Different Sources of Social Support and Their Associations with Academic Engagement and Life Satisfaction Among Nursing and Physiotherapy Students

Abstract

Background: The role of social support in fostering academic engagement (AE) and life satisfaction (LS) is well established, as both are essential for student well-being and success. Theoretical frameworks suggest that distinct sources of social support contribute uniquely to these outcomes. However, limited research has explored these relationships in specific populations, such as nursing and physiotherapy students. Methods: This study aimed to examine the differential relationships between three sources of social support (family, significant other, and friends), AE and LS in a sample of 363 university students (84% female, M_age = 26.52, SD = 8.13) in Croatia (nursing and physiotherapy program). A cross-sectional design was employed using self-reported data. Results: The results of a hierarchical regression analysis revealed that family support was the only significant positive predictor of both AE and LS. Among the demographic variables, an older age was associated with higher AE and LS, while employed students reported lower LS. The findings were consistent across younger (18–24 years) and older (25–48 years) subgroups, with the exception of older students’ AE, where none of the social support sources emerged as significant predictors. Regarding bivariate correlations, all three social support subscales correlated positively with AE and LS, but the correlations were higher for LS. Conclusions: These results highlight the central role of family support in academic and life outcomes for nursing and physiotherapy students, offering insights for targeted interventions to enhance student well-being.

1. Introduction

1.1. Academic Engagement and Social Support

Research consistently demonstrates a positive relationship between social support and academic engagement (AE) or performance among university students. Multiple studies have found significant correlations between social support and academic achievement (Li et al., 2018; Song et al., 2015; Tayfur & Ulupinar, 2016), while fewer are focused specifically on AE. Performance measures are typically based on objective data, such as test results or GPA (grade point average). In contrast, AE reflects students’ psychological and behavioral investment in academics, including their motivation, participation, and effort in learning activities. A meta-analysis revealed that AE has a moderate, positive association with academic achievement and is often viewed as its antecedent (Lei et al., 2018).

AE and academic performance are positively associated, but the degree or their association varies across research (Delfino, 2019; Martínez et al., 2019). AE is commonly conceptualized as consisting of three components: vigor, dedication, and absorption (Schaufeli et al., 2002a). This framework originates from the initial conceptualization of work engagement. The Utrecht Work Engagement Scale (UWES), originally designed to measure work engagement, was adapted to assess AE by modifying its items for an academic context. Vigor reflects mental resilience, high energy levels, and persistence in overcoming challenges. Dedication is characterized by a deep sense of involvement, enthusiasm, pride, and inspiration in one’s academic tasks. Lastly, absorption refers to a state of full immersion and concentration in tasks (Schaufeli et al., 2002b). While AE largely focuses on individual psychological and behavioral engagement, external factors like social support play a critical role in shaping students’ ability to maintain and enhance their academic involvement. Social support is defined in various ways, but usually all authors agree that it involves some kind of relationship transaction between individuals (Zimet et al., 1988). Shumaker and Brownell (1984) pointed out that individuals need to perceive the transaction as enhancing one’s well-being.

Mulyadi et al. (2020) considered the context of social support (emotional, esteem support, instrumental, information, network), as well as sources of social support (family, significant other, friends), in explaining university students’ AE. Their results revealed that both form/context and all three sources of social support are positive predictors of AE. Social support from family and friends, but not from significant others, was found to be positively related to AE in a sample of undergraduate management students (Jayarathna, 2015). El-Sayed et al. (2021) found that higher perceived social support was a predictor of academic motivation in a sample of Egyptian nursing students. They did not differentiate between different sources of social support in their study. Similar positive associations between social support and AE were found among nursing students in Korea (An et al., 2017), as well as in heterogenous groups of students in Indonesia (Qurratuaini et al., 2022) and China (Liu, 2024; Xin, 2022). Koyanagi et al. (2021) concluded that students’ academic motivation (intrinsic, extrinsic, amotivation) is influenced by students’ individual social capital (support from family, peers and community). Xerri et al. (2018) found that peer and teacher support played a central role in students’ AE. Furthermore, An et al. (2017) found that the three sources of social support, namely family, friends, and significant others, remained stable throughout the four years of nursing studies.

However, in a sample of doctoral students, peer support and family support did not significantly predict AE, while institutional and supervisory support did (Cao et al., 2024). This may suggest that doctoral students are not fully representative of the broader university student population, as they might rely more heavily on intrinsic motivation to achieve academic goals and maintain AE.

1.2. Life Satisfaction and Social Support

Life satisfaction (LS) is usually defined as an individual’s overall evaluation of their life and is one of the key components of mental health and quality of life (E. Diener, 1984; Fujita & Diener, 2005). According to the tripartite model of subjective well-being (SWB), LS is considered the cognitive component. In contrast, positive and negative affect are categorized as affective measures (Džida et al., 2023; Metler & Busseri, 2017). While the tripartite model provides a theoretical framework for understanding LS, empirical research highlights the role of social support in influencing this construct across diverse populations.

On a representative sample of over one thousand Croatian citizens, Brajša-Žganec et al. (2018) reported that among the three sources of social support, only the perceived social support of friends was significantly related to subjective well-being. The sample was divided into three age groups and support from friends emerged as a significant predictor for all age groups. Gayathri and Karthikeyan (2016) found that social support from family and work colleagues was positively linked to LS in a sample of employed adults. Alorani and Alradaydeh (2018) found a positive relationship between the three sources of social support and LS. The correlation was highest for family, second highest for significant others, and lowest for friends. Alsubaie et al. (2019) reported that support from family and friends predicts individuals’ psychological quality of life, while support from significant others and friends predicts individuals’ social quality of life (mainly social relationships) on a sample of university students. Positive relationships between LS and social support have been reported in various studies, including one among the adult population (Kong et al., 2019) and adults with limitations on their activities of daily living (ADL) (Vaitsiakhovich et al., 2025), and among a heterogenous student population (Mahanta & Aggarwal, 2013; Ye et al., 2021; Yıldırım & Tanrıverdi, 2021). On the other hand, Mayungbo and Sunmola (2016) did not find that social support had an effect on college students’ LS. Perceived social support, especially from family, played a moderating role between emotional exhaustion and subjective well-being in a sample of female doctors (Wang et al., 2020).

Overall, research generally reports positive relationships between social support and AE or LS, with support from family or friends more frequently identified as significant compared to support from a significant other. Geerling and Diener (2020) reported that social support typically has a moderate-to-large effect on subjective well-being and is one of the most well-established correlates of it. Since these relationships can be highly dependent on age and developmental stage, more research is needed in specific groups.

One such specific group consists of students in helping professions, such as nursing and physiotherapy. These students face significant stress from both academic and clinical demands. Academically, they must manage intense workloads, frequent assessments, and challenges related to studying. Clinically, they encounter additional pressures, including the fear of unfamiliar situations, making mistakes with patients, and handling complex medical equipment (Pulido-Martos et al., 2012). Additionally, their schedules often include irregular hours due to clinical placements, which can disrupt their work–life balance. Some combine full-time employment alongside their studies, adding financial and time management pressures, while others focus solely on academics. Given these challenges, social support from family, friends, and significant others may play a crucial role in sustaining their LS. Despite the importance of social support in these contexts, research on university students in helping professions remains limited. A study focused on nursing students found that they were mostly satisfied with their perceived social support, and this satisfaction was associated with lower levels of perceived stress regarding their coursework (Almeida et al., 2018). However, more research is needed to further explore these dynamics.

The purpose of this study is to examine the differential relationships between three sources of social support (family, significant other, and friends) with AE and LS among physiotherapy and nursing students. Additionally, the study aims to determine whether these sources of social support predict AE and LS differently for younger and older students.

2. Materials and Methods

This study employs a cross-sectional, correlational design, and is based on self-reported data from frequently used questionnaires.

2.1. Participants

All the participants in this study (N = 363, M_age = 26.52, SD = 8.13) are university students enrolled in either Nursing studies or Physiotherapy at University North (Croatia). The Nursing department has bachelor’s and master’s-level studies, while Physiotherapy has only a bachelor-level (undergraduate) program. In total, 66.9% of the sample was enrolled in bachelor’s programs and 33.1% were enrolled in a master’s program (Nursing). The majority of the sample was female (84%), which reflects the gender distribution of these programs. More details of the sample are shown in

Table 1. Characteristics of the sample.

Variable	% of the Sample
Gender
Female	84% (305)
Male	16% (58)
Age
18–19	25.3% (92)
20–21	12.7% (43)
22–23	11.3% (41)
24–30	21.7% (79)
31–48	28.9% (108)
Year of study
First year 1 (bachelor’s level)	58.7% (213)
Second year 2 (bachelor’s level)	3% (11)
Third year 3 (bachelor’s level)	5.2% (19)
Fourth year 4 (1st master’s level)	31.4% (114)
Fifth year 5 (2nd master’s level)	1.7% (6)
Employed
No	31.4% (114)
Yes, in their study field	55.1% (200)
Yes, not in their study field	13.5% (49)
Study program
Nursing	78.2% (284)
Physiotherapy	21.8 (79)

. Participants were selected using a convenience sampling approach within a targeted population. Given that the study focuses on students in helping professions, data collection was conducted in classes where the researchers had access due to their teaching roles.

2.2. Instruments

The Multidimensional Scale of Perceived Social Support—MSPSS (Zimet et al., 1988) was used to evaluate participants’ perceptions of social support from three sources: family, friends, and significant others. The MSPSS consists of 12 items, each rated on a 7-point Likert scale, ranging from 0 (very strongly disagree) to 6 (very strongly agree). The scale yields a total score representing overall perceived social support, as well as subscale scores for the three sources of support. There are four items for each of the three subscales. Scores are calculated by summing the items within each subscale and dividing by the number of items, with higher scores indicating greater perceived social support. In the present study, the MSPSS demonstrated high internal consistency, with Cronbach’s alpha values of 0.95, 0.93, and 0.95 for the family, significant other, and friends support subscales, respectively.

The Utrecht Work Engagement Scale—Student Version (UWES-S-9) was employed to assess participants’ AE (Schaufeli et al., 2002a). Responses were recorded on a Likert-type scale ranging from 0 (never) to 6 (every day/all the time) for nine items in the scale. The instrument generates both a total score and subscale scores for three dimensions of AE: vigor, dedication, and absorption. Each subscale consists of three items. Scores (total and subscales) were calculated as indexes, with the sum of the items divided by the number of items on each scale. A higher total score indicates greater overall AE, while higher subscale scores represent greater vigor, dedication, or absorption. However, an exploratory factor analysis was conducted and revealed a clear one-factor solution for UWES-S-9. One factor explained 70.13% of the variance. In the present study, UWES-S-9 demonstrated high internal consistency, with a Cronbach’s alpha of 0.95 for the total scale and 0.90, 0.91. and 0.93 for vigor, dedication, and absorption.

Satisfaction with Life Scale (SWLS) was used to measure participants’ overall LS (E. D. Diener et al., 1985). The SWLS consists of 5 items, each rated on a 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). The total score is calculated as a mean of all responses to the five items, with higher scores reflecting greater LS. The scale provides a unidimensional measure of LS. In the present study, the SWLS demonstrated high internal consistency and reliability (Cronbach’s alpha = 0.89).

2.3. Procedure

The study was based on an online survey (Google Forms) and administered to students in person during their university lectures. Participants completed the survey using their personal smartphones. A researcher was present throughout the administration process, which lasted approximately 15–20 min. Participation in the study was entirely voluntary, and no compensation was provided to participants. Ethical approval was obtained from the author’s institution (protocol code 11-73/22-2381, 23 May 2022) and the study was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all subjects involved in the study.

2.4. Data Analysis

A robust methodological approach was employed, using validated instruments to ensure the reliability of the findings and comparability with prior studies. Data analysis began with an inspection of descriptive statistics to summarize the characteristics of the sample. Following this, a correlation matrix was conducted to examine the relationships between key variables. Hierarchical regression analyses were used to identify predictors of AE and LS. This approach was implemented on the full sample, as well as within two distinct age groups.

3. Results

Before conducting the main analysis, descriptive parameters for all study variables were examined. The characteristics of the sample are shown in more detail in Table 1.

The descriptive statistics for the main study variables are shown in

Table 2. Descriptive statistics of the main study variables.

Variable	Min	Max	Mean	SD	N
MSPSS—family	1	7	6.04	1.36	363
MSPSS—significant other	1	7	6.19	1.26	363
MSPSS—friends	1	7	5.97	1.28	363
UWES-S-9—Academic engagement	0	6	3.04	1.43	363
SWLS—Life satisfaction	1	7	4.76	1.29	363

Note: MSPSS—Multidimensional Scale of Perceived Social Support, UWES-S-9—Utrecht Work Engagement Scale, Student version, SWLS—Satisfaction with Life Scale, SD—standard deviation.

(three sources of social support, AE and LS).

The index scores for support and LS measures are high, compared to the average AE (M = 3.04), which can be described as moderate. After the analysis of descriptive parameters, a correlational matrix was made for all the main study variables. The correlation matrix is shown in

Table 3. Correlation matrix for the main study variables.

	AE	LS	Family	Sig. Other	Friends
AE	1
LS	0.25 **	1
Family	0.31 **	0.48 **	1
Sig. other	0.28 **	0.35 **	0.63 **	1
Friends	0.27 **	0.34 **	0.62 **	0.79 **	1

Note: AE—Academic Engagement, LS—Life Satisfaction, Sig. other—Significant other, ** p < 0.01.

.

There is a significant, positive relationship between all the main study variables. Correlations are highest for the MSPSS subscales (0.62–0.79). The three sources of social support are all positively related to both AE and LS, but the correlations are somewhat higher for LS.

A hierarchical multiple regression was performed to examine the possible predictors of AE and LS. The main assumptions for regression analysis, including the ratio of cases to independent variables, the absence of outliers, the absence of multicollinearity and singularity, the normality, linearity, and homoscedasticity of residuals, and the independence of errors, were satisfied (Tabachnick et al., 2013).

Two models were performed, one for AE and one for LS. Step one in both models included control (demographic) variables, namely age, gender (0—male, 1—female), year of study (1—first year, …, 5—fifth year of studying/second year at master’s level), and employment status (0—unemployed, 1—employed).

Since the group of students who were employed outside the medical field was small (N = 49), those students were added to the group that is employed in their study field. Step two included three social support subscales (family, significant other, friends).

In model 1 (

Table 4. Demographic variables and three sources of social support as predictors of students’ academic engagement (AE).

Model 1—Academic Engagement (AE)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.078	0.067	0.078	7.533(4,358)		<0.001
Age					0.199 **	0.002
Gender					0.068	0.180
Year of study					−0.062	0.296
Employment					0.126 *	0.05
Step 2	0.162	0.145	0.084	11.866(3,355)		<0.001
Age					0.178 **	0.005
Gender					0.018	0.723
Year of study					−0.060	0.289
Employment					0.109	0.084
Family					0.165 *	0.013
Sig. other					0.104	0.224
Friends					0.065	0.44

** Significant at the 0.01 level, * significant at the 0.05 level.

), the demographic variables significantly predict AE in the first step (F_(4,358) = 7.53, p < 0.001). Moreover, older age and employment are related to higher AE in the first step. Adding three sources of social support significantly increased the explained variance (F_(3,355) = 11.87, ΔR² = 8.4%, p < 0.001). Age was still a significant predictor of AE, but employment was not. Family support was the only significant predictor of AE among the social support variables, with higher support from family being associated with higher AE. Overall, seven predictor variables explained only 15% of the variance in AE.

In model 2, the same set of predictors in two steps was used to explain students’ LS (

Table 5. Demographic variables and three sources of social support as predictors of students’ life satisfaction (LS).

Model 2—Life Satisfaction (LS)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.036	0.025	0.036	3.334(4,358)		0.011
Age					0.200 **	0.002
Gender					0.026	0.625
Year of study					0.066	0.275
Employment					−0.174 **	0.009
Step 2	0.264	0.25	0.228	36.747(3,355)		<0.001
Age					0.149 *	0.011
Gender					−0.028	0.562
Year of study					0.068	0.204
Employment					−0.194 **	0.001
Family					0.396 **	0.000
Sig. other					0.051	0.526
Friends					0.078	0.320

** Significant at the 0.01 level, * significant at the 0.05 level.

).

In model 2, demographic variables significantly predicted LS in the first step (F_(3,358) = 3.33, p < 0.05). Older age and unemployment were related to higher LS. Adding three sources of social support significantly increased the explained variance (F_(3,355) = 36.75, ΔR² = 22.8%, p < 0.001). Age and unemployment remained significant predictors of LS. Family support was the only significant predictor of LS among the social support variables, with higher support from family being associated with higher LS. Overall, seven predictor variables explained 25% of the variance in LS.

Since the sample exhibited a wide age range (19–48 years) and age was a significant predictor of both AE and LS, the sample was divided by age to examine whether the same results applied to younger and older students. The sample was split into two groups of approximately equal size. The first group included students aged 18–24 (53.4%), while the second group comprised students aged 25–48 (46.6%). Hierarchical regression analyses were then repeated for each group. Among younger students, family support remained the only significant predictor of both AE and LS (see Appendix A,

Table A1. Demographic variables and three sources of social support as predictors of students’ academic engagement among younger students (18–24 years).

Model 3—Academic Engagement (AE)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.048	0.028	0.048	2.384(4,189)		0.053
Age					0.198	0.054
Gender					−0.004	0.950
Year of study					−0.151	0.119
Employment					0.125	0.135
Step 2	0.188	0.158	0.14	10.707(3,186)		<0.001
Age					0.231	0.018
Gender					−0.086	0.229
Year of study					−0.186 *	0.041
Employment					0.099	0.211
Family					0.186 *	0.023
Sig. other					0.107	0.373
Friends					0.157	0.193

* significant at the 0.05 level.

and

Table A3. Demographic variables and three sources of social support as predictors of students’ life satisfaction among younger students (18–24 years).

Model 5—Life Satisfaction (LS)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.04	0.02	0.04	1.984(4,189)		0.099
Age					0.110	0.286
Gender					0.000	0.997
Year of study					0.079	0.416
Employment					−0.224 **	0.008
Step 2	0.255	0.227	0.214	17.845(3,186)		<0.001
Age					0.153	0.098
Gender					−0.051	0.455
Year of study					0.016	0.856
Employment					−0.232 **	0.003
Family					0.380 **	0.000
Sig. other					0.081	0.483
Friends					0.063	0.585

** Significant at the 0.01 level.

). For older students, family support remained a significant predictor of LS (see Appendix A,

Table A4. Demographic variables and three sources of social support as predictors of students’ life satisfaction among older students (25–48 years).

Model 5—Life Satisfaction (LS)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.04	0.017	0.04	1.710(4,164)		0.150
Age					0.098	0.208
Gender					0.051	0.510
Year of study					0.004	0.960
Employment					−0.176 *	0.026
Step 2	0.286	0.255	0.246	18.477(3,161)		<0.001
Age					0.095	0.168
Gender					0.002	0.974
Year of study					0.055	0.423
Employment					−0.154 *	0.025
Family					0.453 **	0.001
Sig. other					−0.036	0.775
Friends					0.097	0.405

** Significant at the 0.01 level, * significant at the 0.05 level.

), but none of the social support variables were predictors of AE (see Appendix A,

Table A2. Demographic variables and three sources of social support as predictors of students’ academic engagement among older students (25–48 years).

Model 4—Academic Engagement (AE)
Predictor Variables	Unadjusted R2	Adjusted R2	ΔR2	F(df)	β Standardized	p-Value
Step 1	0.055	0.032	0.055	2.389(4,164)		0.053
Age					0.112	0.148
Gender					0.160 *	0.038
Year of study					−0.037	0.629
Employment					−0.107	0.168
Step 2	0.105	0.066	0.05	3.012(3,161)		0.032
Age					0.106	0.167
Gender					0.129	0.097
Year of study					−0.015	0.842
Employment					−0.099	0.198
Family					0.158	0.269
Sig. other					0.076	0.588
Friends					0.005	0.972

* significant at the 0.05 level.

).

4. Discussion

The findings of this study contribute to the growing body of literature on the role of social support in fostering AE and LS among university students, particularly nursing and physiotherapy students. Consistent with previous research highlighting the significance of social support in academic and psychological outcomes (Alsubaie et al., 2019; Brajša-Žganec et al., 2017; Gayathri & Karthikeyan, 2016; Geerling & Diener, 2020), family support emerged as a key predictor in this study. Among the three sources of social support examined—family, significant other, and friends—only family support uniquely predicted both AE and LS, underscoring the crucial role of familial relationships in shaping student outcomes. Many prior studies did not differentiate between sources of social support (El-Sayed et al., 2021; Liu, 2024; Qurratuaini et al., 2022), making it impossible to assess their unique contributions. This study mostly aligns with previous research regarding the overall positive association between social support and AE. On a bivariate level, all three sources of social support were associated with AE; however, the regression analysis revealed that only family support was a significant predictor. A similar pattern emerged for LS; family support was the only significant predictor of LS in the regression analysis. It is worth noting that the model predicting LS explained 25% of the variance, whereas the model predicting AE explained only 15%. These findings suggest that social support, particularly family support, plays a more substantial role in LS than in AE. This trend is also evident in the correlation matrix (Table 3), where family support shows a stronger correlation with LS (0.48, p < 0.01) than with AE (0.31, p < 0.01). Similarly, Hakimzadeh et al. (2016) found that satisfaction with family and peers had a low correlation with AE but a strong correlation with overall LS. Our findings also align with those of Alorani and Alradaydeh (2018), who reported a positive correlation between LS and all three sources of social support, namely family, significant other, and friends, in that order by correlation size. They did not conduct a regression analysis, so their results are not comparable to our main analysis.

The prominence of family support as the sole significant predictor of AE and LS is somewhat surprising, as existing research has not consistently highlighted such a pattern in university populations (Cao et al., 2024; Koyanagi et al., 2021; Mayungbo & Sunmola, 2016). Considering the developmental stage of most participants in the sample (late adolescence, early adulthood) and the typical developmental tasks undertaken in these periods (e.g., forming romantic partnership, belonging to a social group), one would expect friends and significant others to emerge as more relevant predictors of AE and LS (Berk, 2015). This finding invites an investigation within a cultural framework. Croatia can be viewed as a more traditional society within Europe (Črpić et al., 2022) and places significant emphasis on family values and the centrality of familial relationships in all aspects of life. It is possible that these cultural norms amplify the role of family as a primary source of support for university students, overshadowing the contributions of other support networks such as friends or significant others. This cultural dimension may differentiate Croatian university students from those in less family-oriented societies and highlights the need for culturally sensitive interpretations of social support research. It is important to note that multicollinearity analysis was conducted using VIF and tolerance values and that no issues were identified. However, family support surely did account for a portion of the variance explained by other predictors, as the sources of social support were highly correlated with one another (from 0.62 to 0.79, p < 0.01). Although family and significant other support did not emerge as significant predictors in the regression analysis, both showed moderate, positive correlations with AE and LS.

Interestingly, demographic variables such as age and employment status also played a role in AE and LS, with older and employed students reporting higher AE, while employment was negatively associated with LS. This is partially consistent with a study by An et al. (2017) who reported that older (senior) nursing students had higher AE compared to juniors and sophomores. However, in their study, freshmen students also had higher AE, which is not the case in our dataset. The findings in our study suggest that although employment may enhance AE through the practical application of skills, it may simultaneously place additional stressors on students, detracting from overall LS. This is congruent with Behlau (2010), who reported that employed university students had somewhat lower LS, and Li-Ping Tang et al. (2002), who reported that non-employed university students have higher LS. On the other hand, Lundberg (2004) reported that students who work 20 or more hours a week have lower AE, which is not congruent with our findings. However, our study is based on students who work and study in the medical field, so that may be the source of the positive association between AE and employment. Also, unemployed students are, on average, younger than employed ones, so it is possible that some of them are unsure whether they chose the right academic path. On the other hand, students who are older or already working are probably more secure about their academic decisions and are therefore more engaged in their coursework. On that note, almost a third of the sample (27.5%) had an average AE equal to 2 or lower (maximum is 6), which means that they estimated their engagement in coursework and learning as somewhat low. A future study should explore the undermotivated subgroup of students more thoroughly. Students who are employed full time (55.1% of the sample) and are simultaneously undertaking their bachelor’s or master’s degree are a unique population and should be explored further in the academic and psychological context.

A further analysis of age-divided subgroups revealed that the impact of social support varies by age for AE. Family support remained the only significant predictor of LS across both younger and older students but was not predictive of AE for older students. These results indicate that the importance of social support may shift over time, with older students potentially relying more on intrinsic or institutional factors to maintain AE. A study by Cao et al. (2024) on doctoral students revealed a similar result—peer and family support were not related to students’ AE. This finding warrants further exploration to better understand the evolving role of social support across different stages of student development.

As for the strengths of this research, some studies have previously investigated the relationship between social support and AE or LS in the general population or among university students. However, this study focuses on nursing and physiotherapy students, an underexplored population in this research field. Also, more than half of the sample comprises students who are full-time employed in their study field (nursing and physiotherapy), which is a unique sample of participants in this research field. Including this specific group provides insights into an understudied student population with unique challenges. Additionally, this study examines distinct sources of social support rather than social support as a generalized construct. This distinction allows for a nuanced exploration of how specific sources of social support relate to AE and LS. The questionnaires used to measure the primary study variables (MSPSS, UWES-S-9, SWLS) are widely used, well-validated measures with robust psychometric properties. This methodological choice enhances the reliability and validity of the findings but also facilitates a comparison with previous studies. It is also worth noting that some previous studies have only reported the correlations between variables, viewed social support as a unidimensional construct or not controlled for age, gender, the year of study, etc. This study uses regression analysis, which is superior to simple correlation because it allows for the examination of the unique contribution of each predictor while controlling for potential confounding variables such as age, gender, and the year of study. By doing so, it provides a more accurate and nuanced understanding of the relationship between social support and outcomes.

Despite its strengths, the study has certain limitations. As a cross-sectional study with a correlational design, it cannot capture longitudinal changes or establish causal relationships. The reliance on self-reported data introduces the potential for response bias, such as socially desirable answers, which is a common limitation in research using this methodology. Another limitation is the predominantly female sample, which limits our ability to compare results across genders. However, this gender imbalance reflects the actual distribution of students in nursing and physiotherapy programs and is not unique to this study. An additional challenge involves the interpretation of the significant other subscale because of the wording of the MSPSS items. The term ‘special person’ can be understood differently for different participants. For students in a romantic relationship, the term likely refers to their partner, whereas students who are not in a relationship may interpret this as referring to a family member or close friend. This ambiguity complicates the analysis of this subscale. Future research could address this issue by collecting information on participants’ relationship status and providing a more precise definition of a significant other or special person in the questionnaire.

5. Conclusions

This study examined the differential relationships between three sources of social support (family, significant other, and friends) and AE and LS in a sample of nursing and physiotherapy students. Hierarchical regression analyses revealed that family support was the only significant positive predictor (out of social support subscales) of both AE and LS. Among demographic variables, older age was associated with higher AE and LS, while employed students reported lower LS. The findings were consistent across younger (18–24 years) and older (25–48 years) subgroups, except in older students, where none of the social support sources significantly predicted AE.