Introduction
Social support is crucial for enhancing nursing students’ mental health by helping them effectively cope with stress through emotional, informational, and positive support in reciprocal relationships [1,2]. Social support is particularly multifaceted and includes educational support such as academic support, psychosocial support including positive evaluations, and instrumental support with the informational provision. This multifaceted support yields positive effects across various domains, which includes enhanced academic self-efficacy, adaptation to clinical practice, and improved mental health during crises and stressful situations [3-5]. Therefore, social support experienced by nursing students should be accurately defined and measured.
While reviewing instruments used in previous studies to assess social support among nursing students, we found that various instruments focused on support from friends, parents, and significant others. These instruments included general social support scales [6-8] and only one instrument was specifically designed for nursing students with clinical practice experience [9]. Existing scales measuring social support had limitations in that they did not reflect the characteristics of nursing students [6-8]. Specifically, social support should include various aspects by simultaneously reflecting not only structural aspects such as the size and frequency of social networks but also functional aspects such as informational, emotional, and material aspects. However, there were limitations in that these aspects were not considered [6,8], or in some cases, the scale was developed for elementary and middle school students rather than nursing students or college students, making it difficult to apply in terms of the validity of the research instruments [7]. Although there is a social support measurement scale developed for nursing students, this scale was developed so that it could only be applied to nursing students with clinical practice experience [9], and thus has limitations in measuring social support for nursing students of all grades. In addition, since this scale was developed in a Korean version, it can be applied in Korea, but in order to use this scale overseas, there is a difficult procedure of translating it into the language of that country and verifying its reliability and validity, so there are limitations that make it difficult to apply easily.
To compensate for these limitations, we recognized the need for a new social support scale. The scale developed in this study can be applied globally to nursing students of all grades and will sufficiently reflect the characteristics of nursing students. Choi et al. [4] identified the conceptual attributes and sub-attributes of social support of nursing students by considering the characteristics of nursing students. According to Choi et al. [4], social support includes four attributes and nine sub-attributes. These four attributes include structural, instrumental, educational, and psychosocial support, as illustrated in Figure 1. Structural support includes the structural aspect of social support and the range of social relationships available to an individual and support from social networks such as the size, frequency, and intimacy of social support. Instrumental support encompasses the functional characteristics of social support and includes informational and economic support for problem solving provided to nursing students. Educational support encompasses the functional aspect and academic support provided to nursing students and support from role models provided to nursing students by a variety of supporters (e.g., professors, clinical instructors, etc.). Psychosocial support encompasses the functional characteristics and the attributes of positive appraisal, emotional support, and self-esteem support, and is the provision of encouragement, positive feedback, and respect from significant others to nursing students. Consequently, assessing nursing students’ social support requires a multidimensional framework, and an instrument that captures general support and specific characteristics of both theoretical and clinical practice education is imperative.
Therefore, we aimed to develop a social support scale, tailored to the general social support and characteristics of nursing students. The specific objectives were as follows: (1) to formulate preliminary items for the scale; (2) to verify the content validity; (3) to verify the construct validity; (4) to verify the criterion validity; and (5) to verify its reliability.
Methods
Research design
This study is a methodological research. This study comprises two phases: (1) scale development and (2) scale validation.
Phase 1: Scale development
● Preliminary items development
In the study by Choi et al. [4], there were four attributes (structural, educational, instrumental, and psychosocial support), but as a result of the content validity verification by five experts (nursing professors), structural support was changed to “social relationship support.” The experts suggested that “structural support” only emphasized quantitative elements such as the size and frequency of social support, and that it would be better to change the term to “social relationship support” to emphasize the relational characteristics of social support. Since both educational and instrumental support are material supports provided to nursing students, they were combined into “resource provision support.” Finally, psychosocial support was changed to “positive support.” Experts changed the somewhat abstract name “psychosocial support” to “positive support” to make it more intuitive and easier to understand. As a result of the content validity verification, four attributes and nine sub-attributes were changed to three attributes (social relationship, resource provision, and positive support) and eight sub-attributes (Figure 1). Twenty indicators and preliminary items were developed based on eight sub-attributes. The twenty indicators and preliminary items were formulated from the study analyzing the concept of social support of nursing students by Choi et al. [4], the previously developed scale [9], and additional literature reviews by authors, and were completed through several revisions and supplements through the content validity verification process by experts (Table 1).
Table 1
Development of Preliminary Items
● Content validity verification
The content validity was judged by five experts, all of whom were nursing professors with educational and clinical experience. Content validity index (CVI) is a validity check by experts on whether the developed preliminary items are appropriate to measure the concept (social support), and 1 point is given if the item is very inappropriate, 2 points if it is inappropriate, 3 points if it is appropriate, and 4 points if it is very appropriate. Item CVI (I-CVI) is the CVI score of each item, and scale CVI (S-CVI) is the total CVI score for all items. I-CVI is calculated as the proportion of experts who gave 3 or 4 points out of all experts. The S-CVI is calculated by adding the I-CVI of all items. It is considered suitable when both I-CVI and S-CVI are .80 or higher [10,11]. I-CVI and S-CVI for all 20 items were perfect at 1.00, indicating unanimous agreement among the experts that the items are appropriate. While no items were removed, the items were reworded for clarity and ease of understanding for nursing students.
● Preliminary survey
The 20-item questionnaire utilizing a Likert scale with five-point was administered to 20 nursing students. Following the preliminary survey, no items proved difficult to understand or respond to; therefore, all the preliminary survey items were incorporated into the main survey. The preliminary survey included students of all grades (five first graders, five second graders, five third graders, and five fourth graders), comprising eight male and twelve female students.
● Main survey
The main survey was conducted in two phases. The first phase aimed to examine an exploratory factor analysis (EFA), whereas the second survey sought to conduct a confirmatory factor analysis (CFA).
Phase 2: Scale validation
● Construct validity verification
An initial EFA followed by a CFA was performed to confirm the scale’s constructs. Subsequently, convergent and divergent validity were assessed.
● Criterion validity verification
It was established by examining the correlation between the scale of this study and an existing scale [12]. The instrument used to test criterion validity is the perceived social support scale developed by Kim et al. [12] in a shortened form for ease of use of the previously developed tool and verified reliability and validity. The reason for selecting this instrument is that there must be a correlation between the instrument for verifying criterion validity and the scale developed in this study [10], and it is an instrument developed to measure social support, and it is easy to use and has the latest version. The scale comprises eight items assessing aspects, such as “the support provider makes me feel valued,” “shows interest in my work,” “provides problem-solving guidance” and “assists me in overcoming difficult situations.”
Study participants and data collection
Participants were recruited through convenience sampling, and data were collected from January 20 to February 26, 2024. The inclusion criterion was students enrolled in the department of nursing of the universities approached, who voluntarily participated in this study. An online survey was administered to 450 nursing students at universities in Jeonbuk, Daejeon, Chungbuk, and Seoul. The survey was conducted in two phases. The first phase involved 225 students to validate the EFA, and the second phase included 225 students to validate the CFA. The surveys were distributed online via Google Forms (Google Workspace). After excluding one incomplete response, 449 responses (225 for EFA and 224 for CFA) were used in the final analysis. The sample size surpassed the recommended criterion for factor analysis, exceeding 200 participants or 10 times the number of items [11].
Data analysis
To analyze the data of this study, IBM SPSS program version 23.0 (IBM Corp.) and AMOS version 23.0 (IBM Corp.) were used. The item analysis, EFA, criterion validity, and reliability were conducted using the SPSS program. CFA, convergent, and divergent validity were conducted using the AMOS program. The item analysis was performed using skewness, kurtosis, and item-total correlation coefficient values. The acceptable range of skewness should not exceed the absolute value of 2.00, and the absolute value of kurtosis should not exceed 7.00 [11]. In addition, the acceptable range of item-total correlation is .30 to .80 [10,11]. EFA was performed using the maximum likelihood with promax method. CFA fit indices were assessed: chi-square minimum/degree of freedom (CMIN/DF), comparative fit index (CFI), goodness-of-fit index (GFI), standardized root mean residual (SRMR), Tucker-Lewis index (TLI), and root mean square error of approximation (RMSEA). Convergent validity was confirmed through standardized estimates, average variance extracted (AVE), and construct reliability. Convergent validity is considered established when standardized estimates are greater than .50, AVE is greater than .50, and construct reliability is greater than .70 [10,13]. Divergent validity was established by comparing the squared correlation between AVE and factors, and the correlation coefficient±2×standard error (2SE) between factors. Firstly, AVE was greater than the correlation coefficient between factors, and secondly, by checking whether the correlation coefficient at ±2SE between factors did not include 1.00 [10,13]. Criterion validity was examined using Pearson’s correlation coefficients between the two scales.
Ethical considerations
This study was approved by the Institutional Review Board of the Chungbuk National University (IRB No. CBNU-202301 -HR-0005). Recognizing nursing students as a vulnerable group, data collection took place during summer vacation when grades and credits for the semester were awarded. Survey notifications were communicated to the department’s teaching assistants to ensure voluntary participation. After start of the summer vacation, the teaching assistant first announced the study instructions (e.g., the purpose of this study, participation guide, questionnaire, and rewards for participation) in the SNS group chatroom of each grade. The name of the professor was not revealed in the announcement. After sufficient notice, an online questionnaire was distributed to the students who personally contacted the teaching assistants via a URL address. After reading the explanation of the research purpose, survey process, anonymity, and confidentiality, the participants voluntarily signed a written consent form and took the survey. The participants were offered a $10 online gift-certificate as a token of appreciation for completing the survey.
Results
Characteristics of subjects
Subjects’ characteristics are detailed in Table 2. Of the total participants, 401 participants (89.3%) were women, 297 participants (66.2%) were satisfied with major, and 301 participants (67.0%) were satisfied with clinical practice.
Table 2
Demographic Characteristics of Participants
Scale development
The initial version of social support scale for nursing students consisted of 20 items grouped into three dimensions (social relationship support, resource provision support, and positive support). For CVI calculation, five experts (nursing professor) were invited to check the items’ adequacy. The CVI was acceptable for each of items.
● Item analysis
The range of kurtosis was -0.63 to 3.98, remaining within the value of 7.00, indicating that the normality assumption was met. The range of skewness was -1.61 to -0.09, falling within the value of 2.00. [11]. The mean score for each item varied from 3.31~4.55 on a scale of 1.00~5.00, with a standard deviation of 0.60~1.24. Item-total correlation coefficient values ranged from .36~.78, which falls within the suggested range of .30~.80 [10,11]. Consequently, no items were removed.
● Construct validity verification
∙ EFA
To examine the suitability of data for EFA, we checked for a Kaiser-Meyer-Olkin (KMO) value of .80 or higher and performed Bartlett’s test of sphericity (p<.05). The analysis revealed a KMO value of .89 and a significant chi-square value of 1,941.19 (p<.001).
In the initial EFA, five items with communalities less than .30 were removed: Items 5 (.25), 15 (.17), 16 (.25), 19 (.13), and 20 (.11). After removing these five items, a second EFA was conducted. In the second round, all 15 items demonstrated communalities at or above .30 (.34~.99) and were retained entirely. The recommended factor loading is .40, and the recommended cross-factor loading difference from the factor loading is .20 [13]. Three additional items were removed because the cross factor loading value was not satisfied. Specifically, Item 6 (factor loading=.50; cross-factor loading=.40), item 7 (factor loading=.45; cross-factor loading=.40), and item 14 (factor loading=.50; cross-factor loading=.36) were removed. After removing these three items, a third EFA was performed. The third EFA resulted in 3 factors and 12 items and explained a total variance of 62.5%. Factor 1, which comprised items from preliminary items related to emotional, self-esteem, and appraisal support, was named “positive support.” Factor 2, which comprised items from support from social network and support from significant others, was labeled “social relationship support.” Factor 3, which comprised items related to informational support, was named “informational support” (Table 3).
Table 3
Exploratory Factor Analysis
∙ CFA
For the initial CFA of the 12 items and 3 factors derived from the EFA, a new sample of 224 participants was surveyed. The GFI results are presented in Table 4: CMIN/DF=2.18, GFI=.93, CFI=.95, TLI=.93, SRMR=.02, RMSEA=.07. These values meet he cut-off values of CMIN/DF below 3.00, GFI above .90, CFI above .90, TLI above .90, SRMR below .05, and RMSEA below .08 [14,15]. Following the initial CFA, we examined the squared multiple correlation (SMC) for each item. Item 4 had an SMC of .19, which is below the cutoff value of .30 or higher [16]. Consequently, Item 4 was removed, and a second CFA was conducted with the remaining 11 items. The results of the second CFA indicated that the SMC of the 11 items ranged from .32~.99. The remaining items were retained, resulting in a scale with 11 items and 3 factors.
Table 4
Confirmatory Factor Analysis, Convergent Validity, and Divergent Validity
∙ Convergent validity verification
Following the CFA, convergent validity was verified using three criteria: standardized estimates above .50, AVE above .50, and construct reliability above .70. The range of standardized estimates was .56 to .99; AVE values were .76 (factor 1), .54 (factor 2), and .84 (factor 3); construct reliability values were .95 (factor 1), .78 (factor 2), and .91 (factor 3) (Table 4). All these values met the abovementioned criteria and provide support for convergent validity.
∙ Divergent validity verification
The divergent validity was assessed in two ways: first, by checking whether the AVE was greater value than the correlation between factors, and second, by checking whether the correlation coefficient at ±2SE between factors did not include 1.00. First, AVE values for factors 1, 2, and 3 ranged from .54~.84, whereas the correlation coefficient between factors ranged from .07~.36. These results indicate that the AVE values were relatively large and met the specified criteria. Second, the correlation coefficient±2SE between factors 1 and 2 ranged from .32~.40, factors 2 and 3 ranged from .11~.19, and factors 1 and 3 ranged from .03~.11. All these values met the criteria, as they did not include 1.00 (Table 4).
● Criterion validity verification
The Pearson correlation coefficient between the existing eight-item scale developed by Kim et al. [12] and the finalized scale was .61 (p<.001), indicating the validity of the criteria.
Discussion
The aim of this study was to develop a social support scale that reflects the general social support and characteristics of nursing students. The developed scale comprised 11 items categorized into three factors: positive, social relationship, and informational support.
“Positive support” encompassed items measuring emotional, self-esteem, and appraisal support. These aspects include feeling respected, positively evaluated, and recognized as valuable by significant others. These align with previous research that regarded social support as a positive enhancer [17] and emphasized emotional, appraisal, and self-esteem support [18,19]. In a previous study that developed a Korean-style social support scale targeting college students and middle-aged people, emotional factors were identified as an important attribute of social support [20]. In addition, a previous study that developed an online social support scale for college students also found that self-esteem and emotional support were important attributes [21]. A study examining the validity and reliability of a perceived social support measure also included evaluative support as an attribute of social support [12], and a study examining the concept of social support also included self-esteem and emotional support as important attributes of social support, which is similar to the findings of this study [18,22]. Notably, this study differs from previous research in its organization of self-esteem, emotional, and appraisal support as separate sub-factors. Nursing students perceived positive appraisal, respect for their individuality, and emotional support as a cohesive form of support. Consequently, for the effective integration of positive support in future theoretical and clinical-practice education, offering positive appraisal, emotional support, and support for self-esteem is crucial.
Items related to “social relationship support” were designed to measure support from social network and support from significant others, incorporating aspects such as the number of acquaintances, number of close friends, and frequency of interactions. While previous studies have referred to these elements as structural aspects or support from social network [12,18,23], we adopted the term “social relationship support.” The rationale behind this terminology was the belief that the concept of “social relationship support” is more familiar and understandable than “structural aspects and social networks.” Regarding the nursing students of this study, the greater the number of individuals providing social support and the more frequent their interactions, the higher their perceived level of social support. This perspective aligns with the research of Warshawski et al. [24], who noted that student nurse perceived social support as care. Also, the previous study that developed social support for nursing students who experienced clinical practice also derived emotional and economic support from family and friends and support from social networks as important attributes of social support [9], which is consistent with the results of this study. The results of this study are similar to those of a previous study that developed a scale for parental support in college students, which included economic and emotional support as parental support [25]. Given this understanding that nursing students associate social support with care, continually enhancing both the number and frequency of social relationships to foster adaptability is essential.
“Informational support” comprised items focused on providing information and the presence of individuals who supplied the necessary information. Notably, nursing students expressed a specific need for someone to furnish information regarding theory and clinical practice education. This observation is consistent with prior research findings [18,26,27] wherein students have indicated a desire for educational information from various sources, such as seniors, professors, and nurses [4]. In a previous study that developed the social support of nursing students who experienced clinical practice, the school’s support, support from professors and clinical instructors, and support from department seniors were derived as important attributes of social support, which corresponds to the results of this study [9]. In addition, a previous study that developed an online social support scale targeting college students included instrumental and informational support as social support attributes, which was similar to the results of this study [21]. To address this need, constant attention and effort are essential to establish a mentor-mentee system connecting seniors and juniors at the departmental level. Additionally, implementing a dedicated support system of nursing students in clinical practice will help meet the informational support requirements of nursing students.
This study makes significant contribution because it has developed a globally applicable social support scale that considers the general social support and characteristics of nursing students. In particular, it is significant that a social support scale applicable to all nursing students was developed. The scale developed in this study is thought to be applicable to nursing educators, administrators, and researchers as follows. Nursing educators working at universities can utilize the scale developed in this study to measure the level of social support of nursing students at the time of their admission, when they advance to a higher grade, before and after their admission practicum, and at the time of graduation, and use it to establish a system for early detection and intervention of students with problems. Nursing administrators working in clinical settings can use the scale to check and utilize the social support system of nursing students participating in clinical practice education, and nursing researchers can use the scale developed in this study to conduct research on the level of social support of nursing students and how to improve it.
However, the limitation of is that the data collection and analysis were confined to a single country, impeding the generalizability of the findings. Therefore, future studies should validate the scale in other countries to facilitate the generalization of the study’s results. Another limitation of this study is that the survey was conducted using a convenience sampling method, so caution is needed when interpreting and generalizing the results of this study. In particular, since the tool developed in this study focused on nursing students, caution is also required when applying it to general college students. There is also a limitation in the reliability verification procedure that test-retest reliability could not be verified. Test-retest reliability is to check the correlation after applying the scale to the same subjects of 10%~20% of the subjects of the study at two-week intervals. This study had some aspects that could not be confirmed due to the process of collecting data online during the summer vacation period. It is thought that future studies that improve this aspect will be necessary.
Also, the scale of this study was developed to measure not only the general social support of nursing students but also their characteristics. However, during the exploratory and CFA, which is the process of verifying the construct validity of the scale, items (e.g., 4, 7, 15, 16) that reflected the characteristics of nursing students were excluded. This was an unexpected part in the research planning stage. If items reflecting the characteristics of nursing students are recognized as important, they can be added as items of the scale through an expert validation process regardless of the statistical analysis results [28,29]. However, this study was conducted using a convenience sampling method and targeted only nursing students enrolled in some regions, so there are limitations in interpreting the results arbitrarily. In order to correct these important aspects in the future, it is thought that a repeat study should be conducted by extracting a representative sample of nursing students enrolled nationwide, and that a study that can confirm reliable and valid results should be conducted again.
Conclusion
It was developed the scale tailored to assess social support among nursing students. The scale comprised 11 items categorized into 3 factors: positive support, social relationship support, and informational support. Future research should use the scale developed in this study to measure, compare, and analyze the social support experienced by student nurse in various contexts. The scale developed in this study is expected to be used as a screening tool to measure the level of social support of nursing students at the time of admission, before and after clinical practice, and at the time of graduation, and to identify problems at an early stage. In addition, it is anticipated that a large sample study will be attempted in the future to increase the applicability and usability of the scale developed in this study, and to develop a tool that better reflects the characteristics of nursing students.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
