Data concerning the effect of plyometric training on jump performance in soccer players: A meta-analysis

Plyometric training (PT) enhances soccer performance, particularly vertical jump. However, the effectiveness of PT depends on various factors. A systematic search of the research literature was conducted for randomized controlled trials (RCTs) studying the effects of PT on countermovement jump (CMJ) height in soccer players. Ten studies were obtained through manual and electronic journal searches (up to April 2017). Significant differences were observed when compared: (1) PT group vs. control group (ES=0.85; 95% CI 0.47–1.23; I2=68.71%; p<0.001), (2) male vs. female soccer players (Q=4.52; p=0.033), (3) amateur vs. high-level players (Q=6.56; p=0.010), (4) single session volume (<120 jumps vs. ≥120 jumps; Q=6.12, p=0.013), (5) rest between repetitions (5 s vs. 10 s vs. 15 s vs. 30 s; Q=19.10, p<0.001), (6) rest between sets (30 s vs. 60 s vs. 90 s vs. 120 s vs. 240 s; Q=19.83, p=0.001) and (7) and overall training volume (low: <1600 jumps vs. high: ≥1600 jumps; Q=5.08, p=0.024). PT is an effective form of training to improve vertical jump performance (i.e., CMJ) in soccer players. The benefits of PT on CMJ performance are greater for interventions of longer rest interval between repetitions (30 s) and sets (240 s) with higher volume of more than 120 jumps per session and 1600 jumps in total. Gender and competitive level differences should be considered when planning PT programs in soccer players.


Subject area
Sports sciences More specific subject area

Sports physiology
Type of data Raw and analyzed How data was acquired Data were acquired from articles included in the current meta-analysis. Table  Experimental factors

Data format
Data concerning type of intervention, gender, age, and competitive levels, volume of training sessions based on number of jumps per single session, types of plyometric exercises, training program duration in weeks, weekly frequency of training, rest interval between repetitions, rest interval between sets, and overall training volume were extracted from the included studies.

Experimental features
Meta-analysis according to different moderator variables (type of intervention, gender, age, and competitive levels, volume of training sessions based on number of jumps per single session, types of plyometric exercises, training program duration in weeks, weekly frequency of training, rest interval between repetitions, rest interval between sets, and overall training volume) were performed. Data source location NA Data accessibility Data are within this article.

Value of the data
Largest plyometric training effects were verified in female and high-level soccer players compared to male and amateur counterparts, respectively.
Higher volume of more than 120 jumps per session leads to greater effect of plyometric training on jump performance (countermovement jump (CMJ) without arm swing) when compared to less than 120 jumps per session.
Longer rest interval between repetitions (30 s) and sets (240 s) provide larger improvements in jump performance than shorter rest.
The benefits of plyometric training on jump performance are greater in participants who performed more than 1600 jumps in total than who performed less than 1600 jumps.

Data
Soccer is a most widely practiced sport around the world that combines cyclic and acyclic movements in competitive success [1,2]. However, soccer players perform numerous explosive movements like kicking, tackling, jumping, turning, sprinting, and changing pace and directions during the match [2]. For instance, many athletic and technical movements in soccer require rapid rates of force production or power in addition to high levels of coordination and reactivity, which called plyometrics [3,4].
When jump height performance was evaluated after plyometric training (PT), results from the literature were contradictory [3,5]. For instance, the effectiveness of PT depends on various factors, such as age [6], competitive level [3], training volume [7,8], and types of plyometric drills [3,5,9]. Therefore, the aim of this review was to examine the influence of various factors on the effectiveness of PT on jump height (i.e., countermovement jump (CMJ) without arm swing) using a meta-analysis approach.

Search strategy
The present meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines ( Fig. 1, [10]). A systematic search of the research literature was conducted for randomized controlled trials (RCTs) studying the effects of PT on jump height in soccer players. Studies were obtained through manual and electronic journal searches (up to April 2017). The present review used the following databases: PubMed, SCOPUS, SportDiscus, PsycINFO, PsycARTICLES, Google Scholar, and ScienceDirect. Electronic databases were searched using keywords and/or MeSH terms: "plyometric" or "plyometrics" alone or together with "soccer", "muscular power" and "jump". Moreover, manual searches of relevant journals and reference lists obtained from articles were conducted. The present meta-analysis includes studies published in journals that have presented original research data on healthy human subjects. No age and gender were imposed during the search stage.

Inclusion and exclusion criteria
Studies were included in the review if they met all the following Population/Intervention/Comparison/Outcome(s) (PICO) criteria: (1) Population: studies recruiting male and female amateur (i.e., amateur, healthy, regional) and/or high-level (i.e., high-level, professional, national, 2 nd league, 1 st league) soccer players of any age as participants; (1) Reviews, comments, opinions and commentaries, interviews, letter to editor, editorial, posters, conference abstracts, book chapters, and books were excluded; available reviews have been anyways scanned for increasing the chance of including potentially relevant articles; (2) Assessing performance changed after PT combined with other intervention (strength or sprint training); (3) Studies not employing true experimental designs and valid and reliable measurements; (4) Studies including enough data to calculate effect size; (5) Lacking quantitative information and details.

Identification of the moderator variables
Two authors independently extracted data using a structured form. Because of the high number of variables that may affect training effectiveness, independent variables were grouped into the following areas: 1) type of intervention: plyometric group vs. control group; 2) subject characteristics: gender (male vs. female), age (o15 vs. between 15 and 21 vs. Z21 years), and competitive levels (amateur vs. high-level); 3) program exercises: volume of training sessions based on number of jumps per single session (low: o 120 vs. high: Z120 jumps); types of plyometric exercises (single vs. combined exercises [two or more than two exercises]); and 3) program elements: a) training program duration in weeks (o8 vs. Z 8 weeks); b) weekly frequency of training (1 vs. 2 vs. 3 sessions per week); c) rest interval between repetitions (5 s vs. 10 s vs. 15 s vs. 30 s); d) rest interval between sets (30 s vs. 60 s vs. 90 s vs. 120 s vs. 240 s); e) overall training volume (low: o1600 jumps and high: Z1600 jumps: calculated as total number of jumps per study).

Statistical analyses
For the meta-analysis part, data were extracted from the included studies using a standardized documentation form. For preliminary analysis the effect estimates were computed as standardized mean differences of experimental and control group with their 95% confidence interval (CIs). Metaanalyses were carried out using the program Comprehensive Meta-Analysis, version 2 [11]. Statistical heterogeneity in our meta-analysis was assessed using the Q and I 2 statistics. The I 2 measure of inconsistency was used to examine between-study variability; values of 25, 50 and 75% represent low, moderate and high statistical heterogeneity, respectively [12]. Although the heterogeneity of effects in our meta-analysis ranged from 0% to 76.09% (see results section), we decided to apply a randomeffects model of meta-analysis in all comparisons, in order to determine the pooled effect of PT on CMJ height. Possible publication bias was visually inspected with a funnel plot, looking at asymmetry of the graph. In addition, regression analysis (method of moments) were used in order to investigate possible predictors of observed effect size (ES) among aforementioned continuous variables (age of athletes, training program duration, weekly frequency, rest intervals between repetitions and sets, single session and overall training program volume. The magnitudes of the ESs were considered either trivial (o0.35), small (0.35-0.80), moderate (0.80-1.50), or large ( 41.5) [13]. Furthermore, a regression analysis was used to verify the effects of potential moderator variables on the ES of study results. The significance level of p o0.05 was used. The search strategies yielded a preliminary pool of 921 possible papers. The full text of 36 articles were retrieved and assessed for eligibility against the inclusion criteria. After a careful review of their full texts, 26 articles were excluded and the remaining 10 articles were eligible for inclusion in the review ( Fig. 1; Table 1).

Competitive level (amateur vs. high-level)
Significant difference was found between high-level soccer players compared to amateur counterparts (Q¼6.56; p ¼0.010) in CMJ height after PT.

Types of plyometric exercises (single vs. combined exercises)
There was no significant difference in magnitude of ES when studies with single or combined exercises (Q ¼2.58; p ¼0.108) were compared.

Single training session volume
There was significant effect of number of jumps per training session, where higher volume of more than 120 jumps leads to greater effect when compared to less than 120 jumps per training session (Q¼ 6.12; p ¼0.013) ( Table 2). In addition, regression analysis confirmed preliminary results, showing that number of jumps as significant predictor of the ES (Z¼2.998; p ¼0.003) ( Table 3; Fig. 3).

Frequency of weekly sessions (1 vs. 2 vs. 3 sessions per week)
Weekly frequency of training showed heterogeneous effects ranging from small (ES ¼ 0.53) to large (ES ¼3.00) for one to three trainings per week, respectively, without significant difference between them (Q¼2.62; p ¼0.270). However, regression analysis showed that weekly frequency of training was significant predictor (Z¼2.194; p¼ 0.028) of CMJ height gains following PT (Fig. 4). Meta-analyzed effect of PT regarding different rest periods between repetitions yielded heterogeneous effects ranging from small (ES ¼0.39; 5 s) to large (ES¼ 1.06; 30 s) ( Table 2). Compared to all other intervals, 30 s of rest between repetitions showed significant difference in gains (po 0.001, Q ranging from 16.85 to 18.68 for 10 s and 15 s intervals). In addition, regression analysis have shown that rest interval between repetitions is significant predictor of observed effect (Z¼2.885; p ¼0.004) ( Table 3; Fig. 5).  The rest between sets showed diversity of effect ranging from small (ES¼ 0.39, 30 s) to large (ES ¼3.89; 240 s). Compared to all other intervals significant difference was observed only for rest of 240 s between sets (p o0.001, Q ranging from 16.384 to 19.233). In addition, regression analysis have shown that rest interval between sets is significant predictor of observed effect (Z¼3,079; p ¼ 0.002) ( Table 3; Fig. 6).

Overall training program volume
Marginally significant effect of training volume was observed among analyzed studies, where more than 1600 jumps showed greater ES (ES ¼1.55) compared to less than 1600 jumps per study (ES ¼0.52). Additional regression analysis indicated overall training volume as significant predictor of observed effects (Z ¼3.339; p ¼0.001) ( Table 3; Fig. 7).
Despite the important conclusions that can be drawn from this meta-analysis concerning the effectiveness of PT on jump performance, it is important to note that some limitations should be considered. For example, there was a considerable amount of small numbers of included studies, and this can be due to the high-quality standards used to select studies (e.g., randomized-controlled), and the highly specific focus of the data (i.e., CMJ, in soccer). This leads to bias or limitations in the generalization of results.

Transparency document. Supplementary material
Transparency document associated with this article can be found in the online version at http://dx. doi.org/10.1016/j.dib.2017.09.054.