Navigating digital resources in dental education: a cross-sectional comparison of the educational quality of rubber dam videos on Youtube and Instagram

Abstract

Background

Applying a rubber dam (RD) is fundamental in both endodontic and restorative procedures to ensure a sterile working area, achieve optimal moisture control, and minimize microbial contamination. While platforms like YouTube and Instagram have become go-to resources for dental education, the actual quality of their content needs closer inspection. Therefore, this study sets out to assess how accurate and reliable the information on RD application techniques is across YouTube videos and Instagram Reels.

Methods

Systematic searches for rubber dam application content were conducted on both platforms on January 18, 2026. After filtering, 39 relevant videos from each source were included. Three independent endodontists evaluated these videos using the Global Quality Scale (GQS), JAMA benchmarking criteria, and the modified Patient Educational Materials Evaluation Tool (mPEMAT). For statistical evaluation, Fleiss Kappa was used to check inter-rater reliability, Mann-Whitney U test was used to compare platforms, and Spearman correlation analysis (p < 0.05) was used.

Results

A strong consensus was observed among the observers; inter-rater reliability across all measures fell within the “good” to “excellent” categories (p < 0.001). YouTube significantly outperformed Instagram Reels, scoring higher on the GQS, JAMA, and mPEMAT scales (p < 0.001 for all). Content uploaded by dentists received the highest scores, regardless of the platform. Longer videos tended to have higher GQS and mPEMAT scores on both platforms.

Conclusion

In this limited cross-sectional study, YouTube videos exhibited higher instructional quality and reliability scores compared to Instagram Reels. Rather than establishing an absolute superiority of one platform over the other, it appears that YouTube currently provides a more comprehensive learning resource.

Introduction

Rubber dam (RD) application, introduced by Barnum in 1864, protects against instrument aspiration, reduces microbial contamination, and provides an operating area under sterile conditions [1]. While the necessity of using RDs for treatment that meets gold standards in endodontics is clearly evident, it is equally critical in restorative dentistry for achieving strict moisture control and ensuring the long-term success of adhesive protocols [2, 3]. This practice not only enhances clinical success but also forms the foundation of preventive care by ensuring patient safety during dental procedures [4]. The clinical importance and application of RDs in dental undergraduate education are taught in terms of their contributions to visibility, accessibility, isolation, and retraction [5]. Textbooks, brochures, presentations, and other text- and still image-based materials may be insufficient for conveying the rationale for practical procedures to students who receive training in laboratory applications prior to clinical treatment [6].

Digitalization has increased the popularity of social media, making it the preferred choice of users in various fields, including education [7]. Instagram stands out among popular social media platforms in terms of the user growth rate, with the number of users estimated to reach 1.5 billion by 2026 [8, 9]. The video-sharing network YouTube has made rapid progress since 2005, reaching over 2 billion daily views by the end of 2005 and doubling that number after October 2009. Given their popularity and accessibility, YouTube videos and Instagram Reels can be integrated into modern teaching frameworks to enhance learning; however, a critical analysis of their digital content quality remains essential [10, 11].

Despite the growing literature evaluating digital platforms in dental education, there is a significant research gap regarding the multidimensional assessment of these visual resources. While previous studies have largely focused on general information quality, few have delved deeply into the concepts of ‘understandability’ and ‘applicability’ from a procedural learning perspective. To address this gap in the literature, the present study aims to comprehensively assess these educational videos by combining established reliability tools (GQS and JAMA) with the modified PEMAT (mPEMAT) scale. The null hypothesis (H0) of this study is that there is no significant difference in terms of the quality, reliability, comprehensibility, and applicability scores of rubber dam application videos between YouTube and Instagram platforms and between different sharing sources.

Methods

Data collection

To ensure high reproducibility of the methodology, the search strategy and parameters were precisely defined. To eliminate temporal variations arising from the dynamic nature of social media algorithms and to capture a real-time snapshot, the data collection process was executed as a single cross-sectional iteration on January 18, 2026. To minimize selection bias from personalized algorithms, YouTube searches were performed in an unauthenticated session using only the search string “rubber dam application,” after browser history and cookies were completely cleared [12]. Due to Instagram’s closed architecture, which prohibits unauthenticated searches, a fake account with no prior search or viewing history was used, and the search was strictly limited to the English hashtags #rubberdam and #rubberdamapplication. Furthermore, to prevent any algorithmic discrepancies caused by different IP addresses, device-specific configurations, or geographic tracking, all searches across both platforms were executed consecutively from a single computer (Windows 11 Home Single Language OS, Google Chrome browser) located in Kirikkale, Turkey. No additional filters were applied, allowing the platforms’ native algorithms to naturally rank visibility [13].

Initially, a predetermined sample size of approximately 300 videos was planned, aiming for 85% statistical power to detect an effect size of 0.35. Videos listed in the search results were evaluated sequentially, starting from the top. However, a methodological limitation emerged on Instagram; content under the relevant hashtags became completely irrelevant after the 100th video. Due to this practical limitation, data collection was stopped; to ensure statistical balance and comparative fairness, the top 100 videos with the highest visibility from each platform were selected for screening.

From this pool, videos were excluded if they did not directly demonstrate clinical rubber dam application, lacked clear English audio, subtitles, or text descriptions, or contained repetitive content. Conversely, videos were strictly included in the final analysis only if they were in English and clearly demonstrated key clinical steps specific to rubber dam application, such as equipment introduction, dam perforation, single- or two-stage placement techniques, inversion, flossing, and safe removal procedures.

In the sample selection process, after applying exclusion criteria, exactly 39 videos from the top 100 videos screened on Instagram were found suitable for analysis. Although a significantly larger data pool was available on YouTube, the sample size was limited to match the Instagram data to ensure statistical balance, robustness against potential heterogeneity of variance, and comparative fairness in the cross-platform analysis [14]. Accordingly, the top 39 videos that ranked highest in YouTube search results and met the study criteria were identified and included in the final analysis (Fig. 1).

Fig. 1.

Flowchart of the video selection and screening process

Reels and videos were examined in terms of duration, likes, comments, reposts, days since initial posting, and views. The number of people followed by the sharing sources, their followers, and the number of posts and videos they shared were recorded. The view rates of the videos were calculated as views/days since upload × 100 [15].

Categorization of sharing sources

Content creators have been divided into four main categories based on their professional qualifications: dentists; health-related businesses (dental equipment manufacturers, continuing education platforms, and corporate healthcare agencies, etc.); private hospitals/university hospitals/clinics; and others (dental supply stores, commercial websites, or individual users) [15]. This classification was created to analyze the differences in expertise levels between institutions and individuals regarding information shared about rubber dam applications.

Measuring video quality

The reviewers evaluated the educational quality by focusing on how thoroughly and accurately the content creators executed the clinical sequence previously defined in the inclusion criteria.

Video quality was assessed by three endodontists specializing in the field via the Global Quality Scale (GQS), the Journal of the American Medical Association (JAMA) benchmark criteria, and the modified Patient Education Materials Assessment Tool (mPEMAT).

The GQS consists of five elements that define the quality, accuracy, and usefulness of the content [16]. Videos were scored from a minimum of 1 to a maximum of 5 on the basis of their quality (Table 1).

Table 1.

Global quality scale

Score	Description
1	Poor quality poor flow, most information missing, not useful for education
2	Generally, poor quality and flow of limited use to patients because only some information is present but many important topics missing
3	Moderate quality, suboptimal flow, somewhat useful for patients as some important information is adequately discussed but others poorly discussed
4	Good quality generally good flow useful to patients because most relevant information is covered but some topics not covered
5	Excellent quality and flow, highly useful to patients

The JAMA benchmark criteria aim to evaluate the reliability and accuracy of sources via authorship, citation, explanation, and timeliness criteria [17]. Each item was assigned a score, and the total score ranged from 0 to 4 (Table 2).

Table 2.

Journal of the American Medical Association (JAMA) benchmark criteria

Criteria	Description
1. Authorship	Authors partnerships and contributors, their affiliations and relevant credentials should be provided
2. Attribution	References and resources used for the content and a copyright statement should be listed clearly
3. Currency	Sponsorship, advertising commercial financing potential conflicts of interest should be prominently and fully disclosed
4. Disclosure	Dates when content was posted and updated should be indicated

PEMAT ensures that written, visual, and audio materials are evaluated in terms of comprehensibility and actionability. It demonstrates how well the material enables consumers with diverse backgrounds and varying literacy levels to process and explain the core messages and understand what they can do with the information provided [18]. In this study, we used the visual and auditory material comprehensibility table of the modified PEMAT prepared by the Minnesota Center of Excellence in Newcomer Health and Nationalities Service Center [19]. This study aimed to evaluate the quality of information accessible to dental students on video sharing platforms; therefore, the items were interpreted in a way that would evaluate material quality from the students’ perspective. The relevant items received scores between 0 and 1, whereas irrelevant items 5, 6, 9, and 10 were not included in the scoring. Eight of the 12 items were scored. The score was calculated as “(total score / maximum possible score) × 100” ((total score / 8) × 100) (Table 3).

Table 3.

Modified Patient Education Materials Assessment Tool (Audio / Visual Understandability)

Topic	Item #	Item	Response Options
Content	1	The material makes its purpose completely evident.	Disagree = 0, Agree = 1
Word Choice & Style	2	The material uses common, everyday language.	Disagree = 0, Agree = 1
	3	Medical terms are used only to familiarize audience with the terms. When used, medical terms are defined.	Disagree = 0, Agree = 1
Organization	4	The material breaks or “chunks” information into short sections.	Disagree = 0, Agree = 1
Layout & Design	5	The material uses visual cues (e.g., arrows, boxes, bullets, bold, larger font, highlighting) to draw attention to key points.	Video = N/A
	6	Text on the screen is easy to read.	No text or all text is narrated = N/A
	7	The material allows the user to hear the words clearly (e.g., not too fast, not garbled).	Disagree = 0, Agree = 1
Use of Visual Aids	8	The material uses illustrations and photographs that are clear and uncluttered.	Disagree = 0, Agree = 1
	9	The material uses simple tables with short and clear row and column headings.	No tables = N/A
	10	Visual aids are not culturally specific, thus can be understood across cultures.	No visual aids = N/A
Actionability	11	The material clearly identifies an action the user can take.	Disagree = 0, Agree = 1
	12	The material breaks down any action into manageable, explicit steps.	Disagree = 0, Agree = 1

Statistical analysis

The normality of the variables was examined via the Shapiro-Wilk test, which revealed that the data did not meet the parametric assumptions. Accordingly, the Mann-Whitney U test, a nonparametric method, was used to evaluate the differences between two independent groups. The direction and strength of the relationships between the variables were determined via Spearman correlation analysis (rho). The correlation coefficient (r) was classified as follows: 0.00–0.19 very weak, 0.20–0.39 weak, 0.40–0.59 moderate, 0.60–0.79 strong, and 0.80–1.00 very strong. The Fleiss’ Kappa coefficient was calculated to determine the level of interrater agreement. To interpreting the kappa values, the following criteria were used: ≤0.20 is poor, 0.21–0.40 is fair, 0.41–0.60 is moderate, 0.61–0.80 is good, and 0.81–1.00 is excellent agreement. For all analyses, the statistical significance level was set at p < 0.05. To control for Type I error inflation (false positives) arising from the simultaneous evaluation of three distinct dependent variable scales (GQS, JAMA, mPEMAT), a Bonferroni correction was applied, adjusting the statistical significance threshold to p < 0.016 (0.05 / 3).

Results

The characteristic features of all included videos and descriptive data of the content sources were recorded on the same date. (Table 4).

Table 4.

Characteristics and scores of Instagram and YouTube videos and their sharing sources

	Instagram		YouTube
	Mean ± SD	Min - Max	Mean ± SD	Min – Max
Duration (s)	52.74 ± 59.40	2–338	453.85 ± 601.06	9–3415
Like	2014.00 ± 9411.76	0–54,488	1683.36 ± 3126.68	0–15,000
Comments	6.41 ± 14.59	0–73	36.87 ± 60.26	0–262
Reposts	106.92 ± 427.41	0–2464	-	-
Day Since Upload	700.00 ± 287.28	360–1350	1676.92 ± 176.20	180–5040
View	25123.18 ± 67250.05	30–333,733	274036.44 ± 956762.13	1316–5,882,430
Viewing Rate	4227.77 ± 12126.02	7.14–65437.84	16446.80 ± 53708.83	121.85–326801.67
Following	1178.36 ± 1727.28	12–7502	-	-
Followers	61884.92 ± 175883.22	459–948,000	65712.92 ± 97468.64	271–438,000
Shared Videos	389.44 ± 909.72	9–4276	256.97 ± 300.16	5–1400
Shared Posts	985.26 ± 1737.01	33–7827	-	-
GQS	2.23 ± 1.25	1–4	4.59 ± 0.75	2–5
JAMA	1.03 ± 0.16	1–2	2.80 ± 0.52	2–4
PEMAT (%)	16.99 ± 9.72	12.50–50	51.60 ± 16.01	25–87.50

The GQS, JAMA, and mPEMAT scores of the videos were evaluated blindly by three independent endodontists. Inter-rater reliability analysis revealed statistically significant Fleiss Kappa coefficients for all scales (p < 0.001), exhibiting values within the ‘good’ to ‘excellent’ agreement range (Table 5). However, while there was complete absolute agreement between observers for the JAMA scores on the Instagram platform (p = 1.000), the corresponding p-value remained above the significance threshold (p = 0.075). This statistically significant situation is explained by the ‘prevalence paradox,’ which is also influenced by the relatively small sample size (n = 39). In cases where a single score is excessively dense and there are a limited number of observations, the Kappa statistic may not reach its traditional level of significance because it becomes mathematically difficult to distinguish perfect agreement from random agreement [20]. Based on this high level of consistency achieved among observers, final scores determined by the consensus of experts were used instead of individual scores in the later stages of the analysis.

Table 5.

Interrater reliability

	Score	Mean ± SD.			K	p
		S1	S2	S3
INSTAGRAM	GQS	2.28 ± 1.21	2.18 ± 1.12	2.21 ± 1.15	0.717	< 0.001
	JAMA	1.03 ± 0.16	1.03 ± 0.16	1.03 ± 0.16	1	0.075
	PEMAT	1.38 ± 0.78	1.44 ± 0.79	1.46 ± 0.82	0.735	< 0.001
YOUTUBE	GQS	4.51 ± 0.76	4.56 ± 0.79	4.49 ± 0.82	0.726	< 0.001
	JAMA	2.79 ± 0.52	2.79 ± 0.52	2.79 ± 0.52	1	< 0.001
	PEMAT	4.08 ± 1.24	4.10 ± 1.23	4.18 ± 1.30	0.733	< 0.001

* S: Specialist. The means of the 1 and 0 scores derived from the PEMAT criteria are shown in the table

When comparing the quality and educational material characteristics of videos shared on Instagram and YouTube platforms, a statistically significant difference was found between the platforms in terms of the GQS, JAMA, and mPEMAT scores (p < 0.001 for all scales). The analyses revealed that the YouTube videos had significantly higher scores than did the Instagram Reels content on all three scales (Table 6).

Table 6.

Comparison of Instagram Reels and YouTube video scores

Score	Instagram		Youtube
	Median (Q1-Q3)	Mean Rank	Median (Q1-Q3)	Mean Rank	r**	p
GQS	2 (1–3)	22.31	5 (4–5)	56.69	0.79	< 0.001*
JAMA	1 (1–1)	20.13	3 (2–3)	58.87	0.93	< 0.001*
PEMAT	12.5 (12.5–12.5)	21.53	50 (37.5–62.5)	57.47	0.82	< 0.001*

*Statistically significant at the Bonferroni-corrected significance level of p < 0.016

**r: Effect size calculated using the Z-statistic from the Mann-Whitney U test

Videos were grouped according to their sharing sources. GQS scores for videos shared by dentists (p < 0.001), health-related organizations (p = 0.006), and others (p = 0.008) differed across platforms; no differences were found across platforms for videos shared by hospitals/universities/clinics (p = 0.057). Dentists, health-related organizations, and others also showed significant differences in JAMA (p < 0.001, p = 0.006, p = 0.229, p < 0.001, respectively) scores for dentists, health-related organizations, and other sharing sources, whereas hospital/university/clinic sharing sources had similar scores on both platforms (Table 7).

Table 7.

Comparison of sharing resources across platforms

	GQS
	INSTAGRAM		YOUTUBE
	Median (Q1-Q3)	Mean Rank	Median (Q1-Q3)	Mean Rank	r**	p
Dentists (nı=17, ny=23)	2 (1–3)	9.88	5 (5–5)	28.35	0.83	< 0.001*
Health Related Business (nı=9, ny=4)	3 (1-3.5)	5.11	5 (4.25-5)	11.25	0.75	0.006*
Hospital/University/Clinics (nı=4, ny=3)	2 (1-3.75)	2.63	5 (4–5)	5.83	0.76	0.057
Others (nı=9, ny=9)	2 (1–4)	6.28	4(3.5-5)	12.72	0.62	0.008*

	JAMA
	INSTAGRAM		YOUTUBE
	Median (Q1-Q3)	Mean Rank	Median (Q1-Q3)	Mean Rank	r**	p
Dentists (nı=17, ny=23)	1 (1–1)	9.00	3 (3–3)	29.00	0.95	< 0.001*
Health Related Business (nı=9, ny=4)	1 (1–1)	5.00	3 (2.5-3)	11.50	0.95	0.003*
Hospital/University/Clinics (nı=4, ny=3)	1 (1-1.75)	2.63	3 (2–3)	5.83	0.78	0.057
Others (nı=9, ny=9)	1(1–1)	5.00	2 (2-2.5)	14.00	0.93	< 0.001*

	PEMAT
	INSTAGRAM		YOUTUBE
	Median (Q1-Q3)	Mean Rank	Median (Q1-Q3)	Mean Rank	r**	p
Dentist (nı=17, ny=23)	1 (1-1.5)	9.56	62.5 (50-62.5)	28.59	0.83	< 0.001*
Health Related Business (nı=9, ny=4)	12.5 (12.5–25)	5.17	50 (28.13–62.5)	11.13	0.76	0.006*
Hospital/University/Clinics (nı=4, ny=3)	25 (12.5-46.88)	3.00	62.5 (25–75)	5.33	0.54	0.229
Others (nı=9, ny=9)	12.5 (12.5–12.5)	5.00	50 (25–50)	14.00	0.91	< 0.001*

*Statistically significant at the Bonferroni-corrected significance level of p < 0.016

**r: Effect size calculated using the Z-statistic from the Mann-Whitney U test

The correlations between Instagram video scores and video characteristics and source information were evaluated. A strong positive correlation was found between video duration and the GQS (p < 0.001), whereas a moderate positive correlation was found between video duration and the PEMAT score (p < 0.001). Additionally, a weak positive correlation was observed between the GQS score and PEMAT (p = 0.02) and between the JAMA score and PEMAT (p = 0.02). No statistically significant correlation was found between the scores of the other videos and the sharing of source information (Table 8).

Table 8.

Correlations between Instagram scores and video characteristics

	Duration	GQS	JAMA	PEMAT
Duration	-	r = 0.61** p < 0.001	r = 0.27 p = 0.09	r = 0.54** p < 0.001
GQS	r = 0.61** p < 0.001	-	r = 0.23 p = 0.16	r = 0.36* p = 0.02
JAMA	r = 0.27 p = 0.09	r = 0.23 p = 0.16	-	r = 0.36* p = 0.02
PEMAT	r = 0.54** p < 0.001	r = 0.36* p = 0.02	r = 0.36* p = 0.02	-

* Correlation is significant at the 0.05 level

** Correlation is significant at the 0.01 level

The correlations of YouTube scores with video characteristics and source information were evaluated. A moderate positive correlation was observed between duration and comments, GQS, JAMA, and PEMAT scores (p = 0.01, p < 0.001, p = 0.001, p = 0.004, respectively). There was a moderate positive correlation between the number of comments and the viewing rate (p < 0.001), whereas a weak positive correlation was found between the JAMA score and the viewing rate (p = 0.04). A moderate positive correlation was found between the number of days from the first day the videos were shared to the present and the JAMA score, and a weak positive correlation was found with the PEMAT score. The only negative correlation was weak between the viewing rate and GQS (p = 0.03). Among the scores, GQS and JAMA were moderately positively correlated (p < 0.001); a strong positive correlation was found between GQS and PEMAT (p < 0.001) and between JAMA and PEMAT (p < 0.001). No correlation was found between other video characteristics and sharing source information or scores (Table 9).

Table 9.

Correlation of YouTube scores with video characteristics

	Duration	Comments	Day Since Upload	Viewing Rate	GQS	JAMA	PEMAT
Duration	-	r = 0.40* p = 0.01	r = 0.07 p = 0.68	r = -0.06 p = 0.72	r = 0.55** p < 0.001	r = 0.50** p = 0.001	r = 0.45** p = 0.004
Comments	r = 0.40* p = 0.01	-	r = 0.29 p = 0.07	r = 0.56** p < 0.001	r = 0.22 p = 0.17	r = 0.34* p = 0.04	r = 0.23 p = 0.17
Day Since Upload	r = 0.07 p = 0.68	r = 0.29 p = 0.07	-	r = 0.01 p = 0.93	r = 0.28 p = 0.09	r = 0.45** p = 0.004	r = 0.32* p = 0.048
Viewing Rate	r = -0.06 p = 0.72	r = 0.56** p < 0.001	r = 0.01 p = 0.93	-	r = -0.34* p = 0.03	r = -0.10 p = 0.55	r = -0.09 p = 0.57
GQS	r = 0.55** p < 0.001	r = 0.22 p = 0.17	r = 0.28 p = 0.09	r = -0.34* p = 0.03	-	r = 0.54** p < 0.001	r = 0.69** p < 0.001
JAMA	r = 0.50** p = 0.001	r = 0.34* p = 0.04	r = 0.45** p = 0.004	r = -0.10 p = 0.55	r = 0.54** p < 0.001	-	r = 0.61** p < 0.001
PEMAT	r = 0.45** p = 0.004	r = 0.23 p = 0.17	r = 0.32* p = 0.048	r = -0.09 p = 0.57	r = 0.69** p < 0.001	r = 0.61** p < 0.001	-

* Correlation is significant at the 0.05 level

** Correlation is significant at the 0.01 level

Discussion

Recent technological developments have led to a digital transformation in educational materials, with innovative approaches replacing traditional methods. Video sharing platforms are considered a primary source of information among the younger generation because of their ability to provide quick and easy access to knowledge. Social media is widely used and viewed as a material that can support education by dental students [21]. Digital 2026: Turkiye data show that YouTube ranks first in user preference in terms of information and entertainment-based content, while Instagram ranks first for visual interaction-focused sharing [22]. This study aimed to evaluate the quality and scientific accuracy of videos shared on these platforms and containing content related to “rubber dam application” via objective criteria.

GQS, which is used in the evaluation of health-related websites, has recently become one of the methods used to evaluate the quality of digital content on online platforms [16, 23]. While the GQS generally evaluates content quality, the JAMA assesses reliability and transparency (authorship, citations, recency, etc.) [24]. The PEMAT is a tool developed to evaluate patient education materials [25]. However, the “Rubber Dam application,” which is the focus of the current study, is a clinical procedure performed directly by a dentist. In this context, the PEMAT scale could be adapted to this group to determine the levels of ‘understandability’ and ‘applicability’ of the videos used by dental students in their clinical training process, and the scale was included in the study protocol.

Educating individuals who lack medical terminology and specific field knowledge and ensuring that knowledge is translated into practice is a complex process. PEMAT is a critical tool that provides evidence-based data for the evaluation and development of such materials [26–28]. In this study, the ability of the PEMAT criteria to simplify complex procedures was simulated through the ‘learner’ role of dental students who have not yet achieved clinical competence. Although videos on platforms receive high scores for comprehensibility, the lack of full competence persists because of low information quality and reliability [29, 30]. Chen et al., who compared different platforms, reported that YouTube had a higher PEMAT score than other platforms did [31]. Studies evaluating Instagram alone reported high PEMAT comprehensibility scores, whereas other studies comparing it with other platforms reported that YouTube lagged behind [32–34]. Our results revealed that YouTube and Instagram had lower scores than did compared to similar studies in the literature; however, the better performance of YouTube compared with Instagram is consistent with the literature. On Instagram, the fast-paced delivery of information in short Reels makes comprehension difficult, particularly because the video durations are significantly shorter than the actual clinical time required for rubber dam application. Insufficient video duration may restrict the comprehensive demonstration of the procedure, potentially affecting both comprehensibility and actionability.

Videos on dentistry topics have generally performed poorly in studies evaluating information quality and accuracy, with low GQS and JAMA scores on Instagram and YouTube [35–39]. The findings of our current study align with this trend of low performance in the literature, specifically with respect to Instagram Reels content; however, YouTube videos demonstrated higher success in GQS scoring than did previous studies [15, 40]. In this study, there was also a difference between platforms, with YouTube videos achieving the highest value in both scores. The higher GQS and JAMA scores on YouTube than those reported in studies in the literature on dental topics can be attributed to the adequacy of the sharing sources regarding “rubber dam application” and the improvement in the quality of information by sharing sources to attract the attention of the audience each year [41, 42]. The lack of positive increases across all the scores examined on Instagram can be attributed to the platform being used more for social interaction than information sharing, and to target audiences tending to consume real-life videos for quick visual engagement rather than in-depth academic learning. Consequently, it is crucial to acknowledge that these score disparities were likely influenced by unmeasured variables, such as video production quality and the presenter’s communication skills. Furthermore, since the videos evaluated were selected based on search visibility rather than random sampling, platform-specific algorithmic biases undoubtedly shaped the sampling. Therefore, these findings reflect a broader digital ecosystem where algorithms, creator intent, and audience expectations collectively determine the educational value of medical content.

Video sharing platforms offer a wide range of content across various disciplines owing to their heterogeneous uploader structure. Recent studies evaluating health videos on different platforms have shown that quality scores and reliability are higher for content shared by experts in the field or academic institutions [39, 43, 44]. One of the notable findings of this study is that videos uploaded by individual dentists received higher quality scores than those from academic sources. This is thought to be due to a significant disparity in sample sizes between the two sources of content sharing. The fact that content produced by individual dentists is far more numerous than that from academic institutions naturally increases the likelihood of encountering high-quality videos within this larger pool.

In the literature, studies have reported that YouTube video GQS scores increase in direct proportion to duration [24, 45]. Rachmawati et al. reported that video quality scores for videos longer than 6 min were higher than those for videos shorter than 6 min [46]. The findings of this study are consistent with these data; both the GQS and PEMAT scores are positively correlated with video duration on both platforms. The number of studies evaluating the correlation between video characteristics and Instagram quality scores is limited, with some studies reporting a positive correlation and others reporting no correlation [15, 47]. Video duration appears to be a strongly associated factor with information quality in terms of conveying technical details and content depth. In this study, Instagram Reels videos were found to last a minimum of 2 s, whereas this duration was significantly greater for YouTube videos. The long video format not only increases the sharing of theoretical knowledge but also provides advantages in terms of the detailed demonstration of different clinical techniques and the capacity to guide the viewer. Therefore, the lower GQS scores on Instagram are likely intertwined with the severe time constraints inherent in the Reels format.

The main limitation of this study is the limited sample size (n = 39 per platform). Because completely irrelevant content was obtained after the first 100 search results on Instagram, only 39 videos met the inclusion criteria. To maintain statistical balance and ensure a fair comparison of high-visibility content, the YouTube sample was matched to 39. Consequently, while this allows for direct comparison of top-ranking results, our findings do not fully represent the large number of educational videos found on YouTube. Furthermore, the scope was limited to YouTube and Instagram; another methodological limitation is that the modified PEMAT (mPEMAT) scale used in the assessment has not undergone formal validation in the field of professional clinical education.

Conclusion

The findings of this cross-sectional study indicate that YouTube videos exhibit higher scores in terms of overall quality, reliability, and comprehensibility compared to Instagram Reels. However, due to the substantial variability in content quality across both platforms, social media should be viewed as a supplementary tool rather than a definitive guide. To safely harness the potential of digital learning, it is essential for academic institutions to foster critical evaluation skills among dental students and for experts to actively contribute evidence-based clinical resources to these platforms.

Authors’ contributions

Conceptualization: EC; Methodology: EC; Validation: EC, DH, BY; Formal Analysis: EC; Investigation: EC, DH, BY; Resources: EC, BY; Data Curation: EC; Writing - Original Draft: EC, DH; Writing - Review & Editing: DH; Visualization: EC, BY; Supervision: DH; Project Administration: EC, DH, BY.

Funding

There is no funding.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study comprises only observational analysis of publicly available educational videos on YouTube and Instagram. No human participants or animal subjects were included in this research; no specific, clinically, or personally identifiable data were collected or analyzed. Therefore, in accordance with standard research guidelines regarding the use of publicly available data, formal ethical approval and informed consent were not required for this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.