fear of negative evaluation also showed a positive correlation with fear of positive evaluation (r = 0.479, p < 0.001) and a negative correlation with self‐efficacy (r = −0.188, p < 0.001). Similarly, fear of positive evaluation was negatively correlated with self‐efficacy (r = −0.197, p < 0.001), as illustrated in Table 2.

TABLE 2.

Correlation between social anxiety, self‐efficacy, and fear of positive and negative evaluation.

Variables M ± SD FNE FPE Self‐efficacy Social anxiety
FNE 15.90 ± 7.18 1
FPE 26.02 ± 13.10 0.479 a 1
Self‐efficacy 2.38 ± 0.53 −0.188 a −0.197 a 1
Social anxiety 11.20 ± 4.71 0.648 a 0.521 a −0.273 a 1
Open in a new tab

Abbreviations: FNE, fear of negative evaluation; FPE, fear of positive evaluation.

a

Correlation significant at 0.01 level (two‐tailed).

3.3. Importance Ratings and Ranking of Variables Affecting Social Anxiety Among Undergraduate Nursing Students

A random forest model was constructed using social anxiety as the dependent variable and 12 characteristics as the independent variables, including gender, the only child, etc. The results showed that variables in descending order of importance were fear of negative evaluation (%IncMSE = 67.121), fear of positive evaluation (%IncMSE = 31.942), personality (%IncMSE = 23.596), self‐efficacy (%IncMSE = 12.049), academic performance (%IncMSE = 6.882), student leader (%IncMSE = 1.599), left‐behind experience (%IncMSE = 1.884), the only child (%IncMSE = 1.196), gender (%IncMSE = 1.011), practice experience (%IncMSE = 0.187), and promotion‐seeking (%IncMSE = −0.726), as shown in Table 3 and Figure 1.

TABLE 3.

Coding and importance score of independent variables (%IncMSE, %).

Independent variable Coding Social anxiety
Gender Gender 1.011
The only child The only child 1.196
Place of birth Place of birth 0.770
Left‐behind experience Left‐behind experience 1.884
Student leaders Student leaders 1.599
Practice experience Practice experience 0.187
Personality Personality 24.596
Academic performance Academic performance 6.882
Promotion‐seeking Promotion‐seeking −0.726
Fear of positive evaluation FPE 31.942
Fear of negative evaluation FNE 67.121
Self‐efficacy Self‐efficacy 12.049
Open in a new tab

FIGURE 1.

FIGURE 1

Open in a new tab

Ranking of importance of factors associated with social anxiety. (FNE, fear of negative evaluation; FPE, fear of positive evaluation).

4. Discussion

In this cross‐sectional study, we surveyed undergraduate nursing students to identify factors associated with social anxiety. We used a random forest algorithm to assess the relative importance of various influencing factors. Our findings revealed that the top five factors affecting social anxiety among undergraduate nursing students were fear of negative evaluation, fear of positive evaluation, personality traits, self‐efficacy and academic achievement. These results highlight critical areas where nursing educators can focus on developing targeted interventions and providing practical support to help reduce social anxiety in undergraduate nursing students.

4.1. Social Anxiety Among Undergraduate Nursing Students Is Not Optimistic

In this study, undergraduate nursing students exhibited lower levels of social anxiety compared to general university students (Xu, Liu, and Liu 2024), yet higher levels than college students facing financial difficulties (Zeng et al. 2017). This disparity may stem from the use of different assessment tools for social anxiety. Xu, Liu, and Liu (2024) employed the Interaction Anxiousness Scale (IAS), while both Zeng et al. (2017) and the current study utilised the Social Anxiety Subscale of the Self‐Consciousness Scale (SASS), which has fewer items and a lower total score. A systematic review involving 4983 Chinese nursing students indicated that when measured by the IAS, the prevalence of social anxiety in this group was significantly higher than in general college students (Ye et al. 2017). Although this review primarily focused on students from vocational colleges and interns, it still aligns with our findings, emphasising that social anxiety among nursing students is a pressing concern. One potential reason for this elevated level of social anxiety could be the predominance of females in nursing programs. Our previous research has shown that females are more likely to experience a greater fear of negative evaluation (Yue and Jia 2022), which may contribute to the heightened levels of social anxiety observed in this population. However, most existing studies are cross‐sectional and do not delve into the qualitative experiences of nursing students regarding social anxiety. Thus, it is essential to conduct longitudinal studies and utilise qualitative interviews, along with other methods, to thoroughly examine the long‐term effects and implications of social anxiety on undergraduate nursing students.

4.2. Factors Influencing Social Anxiety Among Undergraduate Nursing Students

Our study found that both fears of positive and negative evaluation are positively related to social anxiety, consistent with previous research (Johnson et al. 2020). Additionally, the random forest analysis revealed that fear of negative evaluation was the primary influence, followed closely by fear of positive evaluation on social anxiety, as supported by Johnson et al. (2020). Fear of evaluation is a core characteristic of social anxiety (Rose and Tadi 2022). When individuals anticipate the possibility of negative evaluation in social settings, they often experience heightened anxiety. The intensity of this social anxiety is positively correlated with the level of fear of negative evaluation (Villarosa‐Hurlocker et al. 2018). Furthermore, individuals with a strong fear of negative evaluation tend to be more sensitive to ambiguous stimuli, resulting to elevated stress responses when faced with uncertainty. This sensitivity may predispose them to mood disorders such as anxiety (Peng et al. 2019).

Similarly, fear of positive evaluation has also been identified as a significant contributor to the development and maintenance of social anxiety. It affects social anxiety both directly and indirectly by influencing fear of negative evaluation and has been increasingly highlighted as a target for interventions (Fredrick and Luebbe 2020; Reichenberger et al. 2019). Research has shown that socially anxious individuals who fear positive evaluation often engage in maladaptive behaviours and cognitive patterns, such as denying positive thoughts, avoiding attention, and processing emotions negatively. This fear reinforces cognitive biases and inhibits positive emotions through a process known as de‐positivization, thereby perpetuating and intensifying symptoms of social anxiety (Ye, Lin, and Yan 2021).

Given the crucial role of evaluating fear in social anxiety, nursing educators should carefully consider their teaching and management evaluation strategies. While encouragement and criticism are well‐established pedagogical tools supported by most literature on positive reinforcement, it is crucial to recognise that individual differences in students' evaluation fears may affect their response to these methods. Although nursing educators have demanding schedules, they might find it feasible to incorporate simple methods like surveys or brief assessments to gain insights into students' evaluation fears, potentially providing more targeted and effective feedback.

In this study, differences in social anxiety scores among nursing students are associated with differing personality types, aligning with the findings of Shan et al. (2018). According to Lakuta (2019), introverted individuals are more likely to developing negative psychological and behavioural responses, such as fear and avoidance in interpersonal interactions, which contribute to higher anxiety levels. From the perspective of Chinese medicine, human personality can be divided into five states: Tai Yang, Shao Yang, Yin Yang Ping He, Shao Yin, and Tai Yin. Those characterised by Tai Yang and Shao Yang tend to be extroverted, active, and sociable. Individuals with a Yin Yang Ping He personality are typically calm, emotionally stable, and exhibit strong self‐control and adaptability, making them less likely to experience social anxiety. In contrast, individuals identified as Tai Yin and Shao Yin are introverted, shy, timid, repressed, self‐denying, easily stressed, sensitive, and prone to withdrawal, which increases their susceptibility to social anxiety (Liu and Yang 2015). Collectively, these studies elucidate the differences in social anxiety behaviours among individuals with varying personality traits. This may be attributed to the fact that individuals with distinct personality characteristics employ different emotional strategies to manage or cope with negative events or experiences (Cao and Qi 2021).

This study also found that lower self‐efficacy is associated with higher social anxiety, consistent with the findings of Lee and Yeghiazarian (2021). Self‐efficacy influences an individual's stress response, anxiety and depression levels, and other physical and mental reactions when facing stressors (Molero et al. 2019). Individuals with high self‐efficacy are likely to adopt positive coping strategies during social interactions, making them more inclined to approach others and engage with their environment, thus experiencing less social anxiety (Bandura et al. 1988). Furthermore, past experiences of failure, such as rejection in social activities, can lower self‐efficacy, leading to negative emotional experiences like anxiety and loneliness (Bandura et al. 1988; Jia et al. 2019).

In this study, we also found that students with poor academic performance exhibited higher levels of social anxiety (p < 0.05), consistent with previous studies (Al‐Hazmi, Sabur, and Al‐Hazmi 2020; Vilaplana‐Perez et al. 2021). Underachieving students may have lower self‐esteem than their high‐achieving peers. They might avoid interactions due to fear of rejection (He 2022), contributing to higher social anxiety levels. Therefore, nursing educators should pay particular attention to these students to promptly identify and address their social anxiety.

Additionally, some studies have shown that factors such as being a student leader, left‐behind experience, gender, being an only child, practice experience, and promotion‐seeking might also be associated with students' social anxiety to varying degrees (Hou et al. 2022; Ye et al. 2017). However, the random forest model indicated that the explanatory power of these factors for social anxiety was relatively low, which needs further verification in subsequent studies.

4.3. Limitations

However, the limitations of this study still need to be considered. Firstly, using a convenience sampling method to include undergraduate nursing students from only one medical school limits the generalizability of our findings. Future research should consider including students from different grades or schools to analyse potential differences in social anxiety. Additionally, this study's cross‐sectional design precludes any implication of causality between variables. Therefore, cohort or longitudinal studies are needed to explore the causal relationships between influencing factors and social anxiety, as well as trends in social anxiety among nursing students over time. Secondly, while the influencing factors were selected based on a literature review, their scope was limited, potentially omitting important psychological or contextual factors. Therefore, the results should be interpreted cautiously. Future research should utilise robust theoretical frameworks to guide investigations and adopt mixed methods to explore the diverse influences on social anxiety comprehensively. As new influencing factors are identified, the results of this study should be interpreted with caution. Finally, while the random forest algorithm was used to rank the importance of the influencing factors, it does not indicate potential interactions, intermediate relationships, or specific action pathways between the variables. Although these were not the focus of this study, future research could address them through path analysis and structural equation modelling (SEM).

5. Conclusion

Using a random forest model, this cross‐sectional study identified factors associated with social anxiety among 546 undergraduate nursing students. Key factors included fear of negative and positive evaluation, personality, self‐efficacy, and academic performance. These findings indicate that it may be beneficial for nursing educators to regularly screen for social anxiety and consider implementing supportive measures such as modifying evaluation strategies, establishing support groups, and providing psychological counselling and stress management training. Future research should explore the effectiveness of these interventions and investigate additional factors influencing social anxiety in nursing students.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgements

The authors thank all tutors and undergraduate nursing students who participated in the survey.

Funding: This work was supported by Guizhou Medical University Nursing Discipline Special Project, YJ22002.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Al‐Hazmi, B. H. , Sabur S. S., and Al‐Hazmi R. H.. 2020. “Social Anxiety Disorder in Medical Students at Taibah University, Saudi Arabia.” Journal of Family Medicine and Primary Care 9, no. 8: 4329–4332. 10.4103/jfmpc.jfmpc_915_20 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Annoni, A. M. , Petrocchi S., Camerini A. L., and Marciano L.. 2021. “The Relationship Between Social Anxiety, Smartphone Use, Dispositional Trust, and Problematic Smartphone Use: A Moderated Mediation Model.” International Journal of Environmental Research and Public Health 18, no. 5: 2452. 10.3390/ijerph18052452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Azad‐Marzabadi, E. , and Amiri S.. 2017. “Morningness‐Eveningness and Emotion Dysregulation Incremental Validity in Predicting Social Anxiety Dimensions.” International Journal of General Medicine 10: 275–279. 10.2147/IJGM.S144376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bandura, A. , Cioffi D., Taylor C. B., and Brouillard M. E.. 1988. “Perceived Self‐Efficacy in Coping With Cognitive Stressors and Opioid Activation.” Journal of Personality and Social Psychology 55, no. 3: 479–488. 10.1037//0022-3514.55.3.479. [DOI] [PubMed] [Google Scholar]
  5. Cao, C. X. , and Qi S. S.. 2021. “The Relationship Between Potential Categories of Emotion Regulation Self‐Efficacy and Social Anxiety in College Students.” Chinese Journal of Behavioral Medicine and Brain Science 30, no. 11: 1040–1044. 10.3760/cma.j.cn371468-20210712-00393. [DOI] [Google Scholar]
  6. Chung, M. L. , Forstner A. J., Mucke M., Geiser F., Schumacher J., and Conrad R.. 2022. “Predictors of Suicidal Ideation in Social Anxiety Disorder—Evidence for the Validity of the Interpersonal Theory of Suicide.” Journal of Affective Disorders 298: 400–407. 10.1016/j.jad.2021.11.017. [DOI] [PubMed] [Google Scholar]
  7. Couronne, R. , Probst P., and Boulesteix A. L.. 2018. “Random Forest Versus Logistic Regression: A Large‐Scale Benchmark Experiment.” BMC Bioinformatics 19, no. 1: 270. 10.1186/s12859-018-2264-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ding, T. T. , Tian Y. C., Zhang J. Y., Lei X. X., and Ma C. Y.. 2008. “A Survey on the Current Status of Social Anxiety Among Undergraduate Nursing Students.” Nursing Research 20: 1799–1800. [Google Scholar]
  9. Fang, J. Q. 2007. Statistical Methods for Biomedical Research (1st ed.). Beijing, China: Higher Education Press. [Google Scholar]
  10. Feng, X. T. , and Wang X. T.. 2016. “From One‐Child to Post‐One‐Child Families—The "Comprehensive Two‐Child" Policy and Changes in Chinese Family Patterns.” Chinese Youth Social Science 35, no. 2: 47–53. [Google Scholar]
  11. Fenigstein, A. , Scheier M. F., and Buss A. H.. 1975. “Public and private self‐consciousness: Assessment and theory.” Journal of Consulting and Clinical Psychology 43, no. 4: 522–527. 10.1037/h0076760. [DOI] [Google Scholar]
  12. Fredrick, J. W. , and Luebbe A. M.. 2020. “Fear of Positive Evaluation and Social Anxiety: A Systematic Review of Trait‐Based Findings.” Journal of Affective Disorders 265: 157–168. 10.1016/j.jad.2020.01.042. [DOI] [PubMed] [Google Scholar]
  13. He, X. 2022. “Relationship Between Self‐Esteem, Interpersonal Trust, and Social Anxiety of College Students.” Occupational Therapy International 2022: 8088754. 10.1155/2022/8088754. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  14. Hou, Y. T. , Zhu H. D., Shi Y. H., and Jia X. S.. 2022. “Potential Profiles and Influencing Factors of Social Anxiety Among College Students.” Chinese Journal of Health Psychology 30, no. 9: 1406–1412. [Google Scholar]
  15. Jia, Y. R. , Zhang S. C., Jin T. L., Zhang L., Zhao S. Q., and Li Q.. 2019. “The Effect of Social Exclusion on Social Anxiety Among College Students: The Role of Negative Evaluation Fear and Interpersonal Trust.” Psychological Science 42, no. 3: 653–659. 10.16719/j.cnki.1671-6981.20190321. [DOI] [Google Scholar]
  16. Johnson, A. R. , Bank S. R., Summers M., et al. 2020. “A Longitudinal Assessment of the Bivalent Fear of Evaluation Model With Social Interaction Anxiety in Social Anxiety Disorder.” Depression and Anxiety 37, no. 12: 1253–1260. 10.1002/da.23099. [DOI] [PubMed] [Google Scholar]
  17. Jun, W. H. , and Lee G.. 2017. “The Role of Ego‐Resiliency in the Relationship Between Social Anxiety and Problem Solving Ability Among South Korean Nursing Students.” Nurse Education Today 49: 17–21. 10.1016/j.nedt.2016.11.006. [DOI] [PubMed] [Google Scholar]
  18. Kraft, J. D. , Grant D. M., White E. J., Taylor D. L., and Frosio K. E.. 2021. “Cognitive Mechanisms Influence the Relationship Between Social Anxiety and Depression Among College Students.” Journal of American College Health 69, no. 3: 245–251. 10.1080/07448481.2019.1661844. [DOI] [PubMed] [Google Scholar]
  19. Lakuta, P. 2019. “Personality Trait Interactions in Risk for and Protection Against Social Anxiety Symptoms.” Journal of Psychology 153, no. 6: 599–614. 10.1080/00223980.2019.1581723. [DOI] [PubMed] [Google Scholar]
  20. Lau, Y. , and Wang W.. 2013. “Development and Evaluation of a Learner‐Centered Training Course on Communication Skills for Baccalaureate Nursing Students.” Nurse Education Today 33, no. 12: 1617–1623. 10.1016/j.nedt.2013.02.005. [DOI] [PubMed] [Google Scholar]
  21. Lee, C. S. , and Yeghiazarian C.. 2021. “Personal Value, Self‐Efficacy, and Social Acceptability of Social Behavior as Correlates of Behavioral Action in Social Anxiety.” Trends in Psychiatry and Psychotherapy 43, no. 3: 217–224. 10.47626/2237-6089-2020-0129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li, C. , Liao C., Meng X., et al. 2021. “Effective Analysis of Inpatient Satisfaction: The Random Forest Algorithm.” Patient Preference and Adherence 15: 691–703. 10.2147/PPA.S294402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liu, H. , Zhou Z., Fan X., et al. 2021. “A Mixed Method Study to Examine the Mental Health Problems of College Students Who Had Left‐Behind Experiences.” Journal of Affective Disorders 292: 149–160. 10.1016/j.jad.2021.04.087. [DOI] [PubMed] [Google Scholar]
  24. Liu, X. C. , Gu L., Chen T., Sun Z. K. J. M., and Zhang R.. 2021. “Analysis of the Current Situation and Influencing Factors of Ruminative Rumination in Patients With Gastrointestinal Tumors Based on Random Forest Model.” Modern Preventive Medicine 48, no. 7: 1310–1314. [Google Scholar]
  25. Liu, Z. L. , and Yang Z. N.. 2015. “A Study on the Correlation Between Social Anxiety and Pentadic Personality in First‐Year College Students.” Chinese Journal of Basic Chinese Medicine 21, no. 3: 308–309. [Google Scholar]
  26. Luszczynska, A. , Scholz U., and Schwarzer R.. 2005. “The General Self‐Efficacy Scale: Multicultural Validation Studies.” Journal of Psychology 139, no. 5: 439–457. 10.3200/JRLP.139.5.439-457. [DOI] [PubMed] [Google Scholar]
  27. Lyvers, M. , Hanigan C., and Thorberg F. A.. 2018. “Social Interaction Anxiety, Alexithymia, and Drinking Motives in Australian University Students.” Journal of Psychoactive Drugs 50, no. 5: 402–410. 10.1080/02791072.2018.1517228. [DOI] [PubMed] [Google Scholar]
  28. Manning, R. P. , Dickson J. M., Palmier‐Claus J., Cunliffe A., and Taylor P. J.. 2017. “A Systematic Review of Adult Attachment and Social Anxiety.” Journal of Affective Disorders 211: 44–59. 10.1016/j.jad.2016.12.020. [DOI] [PubMed] [Google Scholar]
  29. Meng, T. , He Y., Zhang Q., et al. 2021. “Analysis of Features of Social Anxiety and Exploring the Relationship Between Childhood Major Adverse Experiences and Social Anxiety in Early Adulthood Among Chinese College Students.” Journal of Affective Disorders 292: 614–622. 10.1016/j.jad.2021.05.105. [DOI] [PubMed] [Google Scholar]
  30. Molero, J. M. , Perez‐Fuentes M., Oropesa R. N., Simon M. M., and Gazquez L. J.. 2019. “Self‐Efficacy and Emotional Intelligence as Predictors of Perceived Stress in Nursing Professionals.” Medicina 55, no. 6: 237. 10.3390/medicina55060237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Oren‐Yagoda, R. , Melamud‐Ganani I., and Aderka I. M.. 2021. “All by Myself: Loneliness in Social Anxiety Disorder.” Journal of Abnormal Psychology 131: 4–13. 10.1037/abn0000705. [DOI] [PubMed] [Google Scholar]
  32. Peng, S. , Wang X., Niu G. F., Zhang H. P., and Hu X. N.. 2019. “The Effect of Fear of Negative Evaluation on Social Anxiety: A Cognitive‐Behavioral Model Based on Social Anxiety.” Psychological Development and Education 35, no. 1: 121–128. 10.16187/j.cnki.issn1001-4918.2019.01.14. [DOI] [Google Scholar]
  33. Perez‐Martin, A. M. , Garcia M. M., Ruiz‐Moral R., Crespo‐Canizares A., Mena C., and Caballero‐Martinez F.. 2022. “Ibero‐American Consensus on Communication Skills for Nursing Degree Students.” Revista Latino‐Americana de Enfermagem 30: e3565. 10.1590/1518-8345.5653.3565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Raita, Y. , Goto T., Faridi M. K., Brown D., Camargo C. J., and Hasegawa K.. 2019. “Emergency Department Triage Prediction of Clinical Outcomes Using Machine Learning Models.” Critical Care 23, no. 1: 64. 10.1186/s13054-019-2351-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Regier, D. A. , Kuhl E. A., and Kupfer D. J.. 2013. “The DSM‐5: Classification and Criteria Changes.” World Psychiatry 12, no. 2: 92–98. 10.1002/wps.20050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reichenberger, J. , Wiggert N., Wilhelm F. H., et al. 2019. “Fear of Negative and Positive Evaluation and Reactivity to Social‐Evaluative Videos in Social Anxiety Disorder.” Behaviour Research and Therapy 116: 140–148. 10.1016/j.brat.2019.03.009. [DOI] [PubMed] [Google Scholar]
  37. Rodebaugh, T. L. , Woods C. M., Thissen D. M., Heimberg R. G., Chambless D. L., and Rapee R. M.. 2004. “More Information From Fewer Questions: The Factor Structure and Item Properties of the Original and Brief Fear of Negative Evaluation Scale.” Psychological Assessment 16, no. 2: 169–181. 10.1037/1040-3590.16.2.169. [DOI] [PubMed] [Google Scholar]
  38. Rose, G. M. , and Tadi P.. 2022. Social Anxiety Disorder. Treasure Island (FL): StatPearls. [Google Scholar]
  39. Shan, Y. X. , Tang F. P., Liu J., and Li J.. 2018. “A Study on the Factors Influencing Social Anxiety Among Nursing Specialist Students.” Nursing Research 32, no. 3: 439–440. 10.3969/j.issn.1009-6493.2018.03.029. [DOI] [Google Scholar]
  40. Song, Y. Y. , Li Y. Q., Hou Y., and Li K.. 2015. “Application of Variable Capture Method of Random Forest in Variable Screening of High‐Dimensional Data.” China Health Statistics 32, no. 1: 49–53. 10.13228/j.boyuan.issn1002-1183.20140343. [DOI] [Google Scholar]
  41. Spinhoven, P. , Elzinga B. M., Penninx B., and Giltay E. J.. 2021. “Temporal Relationships Between Happiness and Psychiatric Disorders and Their Symptom Severity in a Large Cohort Study: The Netherlands Study of Depression and Anxiety (NESDA).” BMC Psychiatry 21, no. 1: 344. 10.1186/s12888-021-03346-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Su, B. B. , Du Z., Jia J. Z., et al. 2018. “A Study on the Current Situation of Nursing Human Resources and Their Allocation in China.” China Health Policy Research 11, no. 12: 56–61. [Google Scholar]
  43. Turan, N. , Ozdemir A. G., Kaya H., Aksel G., and Yilmaz A.. 2019. “Male Nursing Students' Social Appearance Anxiety and Their Coping Attitudes.” American Journal of Mens Health 13, no. 1: 1818103810. 10.1177/1557988319825922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vilaplana‐Perez, A. , Perez‐Vigil A., Sidorchuk A., et al. 2021. “Much More Than Just Shyness: The Impact of Social Anxiety Disorder on Educational Performance Across the Lifespan.” Psychological Medicine 51, no. 5: 861–869. 10.1017/S0033291719003908. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Villarosa‐Hurlocker, M. C. , Whitley R. B., Capron D. W., and Madson M. B.. 2018. “Thinking While Drinking: Fear of Negative Evaluation Predicts Drinking Behaviors of Students With Social Anxiety.” Addictive Behaviors 78: 160–165. 10.1016/j.addbeh.2017.10.021. [DOI] [PubMed] [Google Scholar]
  46. Wang, C. K. , Hu Z. F., and Liu Y.. 2001. “A Study of the Reliability and Validity of the General Self‐Efficacy Scale.” Applied Psychology 1: 37–40. [Google Scholar]
  47. Wang, X. D. , Wang X. L., and Ma H.. 1999. Handbook of Mental Health Rating Scales (Updated Edition). Beijing: China Mental Health Journal Press. [Google Scholar]
  48. Weeks, J. W. , Heimberg R. G., Fresco D. M., et al. 2005. “Empirical Validation and Psychometric Evaluation of the Brief Fear of Negative Evaluation Scale in Patients With Social Anxiety Disorder.” Psychological Assessment 17, no. 2: 179–190. 10.1037/1040-3590.17.2.179. [DOI] [PubMed] [Google Scholar]
  49. Weeks, J. W. , Heimberg R. G., and Rodebaugh T. L.. 2008. “The Fear of Positive Evaluation Scale: Assessing a Proposed Cognitive Component of Social Anxiety.” Journal of Anxiety Disorders 22, no. 1: 44–55. 10.1016/j.janxdis.2007.08.002. [DOI] [PubMed] [Google Scholar]
  50. Wu, T. , Yang N. X., Cai L., Zhang Y., and Sang Z. Q.. 2021. “The Effect of Self‐Compassion on Social Anxiety: The Mediating Role of Self‐Esteem and Evaluative Fear.” Chinese Journal of Clinical Psychology 29, no. 1: 169. [Google Scholar]
  51. Xie, H. , Gao Q., and Yang Y. W.. 2021. “Analysis of Social Anxiety Among Intern Nursing Students and Its Influencing Factors.” Medical Theory and Practice 34, no. 23: 4189–4191. [Google Scholar]
  52. Xu, C. L. , and Yang X.. 2019. “Research Progress of Social Anxiety Among Nursing Students.” General Practice Nursing 17, no. 24: 2968–2970. [Google Scholar]
  53. Xu, Y. , Liu H. Y., and Liu H. M.. 2024. “Parallel Mediating Role of Technology Intrusion and Fear of Negative Evaluation in the Relationship Between Social Anxiety and Smartphone Addiction of College Students.” Chinese Journal of Health Psychology 32, no. 6: 926–933. 10.13342/j.cnki.cjhp.2024.06.023. [DOI] [Google Scholar]
  54. Ye, B. , Wan J. B., Yue X. L., and Ni J. Y.. 2017. “A Systematic Evaluation on the Status Quo of Social Anxiety of Nursing Students in China and Its Characteristics.” Nursing Research 31, no. 1: 18–22. 10.3969/j.issn.1009-6493.2017.01.006. [DOI] [Google Scholar]
  55. Ye, Y. C. , Lin R. M., and Yan Y. W.. 2021. “Big Trees Attract Big People: Fear of Positive Evaluation Fear of Positive Evaluation in Socially Anxious People.” Advances in Psychological Science 29, no. 6: 1056. 10.3724/SP.J.1042.2021.01056. [DOI] [Google Scholar]
  56. Yu, Y. , Liu S., Song M., Fan H., and Zhang L.. 2020. “Effect of Parent–Chi Attachment on College Students' Social Anxiety: A Moderated Mediation Model.” Psychological Reports 123, no. 6: 2196–2214. 10.1177/0033294119862981. [DOI] [PubMed] [Google Scholar]
  57. Yue, Y. , and Jia Y.. 2022. “Fear of Negative Evaluation: A Cross‐Sectional Study Among Undergraduate Nursing Students.” Nurse Education Today 121: 105678. 10.1016/j.nedt.2022.105678. [DOI] [PubMed] [Google Scholar]
  58. Zeng, J. H. , Lu A. T., Guo Y. Y., and Cai R. Y.. 2017. “The Relationship Between Social Support and Social Anxiety in College Students With Financial Difficulties: The Mediating Effect of Resilience.” Psychological Research 10, no. 2: 83–89. CNKI:SUN: OXLY.0.2017‐02‐012. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Nursing Open are provided here courtesy of Wiley

RESOURCES