Social media engagement and professional identity among Indonesian medical students: a cross-sectional analysis of patterns, intensity, and correlation

Abstract

Purpose

Professional identity (PI) formation occurs over time through immersive social—digital and in-person—interactions. While digital social interactions may support the PI formation of medical students, their influence remains underexplored. This study aimed to describe the patterns and intensity of social media use and analyze its association with PI.

Methods

This quantitative cross-sectional study was conducted at a private medical school in Jakarta, Indonesia. All undergraduate medical students were invited to participate in this study. Data were collected using the Social Networking Activity Intensity Scale (SNAIS) and the MacLeod Clark Professional Identity Scale (MCPIS-9) questionnaire. Data were analyzed descriptively and statistically.

Results

A total of 451 students participated in this study. Participants reported moderate overall social media use intensity (mean SNAIS=2.96±0.51), with higher use for entertainment than social interaction. The mean MCPIS-9 score was 3.76±0.44. Content engagement, measured using a content engagement index (CEI) reflecting the intensity of engagement with educational and informational content across selected social media activities, was generally low and driven mainly by browsing behaviors (CEI=0.51±0.64). Weak but significant correlations were observed between CEI and MCPIS-9 (ρ=0.139, p=0.003) and between SNAIS and MCPIS-9 (r=0.102, p=0.030). One-way analysis of variance revealed significant differences in MCPIS-9 scores by duration of use for WhatsApp and X, but not for other platforms.

Conclusion

Participants mainly used social media for entertainment purposes. Social media use intensity and content engagement showed weak but significant associations with PI.

Introduction

Professional identity (PI) refers to an individual’s interpretation of the profession’s characteristics, values, and attributes [1]. PI formation is particularly important in healthcare because it provides the foundation for healthcare workers to fulfill their responsibilities, including delivering high-quality care, ensuring patient safety, fostering effective doctor–patient relationships, and upholding the principles of humanism [2,3]. This process occurs through the internalization of professional values and characteristics, primarily through formal and informal social interactions in a community of practice, which represents the widely accepted framework of PI formation in healthcare profession education [1,4].

Numerous studies have indicated that students’ activities outside the formal curriculum, particularly through informal direct interactions with patients, senior colleagues, and other healthcare professionals, heavily influence PI formation [5,6]. Traditionally, these interactions occur in person. However, rapid technological advances have significantly blurred and shifted the boundaries between in-person and digital interactions. This shift was further accelerated by the recent COVID-19 (coronavirus disease 2019) pandemic, which normalized digital interactions across all aspects of life, including personal and educational contexts [7,8].

Social media has emerged as a communication and educational instrument that facilitates interaction among individuals, enabling users to create, engage with, and share content [9]. Social media plays an increasingly important role in shaping students’ PI as they spend more time online [10]. It acts as a virtual “community of practice,” offering alternative access to mentorship, role models, and continuous academic engagement [11,12]. Digital interactions with other professionals are believed to be as effective in supporting PI formation as traditional in-person interactions [13]. Social media can support knowledge building, skill enhancement, and professional growth when utilized effectively, as it offers a digital avenue for expressing and strengthening PI [14].

A recent study on social media and identity formation suggests that social media functions as a digital community that brings together individuals who share similar characteristics, values, or experiences related to identity formation [15,16]. Such communities provide meaningful engagement—a process that facilitates PI formation [17,18]—including social interactions, social connectedness, and psychosocial support, all of which play an important role in PI formation. Social media also enables its users to gain recognition—another important component of PI formation [19]—and to validate their identity through feedback mechanisms such as “likes” and comments on their posts [20]. The diversity of social media platforms, each offering distinct user experiences, allows social media to function as a medium for PI formation across varied populations, including different age groups and individuals with diverse health conditions [15,19,20].

Taken together, social media has potential as a learning environment that supports PI formation. However, to effectively leverage this potential within medical education, it is essential to first understand how medical students use social media, including their usage patterns and intensity—an area that remains insufficiently explored, particularly in relation to its impact on PI formation. Nevertheless, studies that explore the patterns and intensity of social media use by medical students, as well as their impact on PI, are still limited. In countries like Indonesia, where undergraduate medical education emphasizes theoretical mastery through classroom-based approaches, students’ exposure to real clinical practice is limited. Any alternative learning approaches that facilitate both academic and professional interactions will be valuable to support students’ PI formation early in their medical training.

This study aimed to: (1) examine the patterns and intensity of social media use among undergraduate medical students, captured by the modified Social Networking Activity Intensity Scale (SNAIS) questionnaire; (2) assess participants’ engagement with educational and informational content through a content engagement index (CEI); and (3) assess their correlation with PI, as measured by the 9-item MacLeod Clark Professional Identity Scale (MCPIS-9). We hypothesized that students with greater engagement in educational and professional social media content would have higher MCPIS-9 scores.

Methods

This quantitative cross-sectional study was conducted in a private medical school in Jakarta, Indonesia, between August and September 2023. All undergraduate medical students who were actively enrolled during the academic year 2023 were invited to participate in this study. Information about the study was sent to each student cohort via the school’s official Line app. Information was sent twice with a 1-week interval. Interested students were asked to complete the online informed consent and questionnaire forms. Ethical approval was obtained from the Ethics Committee of the School of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia (No: 12/08.KEP-FKIKUAJ/2023), and informed consent was obtained before data collection.

Participants’ social media use intensity was measured using the SNAIS questionnaire. Originally, SNAIS is a 14-item questionnaire that captures two dimensions of social media use, including: entertainment function use intensity (EFUI) (item 1–10) and social function use intensity (SFUI) (item 11–14). To identify the type of contents that participants engaged in, we added a follow-up question (“What kind of content—educational, informational, entertainment, promotional, or others—that they reacted to or engaged in?”) after question numbers 4, 5, 6, 8, 9, and 11 on the SNAIS questionnaire. We also added several questions to assess the participants’ general use of social media, such as which social media they mainly used, how many times they accessed their social media, and how much time they spent each day on it.

The SNAIS used in this study was based on a previously validated Indonesian version. The original SNAIS was translated using forward-backward translation and validated by Bella and Santoso [21], demonstrating excellent internal consistency (Cronbach’s α=0.89) and acceptable construct validity, with CFA factor loadings ranging from 0.48 to 0.76. Additional items were developed based on the literature on social media engagement and PI formation and were reviewed by medical education experts to ensure content relevance and clarity. Item–total correlation analysis indicated that most of the modified SNAIS items demonstrated acceptable correlations with the total scale, thereby indicating internal consistency. Most of the original SNAIS items showed moderate to strong corrected item-total correlations (CITC, 0.329–0.541), with a few items showing lower correlations, particularly items 9 (0.242), 13 (0.100), and 14 (0.266). In contrast, the additional items related to types of content engaged with demonstrated lower correlations, reflecting their role in capturing qualitative aspects of social media use rather than overall activity intensity (CITC, 0.94–0.253). The removal of these items resulted in only marginal increases in Cronbach’s α; therefore, all items were retained to preserve content coverage relevant to the study objectives. The modified SNAIS questionnaire demonstrated acceptable internal consistency (Cronbach’s α=0.76).

A CEI was developed to quantify students’ engagement with educational and professional social media content. Only SNAIS items that were followed by the additional content-type filter questions (items 4–6, 8, 9, and 11) were included in the construction of this index, as these items allowed differentiation between educational/professional and non-educational content. Engagement with educational or professional content was coded as 1, whereas engagement with non-educational content was coded as 0. Activity-specific engagement scores were calculated by weighting each activity’s frequency by its content classification. These weighted scores were then aggregated and averaged to generate the CEI, with higher scores indicating greater engagement with educational and professional social media content.

Participants’ PI was assessed using the MCPIS-9, which is a 9-item questionnaire that specifically assesses the affective and relational dimensions of PI, including participants’ sense of belonging, emotional attachment, and identification with the professional group. Both questionnaires used a 5-point Likert scale ranging from 1 (never) to 5 (always). The MCPIS-9 has been translated (backward and forward translation) and validated by Rahman [22] and had acceptable internal consistency (α_MCPIS-9=0.708).

The participants’ social media use patterns, intensity, and CEI were analyzed using descriptive statistics. A total of 451 students participated in this study. Normality testing using the Saphiro-Wilk test showed abnormal data distribution (W_SNAIS(451)=0.992, p=0.012; W_CEI(451)=0.792, p=0.000; W_MCPIS-9(451)=0.984, p=0.000). However, visual inspection of the Q-Q plots for SNAIS and MCPIS-9 scores demonstrated approximately symmetrical distributions centered around the median. Given the large sample size, the distributions for these two variables were considered to be sufficiently normal for parametric analyses. Accordingly, Pearson’s correlation was used to examine the relationship between social media use intensity and PI scores, whereas Spearman’s rank correlation was used to assess the association between CEI scores and PI scores. In addition, one-way analysis of variance (ANOVA) was conducted to examine differences in PI scores across groups based on duration of use for each social media platform. Data were statistically analyzed using STATA/IC ver. 15.0 (Stata Corp., College Station, USA).

Results

A total of 451 out of 814 students (314 female, 137 male) participated in this study (response rate=55.41%). The proportion of female students across academic years ranged from 61.5% in Year 1 to 72.2% in Year 2, 69.7% in Year 3, and 70.2% in Year 4, indicating a consistently higher representation of female students across all cohorts. Most participants (98.4%) were aged between 17 and 21 years, with fewer aged between 22 and 26 years. Overall, most participants accessed social media multiple times per day, with the largest proportion reporting use 5–9 times daily, followed by 10–13 times and more than 13 times per day (Table 1).

Table 1.

Distribution of Participants by Academic Year and Frequency of Social Media Use

Frequency of use (times/day)	Academic year				Total
	Year 1	Year 2	Year 3	Year 4
1–4	0 (0)	8 (6.3)	8 (6.6)	10 (6.6)	26 (5.8)
5–9	21 (40.4)	52 (41.3)	47 (38.5)	58 (38.4)	178 (39.5)
10–13	16 (30.8)	32 (25.4)	38 (31.1)	45 (29.8)	131 (29.0)
>13	15 (28.8)	34 (27.0)	29 (23.8)	38 (25.2)	116 (25.7)

Data are presented as number (%).

1. Social media use patterns and intensity

We asked participants to list all kinds of social media they used and how much time they spent on each platform each day. Eight social media platforms were identified, including three messaging apps (WhatsApp, Line, and Telegram), three multimedia sharing platforms (TikTok, YouTube, and Instagram), one microblogging site (X), and one social networking site (Facebook). Out of those eight platforms, participants spent the most time (more than 3 hours) on TikTok, YouTube, and Line. On the other hand, almost all participants spent less than an hour on Facebook and Telegram. Fig. 1 presents a detailed distribution of the types of social media used and the duration of use for each platform.

Fig. 1.

Types of social media used by participants and the duration of use for each platform each day.

On average, participants primarily used social media for entertainment purposes, with a mean±standard deviation EFUI score of 3.76±0.65. The mean SFUI score was 2.64±0.58, indicating that participants engaged in digital social interactions only occasionally. Overall, the intensity of social media use was between “occasionally” and “often,” with a mean SNAIS score of 2.96±0.51 (Table 2).

Table 2.

Descriptive Statistics for SNAIS Items and Participants’ Professional Identity Level Using MCPIS-9

Activity/statement	Mean±SD	Median	IQR
SNAIS items
Social function use intensity	2.64±0.58	2.8	1.1
1. Sent messages to friends on a message board	2.66±1.31	2	2.0
2. Chatted with friends via the instant messaging function	4.46±0.79	5	1.0
3. Replied to comments made by social networking friends	2.68±1.15.	3	1.0
4. Commented on friends’ status, logs, and photos	2.44±1.00	2	1.0
5. Shared/Forwarded content	3.15±1.10	3	2.0
6. Browsed others’ logs/photos/statuses/albums	2.82±1.06	3	0.0
7. Updated self-status	2.19±0.91	3	2.0
8. Posted photos/videos on personal web profile	2.34±0.94	2	1.0
9. Wrote logs/Weibo	1.21±0.61	3	0.0
10. Decorated personal web profile (changed image/contact information/privacy setting)	2.41±0.99	2	1.0
Entertainment function use intensity	3.76±0.65	4	1.5
11. Surfed entertainment/current news	3.76±1.02	4	2.0
12. Watched video/listened to music	4.41±0.85	5	1.0
13. Played games/applications	3.28±1.27	3	2.0
14. Bought/gave virtual goods (e.g., birthday gifts)	3.59±1.11	4	1.0
Total SNAIS average score	2.96±0.51	3.2	1.2
MCPIS-9
1. I feel like I am a member of this profession.	3.47±0.75	3	1
2. I feel that I have strong ties with members of this profession.	3.47±0.73	3	1
3. I am often ashamed to admit that I am studying for this profession.a)	3.95±0.99a)	4	2
4. I find myself making excuses for belonging to this profession.a)	3.47±0.86a)	3	1
5. I try to hide that I am studying to be a part of this profession.a)	3.76±1.07a)	2	2
6. I am pleased to belong to this profession	4.25±0.71	4	1
7. I can identify positively with the members of this profession.	3.92±0.68	4	0
8. Being a member of this profession is important to me.	4.16±0.73	4	1
9. I feel that I share characteristics with other members of the profession.	3.43±0.68	3	1
Total MCPIS-9 average score	3.76±0.44	3.8	0.56

SNAIS: Social Networking Activity Intensity Scale, MCPIS-9: 9-Item MacLeod Clark Professional Identity Scale, SD: Standard deviation, IQR: Interquartile range.

^a)Reversed score.

2. professional identity and social media engagement patterns

1) Professional identity and content engagement

Overall, the participants had a relatively high sense of belonging to the medical community (MCPIS-9 score=3.76±0.44). Participants scored the highest on items “I am pleased to belong to this profession” (4.25±0.71) and “Being a member of this profession is important to me” (4.16±0.73). In contrast, the participants scored the lowest on the statement “I feel I share characteristics with other members of the professions” (3.43±0.68) (Table 2).

The average CEI score was 0.51±0.64, with a median of 0.17 (IQR, 0.83). Browsing activities demonstrated the highest engagement intensity and contributed more to the overall CEI. In contrast, active content production behaviors, such as writing blogs or vlogs, showed the lowest engagement intensity and contributed less. Across all activities, engagement with educational or informational content constituted a smaller proportion of total engagement than other content types (Table 3).

Table 3.

Components of the CEI: Engagement Intensity and Content Type Across Social Media Activities

Variable	Engagement intensity	% Engagement		Engagement score	CEI
		Educational/informational content	Other contents
CEI					0.51±0.64
Activity
1. Commented content	2.44±1.00	19.96	80.04	0.44±0.99
2. Shared/forwarded content	3.15±1.10	14.86	85.14	0.41±1.07
3. Uploaded content	2.34±0.94	7.76	92.24	0.15±0.55
4. Written content (i.e., blog or vlog)	1.21±0.61	2.22	97.78	0.05±0.33
5. Browsed content (on friend’s social media)	2.82±1.06	29.93	70.07	0.81±1.38
6. Browsed content (public channel)	3.76±1.02	32.82	67.18	1.18±1.79

Data are presented as mean±standard deviation or %.

CEI: Content engagement index.

2) Associations between social media engagement and professional identity

A weak yet significant positive correlation was observed between the CEI and MCPIS-9 scores (ρ=0.139; 95% confidence interval [CI], 0.042–0.236; p=0.003). A similarly weak but significant positive correlation was found between the SNAIS and MCPIS-9 scores (r(449)=0.102; 95% CI, 0.009–0.190; p=0.030).

To further examine the differences in MCPIS-9 scores across social media use patterns, one-way ANOVA analysis was conducted. To account for platform-level multiplicity, p-values were adjusted using the Benjamini-Hochberg false discovery rate (FDR) procedure. After FDR correction, significant differences in MCPIS-9 scores were observed across duration-of-use groups for WhatsApp (F(2, 448)=5.377, p_adj=0.004, η²=0.023) and X (F(2, 448)=4.953, p_adj=0.003, η²=0.022), while no significant differences were found for other platforms (Table 4).

Table 4.

Average MCPIS-9 Score by Usage Duration of Each Social Media

Social media	MCPIS-9 by usage duration						One-way ANOVA
	<1 hr		1–3 hr		>3 hr		p-value	padj-value	F (df1, df2)	η²
	Mean±SD	No.	Mean±SD	No.	Mean±SD	No.
WhatsApp	3.69±0.45	215	3.83±0.44	137	3.83±0.41	99	0.0049	0.004	5.377 (2, 448)	0.023
Line	3.71±0.44	101	3.75±0.46	212	3.82±0.42	138	0.1352	0.216	2.010 (2, 448)	0.009
Telegram	3.76±0.44	442	3.87±0.53	7	3.95±0.08	2	0.6812	0.681	0.384 (2, 448)	0.002
TikTok	3.71±0.49	181	3.79±0.39	119	3.81±0.42	151	0.0732	0.162	2.631 (2, 448)	0.012
YouTube	3.79±0.46	181	3.77±0.38	171	3.71±0.51	99	0.3270	0.436	1.121 (2, 448)	0.005
Instagram	3.73±0.47	103	3.76±0.44	203	3.79±0.44	145	0.5965	0.682	0.517 (2, 448)	0.002
X	3.79±0.45	373	3.62±0.38	58	3.64±0.33	20	0.0075	0.028	4.953 (2, 448)	0.022
Facebook	3.76±0.44	448	3.73±0.90	3	-		0.9364	0.171	0.006 (1, 449)	0.000

Statistically significant results are marked in bold.

MCPIS-9: 9-Item MacLeod Clark Professional Identity Scale, ANOVA: Analysis of variance, SD: Standard deviation.

Specifically, participants who indicated that they used WhatsApp at least 1 hour per day had a significantly higher MCPIS-9 score (mean_{<1 hour}=3.69, SD_{>1 hour}=0.452; mean_{≥1 hour}=3.83, SD_{≥1 hour}=0.43). On the other hand, participants who used X less than 1 hour per day had a significantly higher MCPIS-9 score (mean_{<1 hour}=3.79, SD_{<1 hour}=0.453; mean_{≥1 hour}=3.62, SD_{≥1 hour}=0.368).

Discussion

Our findings revealed high-frequency, high-intensity social media use, with 94.2% of participants accessing social media at least 5 times daily and 30.6%–33.5% spending more than 3 hours per day on TikTok, Instagram, or Line. These findings are consistent with a previous study showing that most adolescents spend approximately 1 to 4 hours a day on social media [23]. However, unlike several studies that identified socializing with friends and relatives as the main reason behind social media use [23-25], our findings indicated that participants primarily used social media for leisure purposes, such as browsing the internet, playing games, watching movies, or listening to music. This is evident from the higher EFUI score relative to the SFUI score in this study, which was further highlighted by participants’ preference for engaging with entertainment content over educational, informational, or promotional posts.

Beyond leisure and entertainment, social media platforms such as TikTok and Instagram can also serve educational purposes, particularly as platforms for sharing knowledge and skills during learning activities [26]. Social media platforms also provide networking opportunities with a broader professional community and help users connect with peers who share similar interests and with other professionals who can serve as digital role models [17]. The digital interactions performed on social media (including commenting, sharing, forwarding, browsing, and posting content) are forms of participation in a digital community, which enable them to gain recognition as a member of a community and enact their identity, both of which are crucial parts in PI formation process [18].

Our data showed that participants’ social media engagement was predominantly passive, with browsing activities contributing most to content engagement, whereas active content production was less common. Such patterns indicate that students’ digital participation may offer exposure to professional content and communities but limited opportunities for active identity expression and reflective engagement. Engagement is a crucial component in PI formation because it fosters knowledge and skill development while deepening one’s understanding of professional norms, values, responsibilities, and expectations [14]. However, depending on the content and interactions, social media can either have a positive or negative impact on students’ PI formation. Excessive social media use for social, entertainment, or promotional purposes may encourage consumerism, narcissism, and hedonism, which can negatively affect identity formation [27,28]. Conversely, engagement with educational and informational content is associated with positive identity development, including enhanced communication skills, intercultural awareness, and cognitive abilities [13,29].

The weak yet significant associations between MCPIS-9 and both SNAIS and CEI further suggest that digital interactions play a supportive but secondary role in PI formation, with other educational and social influences likely exerting stronger effects [30]. This is reinforced by the differences in the MCPIS-9 scores across the social media platforms. Specifically, greater use of WhatsApp, which is commonly used for peer and academic communication [31], was associated with a higher MCPIS-9 score, whereas heavier use of X showed the opposite trend. This finding also possibly reflects differences in the nature and tone of interactions across platforms. Together, these findings highlight the importance of moving beyond mere exposure to social media toward more intentional, guided digital engagement that may better support PI formation in medical education.

A key limitation of this study is that the observed correlations were weak, although the associations between PI as measured by MCPIS-9 and both social media use intensity (SNAIS) and content engagement (CEI) were statistically significant. Given the relatively large sample size, statistical significance may have been amplified despite small effect sizes, indicating that social media engagement explains only a limited proportion of the variance in PI and that other factors likely play a more substantial role in PI formation. Nevertheless, the findings of this study indicate that intentional integration of professionally oriented digital interactions—such as guided engagement with educational or professional content—may offer incremental support to students’ PI formation when combined with formal curricular and clinical learning experiences.