A systematic search was conducted across seven electronic databases (Web of Science Core Collection, PubMed, SPORTDiscus, SCOPUS, Cochrane library, CINAHL, and Embase) to locate relevant studies from their inception until 10 December 2024. The selected databases were chosen for their extensive coverage of topics in physical activity, sports science, and related fields, as well as their access to high-quality sources. Previous reviews and meta-analyses (Slimani et al., 2018; Hughes et al., 2023; Ngo et al., 2024) were used to help define our search strategy. The Boolean search strategy utilized the operators “AND” and “OR” to combine and refine search terms: (danc* OR ballet OR salsa OR ballroom OR “hip-hop” OR jazz OR tap OR cha-cha OR rumba OR samba OR flamenco OR tango OR waltz OR folk) AND (“strength training” OR “resistance training” OR “weight training” OR “power training” OR “plyometric training” OR “complex training” OR “compound training” OR “neuromuscular training”). Following the formal systematic searches, additional hand-searches were conducted using Google Scholar and by reviewing the reference lists of included studies to identify relevant works. The search strings for each database are provided in Supplementary Appendix S1.

The present meta-analysis included studies that followed the PICOS framework (Liberati et al., 2009). The inclusion criteria were as follows: (a) participants were dancers of any age, sex, or dance genre; (b) studies involved strength training programs lasting more than 2 weeks; (c) included an active control group; (d) measured at least one physical fitness parameter (e.g., power, strength, flexibility) before and after the intervention; (e) were controlled trials published in peer-reviewed journals; and (f) were written in English to avoid translation challenges.

Studies were excluded if they met the following criteria: (a) lacked original data (e.g., protocols, reviews, or patents); (b) included dancers with specific injuries (e.g., knee ligament injuries); (c) did not use a control group to assess the impact of strength training; (d) failed to report outcomes related to physical fitness; or (e) integrated strength training programs with psychological methods (e.g., self-talk).

Two independent reviewers (YY, ND) assessed potentially relevant articles for eligibility by reviewing titles, abstracts, and full texts, based on the established inclusion and exclusion criteria (Table 1). If YY and ND disagreed on the inclusion or exclusion of an article, a third author (XY) was consulted for a final decision.

The first reviewer (YY) extracted data from the selected studies using a standardized form developed in Microsoft Excel, while the second reviewer (ND) independently cross-checked all the extracted information to ensure accuracy. In cases where discrepancies arose between the two reviewers, the study information was revisited for verification. The extracted data included the following: (1) general publication details (e.g., author names, publication year); (2) participant information (e.g., sample size, age, dance genre); (3) intervention specifics (e.g., duration, frequency, session length); and (4) outcome variables (e.g., power, strength, flexibility). For the meta-analysis, only original research papers that provided data compatible for calculations and used standardized outcome measures were selected.

The risk of bias for each randomized controlled trial (RCT) included in the study was carefully assessed using the most recent version of the Cochrane risk of bias tool for randomized trials (RoB-2) (Flemyng et al., 2023). For non-RCTs, the evaluation was conducted using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool (Sterne et al., 2016). Two independent reviewers (YY, ND) conducted the assessments. In instances where the reviewers disagreed, a consensus meeting was convened, and a third assessor (XY) provided an additional rating to resolve the differences and reach an agreement.

All statistical analyses were conducted using Comprehensive Meta-Analysis software (Version 3.0; Biostat, Englewood, NJ, USA), which facilitated tasks such as the calculation of ESs, heterogeneity assessments, and the generation of forest plots. A meta-analysis was conducted when the outcome measure included sufficient data (at least three studies) (Deng et al., 2023; Deng et al., 2024). Many of the included studies had small sample sizes (fewer than 20 participants), necessitating an adjustment for small sample calculations (Anderson et al., 2017). Consequently, Hedges’ g ESs were calculated using pre- and post-intervention data to evaluate performance outcomes in both the experimental and control groups. In studies with multiple intervention groups, the control group’s sample size was proportionally divided to facilitate comparisons among all subjects (Higgins et al., 2008). The meta-analysis employed a random effects model for continuous data, incorporating inverse variance and a 95% confidence interval. The ES values obtained were classified according to the following scale: trivial (ES < 0.2), small (0.2 ≤ ES ≤ 0.6), moderate (0.6 < ES ≤ 1.2), large (1.2 < ES ≤ 2.0), very large (2.0 < ES ≤ 4.0), and extremely large (ES > 4.0) (Hopkins et al., 2009). To assess heterogeneity among studies, the I² statistic was used, expressed as a percentage ranging from 0 to 100, indicating the extent of variability attributable to heterogeneity rather than random chance. Values below 25% signified low heterogeneity, those between 25% and 75% indicated moderate heterogeneity, and values exceeding 75% suggested high heterogeneity (Higgins and Thompson, 2002). The extended Egger’s test was conducted to assess the potential risk of publication bias among the studies (Egger et al., 1997), with a significance threshold set at p < 0.05 to establish statistical relevance.

As shown in Figure 1, the initial database search across multiple platforms identified 1,347 documents, supplemented by 23 studies found through manual searches of reference lists and Google Scholar. After removing duplicates, 875 unique records remained for initial screening. Titles and abstracts of these records were reviewed, narrowing the selection to 170 articles for full-text assessment. Each article was carefully evaluated against the inclusion criteria, leading to the exclusion of 155 studies due to issues such as inadequate study design, irrelevant outcomes, or insufficient data. Consequently, 15 studies met the inclusion criteria, with 13 included in the final meta-analysis.

Twelve RCTs were assessed using the RoB-2 tool, while three non-RCTs were evaluated with the ROBINS-I framework. As depicted in Figures 2, 3, 13 studies presented a moderate risk of bias or some concerns, with only two categorized as having a low risk of bias. Figure 2 provides an overview of the RoB-2 assessment results. Of the RCTs, only three detailed their randomization sequence generation methods, whereas the remaining eleven did not provide sufficient descriptions of their randomization procedures. Three RCTs exhibited a high risk of bias and some concerns due to missing data, with dropout rates of approximately 10% and over 15%, respectively. Three RCTs raised concerns regarding the selection of reported outcomes, while two others had bias related to outcome measurements. Figure 3 visually represents the ROBINS-I findings. One non-RCT showed a moderate risk of bias due to selection bias, while two others had moderate risk stemming from outcome measurement and outcome reporting biases.

Table 2 provides a detailed overview of the primary characteristics of both the participants and the interventions involved in the studies. The studies included a total of 351 dancers from various styles, including ballet, contemporary, modern, jazz, ballroom, and hip-hop, with participants aged between 11 and 27 years. Among these studies, ten focused exclusively on females (Stalder et al., 1990; Koutedakis and Sharp, 2004; Brown et al., 2007; Vetter and Dorgo, 2009; Wyon et al., 2013; Sozbir et al., 2014; Skopal et al., 2020; De Leonardis and Greco, 2020; Sanders et al., 2020; Cofré-Fernández et al., 2023), while five assessed both genders (Koutedakis et al., 2007; Roussel et al., 2014; Zhang et al., 2021; Ávila-Carvalho et al., 2022; Zhao, 2023). The training modalities identified in these studies included lower limb strength training (Ávila-Carvalho et al., 2022; Zhao, 2023), resistance training (Koutedakis and Sharp, 2004; Sanders et al., 2020; Vetter and Dorgo, 2009), plyometric training (Brown et al., 2007; De Leonardis and Greco, 2020), weight training (Stalder et al., 1990; Brown et al., 2007), neuromuscular training (Zhang et al., 2021; Cofré-Fernández et al., 2023), and combined training (Koutedakis et al., 2007; Skopal et al., 2020; Roussel et al., 2014). The training interventions lasted between 6 and 16 weeks, with participants training two to three times per week. Each session ranged from 20 to 75 min. In terms of exercise volume, the interventions included between 2 and 10 sets per exercise, with each set consisting of 5–20 repetitions.

Notably, only two studies investigated the effects of neuromuscular training on balance in dancers, and thus, they were excluded from the meta-analysis. Zhang et al. (2021) implemented a 10-week neuromuscular training program with three sessions per week, which significantly improved balance performance in 42 elite collegiate ballroom dancers (aged 17–23 years). The neuromuscular training group scored significantly higher on the post-test (392 ± 36) compared to the control group (329 ± 23). Similarly, Cofré-Fernández et al. (2023) evaluated the effects of a 6-week neuromuscular training program, also comprising three sessions per week, on balance in 14 female competitive hip-hop and jazz dancers (aged 12–20 years). The experimental group exhibited significant improvements in Y-balance test scores (p = 0.01–0.02; ES = 0.58–0.63).

Dancers need to maintain optimal body composition to perform repetitive, high-quality, and physically demanding movements effectively (Wilmerding et al., 2005). However, the current study found no significant reductions in body mass or body fat percentage resulting from strength training. Most studies reported no changes in body mass or body fat percentage after resistance, plyometric, weight, or combined training (Koutedakis and Sharp, 2004; Brown et al., 2007; Roussel et al., 2014; Sozbir et al., 2014; De Leonardis and Greco, 2020; Sanders et al., 2020; Ávila-Carvalho et al., 2022). These findings align with a prior systematic review by Zouita et al. (2023), which concluded that resistance training does not effectively improve body composition in female athletes. In contrast, some studies suggest that strength training can be as effective as aerobic exercise in reducing body mass and body fat percentage (Carvalho et al., 2014; Wewege et al., 2022). Strength training promotes energy expenditure through muscle growth, which can aid in fat loss (Westcott, 2012).

The literature suggests that various factors can influence body composition, such as diet, nutrition, sleep, and stress (Moon, 2013; Poggiogalle et al., 2016; Stefanaki et al., 2018). Notably, experts have indicated the significant role of dietary habits and nutrition in the body composition of dancers (Chaikali et al., 2023). Dance is a physically demanding profession that requires both good athletic performance and a lean physique. Consequently, many dancers restrict their caloric intake to maintain a low body weight (Wilmerding et al., 2005). However, some researchers pointed out that the pressure to sustain low body weight and body fat, particularly in ballet, can lead to unhealthy eating patterns and health issues if not properly managed (Sousa et al., 2013). Physical activity is widely recognized as a safe means to reduce and control body fat (Zaccagni et al., 2014). Indeed, the lack of change observed in this meta-analysis may be attributed to the duration, frequency, or intensity of the training programs, as some studies suggest that interventions lasting more than 14 weeks are typically required to impact body fat and lean muscle mass (Suarez-Arrones et al., 2019; Méndez-Hernández et al., 2022). Additionally, participants in the studies may have already possessed low body fat levels, complicating the detection of any significant changes (Wilmerding et al., 2003). Accordingly, future research should aim to determine the appropriate training dose to more accurately assess its effects on muscle mass and fat reduction. To achieve a more comprehensive understanding of these changes in dancers, studies should also account for dietary factors, initial body composition, and hormonal responses (Hincapié and Cassidy, 2010; Shaw et al., 2021).

Muscle strength is considered essential for improving dance performance and minimizing the risk of injuries in dancers (Koutedakis et al., 2005; Moita et al., 2017). Our findings indicate that strength training significantly improved muscle strength in dancers, closely linked to the physiological effects of the training. By increasing external resistance, strength training subjects muscle fibers to both mechanical tension and metabolic stress (Schoenfeld, 2013). This stimulation results in an increase in muscle fiber cross-sectional area (i.e., hypertrophy) and enhances strength through improved neural adaptations, specifically the efficiency and synchronization of motor neuron recruitment (Carroll et al., 2011). Moreover, strength training boosts muscles’ ability to withstand external forces, strengthening tendons and ligaments, which in turn enhances joint stability and overall force transmission efficiency (Chaudhry et al., 2017). For dancers, especially during complex movements like jumps and spins, greater muscle strength can significantly enhance body control and the precision of technical execution (Özkal et al., 2024).

In line with the findings of our meta-analysis, strength training has been well-established as an effective method for improving strength performance across a range of sports. For example, research shows that strength training significantly enhances leg strength and explosiveness in adolescent soccer players (McKinlay et al., 2018). Similarly, in basketball, strength training improves lower limb strength, contributing to enhanced jump performance and better rebounding capabilities (Yáñez-García et al., 2022). Swimmers also experience benefits from strength training, which enhances muscle strength and positively influences short-to medium-distance performance by improving force transmission and stroke mechanics (González et al., 2023). However, the analysis revealed moderate heterogeneity for muscle strength (I² = 68.77%), which may be attributed to a study that included a diverse group of participants, consisting of ballet, modern, and jazz dancers (Vetter and Dorgo, 2009). This study reported an exceptionally large ES (4.15). The inclusion of participants from multiple dance styles may have influenced the training effects on muscle strength. Based on our findings, strength training appears to be effective in improving muscle strength in dancers.

In dance, effective utilization of muscle power is vital for executing fast and repetitive movements, such as sequences of jumps (Malkogeorgos et al., 2013). For instance, it has been reported that ballet and modern dancers land on a single limb 50% and 89% of the time, respectively (Pappas et al., 2012). This meta-analysis revealed that strength training had a significant impact on muscle power in dancers when compared to regular dancing groups. The observed increase in muscle power from the strength training intervention was expected, given the historical evidence that various strength training programs lead to power gains. For example, Mascarin et al. (2017) reported significant improvements in muscle power and transfer effects on ball speed following a 6-week strength training program for young female handball players. Similarly, Rodríguez-Rosell et al. (2017) found notable increases in muscle power after a 6-week high-speed strength training regimen for young soccer players across different age groups. A recent study by Wang et al. (2024) observed significant improvements in muscle power following an 8-week flywheel strength training program in elite female volleyball players. Furthermore, our findings align with previous meta-analyses, suggesting that incorporating strength training can enhance power adaptations (Behm et al., 2015; Slimani et al., 2018; Uysal et al., 2023)

The observed improvements in muscle power performance among dancers can be partially attributed to physiological adaptations resulting from strength training. These adaptations are largely driven by neural pruning, which enhances motor efficiency and automates movement patterns (Ahtiainen, 2019). Research indicates that the increases in muscle power stem from various muscular adaptations, including changes in muscle size and architecture, as well as variations in the mechanical stiffness of muscle-tendon structures (Bojsen-Møller et al., 2005; Duchateau et al., 2021). In this meta-analysis, jump ability serves as the primary measure of muscle power, as it is widely recognized as a strong predictor of dance performance (Angioi et al., 2009b). Additionally, several studies have incorporated plyometric exercises to further enhance dancers’ performance (Brown et al., 2007; Sozbir et al., 2014; De Leonardis and Greco, 2020; Skopal et al., 2020). These exercises improve muscular function by utilizing a quick eccentric loading phase followed by rapid concentric contraction, thereby training the stretch reflex and harnessing stored elastic energy to enhance vertical jump performance (Markovic and Mikulic, 2010). Overall, this meta-analysis confirms that strength training positively impacts muscle power in dancers.

Good cardiorespiratory fitness is associated with quicker recovery from high-intensity intermittent exercise and a lower risk of musculoskeletal injuries in sports (Tiemens et al., 2023). However, this study found no significant increase in endurance among dancers following strength training. Notably, limited literature addresses the impact of strength training on dancers’ cardiorespiratory endurance. One study identified improvements in cardiorespiratory endurance after a 12-week regimen that combined aerobic and strength training (Koutedakis et al., 2007). In contrast, two other studies reported no significant changes in endurance following 8–16 weeks of strength training (Roussel et al., 2014; Sanders et al., 2020). Similarly, previous research on the effects of strength training on cardiorespiratory endurance has yielded inconsistent outcomes. For example, significant improvements were noted in well-trained runners undergoing strength training (Beattie et al., 2014). Additionally, a meta-analysis by Ramos-Campo et al. (2021) found that resistance circuit-based training effectively enhances cardiorespiratory endurance. Conversely, some strength training programs have demonstrated no effect on cardiorespiratory endurance in various athletes (Ignjatovic et al., 2011; Thiele et al., 2020).

The observed lack of performance gains in cardiorespiratory endurance in this meta-analysis may be attributed to the inherent nature of strength training, which primarily focuses on enhancing muscle strength and anaerobic capacity (Knuttgen, 2007). Typically, improvements in cardiorespiratory endurance necessitate moderate to high-intensity aerobic exercises that foster cardiovascular adaptations through sustained workloads (Hulke et al., 2012; Ahmed et al., 2022). Moreover, high-intensity interval training has been shown to significantly enhance cardiorespiratory endurance by alternating short periods of intense exercise with intervals of lower intensity, effectively promoting cardiovascular adaptations (Sultana et al., 2019). Additionally, the volume, frequency, and duration of the training in the included studies may not have been sufficient to elicit significant cardiovascular changes. While strength training may not directly enhance cardiorespiratory endurance, it could indirectly improve dancers’ movement efficiency and performance by increasing muscle strength and power, potentially reducing fatigue during performances (Izquierdo et al., 2006). Of note, the moderate heterogeneity (I² = 68.85%) observed in this analysis may be explained by one study that combined aerobic endurance, strength, proprioception, and motor control training (Roussel et al., 2014). Such a multimodal intervention introduces variability in training protocols and participant responses, which could contribute to differences in outcomes compared to studies focusing on isolated training modalities. Collectively, further research is needed to determine the role of strength training in developing endurance performance effectively.

Dancers are often perceived by the public as having fit, slim physiques and a wide range of joint mobility, traits considered advantageous in various dance performances (Steinberg et al., 2018). However, the results of this meta-analysis indicate that, despite engaging in strength training, dancers did not demonstrate significant improvements in flexibility. Supporting this observation, research by Morton et al. (2011) indicated that strength-focused exercises were generally less effective than dedicated flexibility training, such as static stretching, for improving flexibility. Conflicting results have been reported in studies involving young men (Leite et al., 2017), older women (Carneiro et al., 2015), and athletes (Saraiva et al., 2014), suggesting that resistance training can sometimes lead to increases in flexibility. The differing goals and mechanisms of strength training may explain the results we observed. While the primary focus of strength training is often on increasing muscle fiber size and enhancing strength output (Suchomel et al., 2018), it typically does not prioritize joint flexibility and range of motion, which are critical for dancers. Although full-range motion exercises may provide some flexibility benefits, the strength training programs utilized in the included studies may not have delivered sufficient stimulus to enhance the dancers’ joint flexibility. Additionally, dancers typically begin with a high baseline level of flexibility (Gupta et al., 2004), which may further restrict the potential for noticeable improvement in this area.

Some researchers argue that stretching is the most frequently recommended approach for improving flexibility (Medeiros et al., 2016; Ikeda and Ryushi, 2021). Numerous studies have highlighted the positive effects of Pilates training on dancers’ flexibility (Amorim et al., 2011; Ahearn et al., 2018), underscoring the importance of flexibility-specific training methods for this group. However, our meta-analysis included a study that compared the effects of resistance training and static stretching on flexibility, finding no significant differences in outcomes between the two conditions (Morton et al., 2011). Furthermore, a recent investigation revealed that resistance training through a full range of motion can be as effective as static stretching in enhancing flexibility (Rosenfeldt et al., 2024). Overall, our findings suggested that strength training might not substantially enhance flexibility in dancers; however, further research is required to reach a definitive conclusion.

The findings indicate that strength training modalities, including resistance, plyometric, weight training, and combined programs, have demonstrated positive effects on improving muscle strength and power in dancers. However, due to limited data, more specific evidence-based recommendations cannot yet be established. This meta-analysis provides preliminary guidelines, suggesting a minimum training duration of 6 weeks with 2–3 sessions per week, lasting 20–75 min per session. Exercise volume typically ranged from 2 to 10 sets per exercise, with 5–20 repetitions per set. Despite these benefits, the meta-analysis found no significant effects of strength training on body composition, cardiorespiratory endurance, or flexibility in dancers. Further research is required to confirm these findings and identify potential influencing factors. Practitioners and coaches should develop strength training programs tailored to individual competencies, aligned with dance-specific movement patterns, and scheduled well in advance of performances.

There are several limitations that warrant attention. Due to the limited number of studies, it was not possible to explore potential moderators such as training frequency, duration, session length, participants’ age, or dance genre. Importantly, the effects of different types of strength training may vary due to the specificity of training stimuli. However, this analysis did not compare various types of strength training, which could be an interesting and valuable direction for future research. Moreover, most studies focused on short-term interventions lasting 6–16 weeks, limiting our understanding of the long-term effects and sustainability of strength training benefits on dancers’ physical fitness. Furthermore, the studies primarily involved a limited range of dancer types, which may restrict the generalizability of the findings to other dance forms not covered in the analysis (e.g., samba, waltz). Additionally, only two studies examined the effects of neuromuscular training on dancers’ balance, and these were excluded from the meta-analysis. The positive findings reported in these studies require confirmation in future research, as balance is a vital fitness component for dancers.