Twenty-four collegiate volleyball athletes (12 males and 12 females; age 18–22 years) voluntarily participated in this study. All participants had a minimum of two years of competitive playing experience at the university or regional level and had no history of lower-limb injuries or musculoskeletal disorders within the past six months. The absence of recent injuries was verified through a standardized self-report health questionnaire and confirmed by a brief musculoskeletal screening conducted by a certified sport science practitioner. Eligibility criteria included regular participation in team training (≥3 sessions per week) and the absence of any contraindication to high-intensity exercise. Before participation, each athlete received a detailed explanation of the study procedures and provided written informed consent.

Participants were stratified by gender and randomly assigned to one of four groups (n = 6 per group): (1) male–plyometric, (2) male–Air Alert, (3) female–plyometric, and (4) female–Air Alert. Randomization was performed using a computer-generated block randomization procedure (block size = 4) to ensure equal allocation across the four groups within each sex. The study protocol was reviewed and approved by the Human Research Ethics Committee of the University of Phayao (Protocol No. UP-HEC 1.1/006/65). All procedures conformed to the Declaration of Helsinki on ethical principles for research involving human subjects.

This study employed a randomized controlled experimental design with two training modalities (Plyometric and Air Alert) and two gender groups (male and female). Each participant completed three testing sessions: pre-test (Week 0), mid-test (Week 4), and post-test (Week 8). This repeated-measures randomized controlled design minimizes inter-individual variability by allowing each participant to serve as their own control across testing points, thereby increasing the sensitivity to detect true training-induced changes.

At baseline, participants underwent familiarization sessions to ensure consistency in jumping technique and measurement procedures. Participants completed two familiarization sessions, each lasting approximately 20–25 min, during which they practiced the standardized jumping protocol and measurement procedures under investigator supervision. All tests and training sessions were conducted at the Motion Analysis Laboratory and Volleyball Training Facility, Department of Sports Science, University of Phayao, Thailand. Environmental conditions (temperature 25–28°C, humidity 50%–60%) were standardized for all sessions. The laboratory was equipped with controlled ventilation, uniform LED overhead lighting, and a non-slip synthetic sport flooring surface to ensure consistent biomechanical measurement conditions across all testing sessions.

The experimental timeline consisted of: •

Week 0 (Pre-test): Assessment of vertical jump height and anthropometric data.

The plyometric group followed a progressive overload model designed to enhance explosive lower-limb power through high-intensity, low-volume jump exercises. Sessions were conducted three times per week for eight weeks, with at least 48 h of recovery between sessions. Each session lasted approximately 45 min and consisted of: •

Week 1–2: Squat jumps (3 × 10 reps), split lunges (3 × 10 reps), bounding (3 × 20 m).

The Air Alert group followed the standardized “Air Alert III” protocol, which involves high-repetition jump exercises focusing on endurance-type neuromuscular adaptation. Participants performed sessions three times per week for eight weeks, increasing total jumps weekly according to the program manual. Exercises included leap-ups, double jumps, calf raises, and squat jumps, starting at approximately 500 jumps per session in Week 1 and progressing to over 1,000 jumps by Week 8. No external load was added, and only short rest intervals (<30 s) were allowed between sets.

Although the Air Alert program is widely circulated among amateur and recreational athletes, its scientific validation remains limited. No peer-reviewed studies have systematically evaluated its efficacy or safety, and existing reports are largely anecdotal. This scarcity of empirical evidence highlights the need to examine the program within a controlled experimental setting, particularly when comparing it with structured plyometric training.

The substantial difference in recovery structure between the two programs reflects the inherent design and intended physiological targets of each training modality rather than an experimental bias. The plyometric protocol incorporated 48 h inter-session rest intervals, consistent with evidence-based recommendations for high-intensity power development to allow for neuromuscular recovery and minimize accumulated fatigue. In contrast, the Air Alert program prescribes very short intra-session rest periods and extremely high repetition volumes to elicit an endurance-type neuromuscular stimulus, as specified in the program manual. Preserving these contrasting recovery patterns ensured the ecological validity of each training approach and allowed for an accurate comparison of how fundamentally different jump-training paradigms influence fatigue and adaptation responses.

Both groups were instructed to avoid additional lower-body resistance training during the intervention period. Attendance and compliance rates were >95% across all participants. Attendance was monitored using researcher-maintained session logs, and each training session was supervised directly by a certified strength and conditioning coach to ensure full compliance with the prescribed protocols.

Vertical jump height was assessed using a Vertec Jump Measurement Device (Vertec™, Sports Imports, Columbus, OH, USA), a valid and reliable instrument for measuring vertical displacement (ICC = 0.97; SEM ±0.5 cm). Participants performed three maximal countermovement jumps with hands free, and the highest value was recorded for analysis.

A subset of participants (n = 8; two from each group) was additionally evaluated using a DMAS high-speed infrared motion capture system (6 cameras, 300 Hz) to confirm kinematic consistency in take-off and landing technique. Given the small subsample (n = 8), the motion-analysis component of the study was considered exploratory. The limited sample size reduces the generalizability of the kinematic findings; therefore, these results should be interpreted as supplementary insights rather than definitive biomechanical conclusions. Key biomechanical parameters recorded included knee flexion angle, ankle plantarflexion, and arm swing amplitude at pre-, mid-, and post-tests.

Body mass and height were recorded at each testing phase using a digital scale (±0.1 kg) and stadiometer (±0.1 cm). Training load consistency was ensured by supervising all sessions and maintaining uniform warm-up and recovery procedures.

All data were analyzed using SPSS Statistics version 28.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics (mean ± SD) were computed for all variables. Assumption checks were performed before conducting the repeated-measures ANOVA. Normality of residuals was evaluated using the Shapiro–Wilk test, which indicated no significant deviations from normal distribution. Sphericity was assessed using Mauchly's test, and when violations were detected, the Greenhouse–Geisser correction was applied to adjust the degrees of freedom accordingly.

Because the fixed roster structure restricted the maximum achievable sample size, an a priori power analysis was not feasible. Therefore, statistical power was evaluated post hoc based on the observed effect sizes. The achieved power for the primary repeated-measures ANOVA model was 0.83, indicating adequate sensitivity to detect meaningful training-related effects within the available sample.

A three-way repeated-measures ANOVA (3 × 2 × 2) was employed to examine the main effects and interactions of time (pre-, mid-, post-), training type (Plyometric vs. Air Alert), and gender (male vs. female) on vertical jump height.

When significant effects were detected, post hoc pairwise comparisons with Bonferroni adjustment were performed. Effect sizes (η²) were calculated and interpreted as small (0.01–0.06), medium (0.06–0.14), and large (>0.14).

Normality and sphericity assumptions were verified using the Shapiro–Wilk test and Mauchly's test, respectively. Homogeneity of variance was assessed via Levene's test. If violations occurred, the Greenhouse–Geisser correction was applied.

The level of statistical significance was set at p < .05. Cohen's d values were calculated to quantify the magnitude of pairwise differences. A post hoc power analysis (G*Power 3.1) confirmed that the achieved power (1–β) was 0.83, indicating adequate sensitivity for detecting moderate effect sizes (f = 0.25).

Regression Equation:Y^post=−0.52+2.14Xpre−0.86Xmid+0.46Xgender+2.15XtrainingNote. Dependent variable: post-training vertical jump height (cm). Predictors included pre- and mid-training jump height, gender (Male = 1, Female = 0), and training type (Plyometric = 1, Air Alert = 0). The model significantly predicted post-test performance (R² = .86, p < .001), explaining 86% of variance.

Changes in vertical jump performance across testing phases and training groups are presented in Table 2.

Effect size values (Cohen's d) for vertical jump performance improvements are reported in Table 3.

The results of the multiple regression analysis predicting post-training jump height are presented in Table 4.

Both pre-test height and training type were the strongest predictors of post-training outcomes. Gender had a moderate positive contribution, indicating that male athletes achieved slightly greater post-training gains, while mid-phase performance negatively predicted final outcomes, suggesting diminishing marginal returns. No multicollinearity was observed (VIF < 2.0 for all predictors).

The progression of vertical jump performance across the three testing phases is illustrated in Figures 1–3, which display the jump-height trend, effect sizes, and standardized regression coefficients, respectively.

The purpose of this study was to compare the effects of two jump training protocols—Plyometric and Air Alert—on vertical jump performance among male and female collegiate volleyball athletes, and to develop a predictive model for post-training outcomes. The findings demonstrated significant improvements across all groups, confirming that both training programs effectively enhanced jump performance. However, athletes who performed plyometric training achieved markedly greater gains in vertical jump height than those who trained with the Air Alert program. These results support the hypothesis that progressive, load-based jump training produces superior performance adaptation and movement efficiency. These findings are consistent with volleyball performance literature, where increases in vertical jump height have been linked to improvements in spiking efficiency, attack success rates, and blocking effectiveness at both collegiate and international levels (1, 8).

The superior outcomes observed in the Plyometric groups align with prior studies emphasizing the benefits of structured progression, controlled loading, and appropriate recovery in maximizing explosive movement performance (4–6). The training design in this study—featuring varied exercise selection and incremental overload—likely maintained high stimulus quality and minimized early plateaus in performance.

The superior performance improvements observed in the plyometric group can be attributed to several neuromuscular and biomechanical mechanisms. Plyometric movements typically enhance stretch–shortening cycle (SSC) efficiency by reducing amortization time, increasing muscle–tendon stiffness, and improving elastic energy reutilization during the eccentric–concentric transition (2). These adaptations contribute to increased rate of force development (RFD), allowing athletes to generate higher propulsive impulses in shorter time frames. Additionally, repeated exposure to high-intensity landing and take-off actions improves intramuscular coordination, motor-unit recruitment, and intermuscular synchrony—all of which are essential determinants of vertical jump performance. Practically, these neuromuscular adaptations directly enhance volleyball-specific actions such as spiking penetration, block height, and transition-speed efficiency, all of which are critical determinants of competitive performance.

Consistent with earlier meta-analyses, the current results showed large to extremely large effect sizes (Cohen's d > 2.0) in both male and female participants (9). These magnitudes are indicative of substantial practical improvement in jump ability and confirm the efficiency of plyometric methods for developing vertical power in volleyball athletes. These physical improvements also align with performance-analytics research demonstrating that greater jump capacity is associated with enhanced spiking efficiency and offensive effectiveness in women's volleyball (18). In contrast, the repetitive, high-volume structure of the Air Alert program may have led to cumulative fatigue, reducing movement quality and overall performance progression—a pattern similarly noted by Ramírez-Campillo et al. (11) in prolonged jump-based interventions.

Although Air Alert training produced statistically significant gains, the effect magnitudes were notably smaller (Cohen's d ≈ 1.5–1.8). The lack of recovery modulation and the monotonous repetition pattern likely contributed to fatigue accumulation, restricting further improvement. Similar findings have been reported in long-term analyses of repetitive jump programs, where excessive workload limits power development despite high total volume (7). Although no physiological fatigue markers such as RPE, heart rate, or lactate were collected in this study, the observed performance pattern is consistent with theoretical and empirical evidence indicating that high-volume, low-recovery jump protocols can induce cumulative neuromuscular fatigue and impair movement quality.

The emphasis on endurance-oriented jumps within the Air Alert program may also explain its lower transferability to power-dominant movements such as vertical jumping. From a practical standpoint, these results underscore the importance of prioritizing training quality and load progression over mere repetition count in jump-based conditioning programs.

Significant gender differences in absolute jump height were observed, with male athletes exhibiting higher performance values throughout all phases. This outcome is consistent with established physiological factors such as greater muscle mass and force capacity in males (10). Nevertheless, relative improvements were comparable between males and females (+18.4% vs. +16.6%), demonstrating that well-structured plyometric training can effectively enhance performance across sexes when loads are scaled appropriately.

These results confirm earlier reports of gender-equitable adaptation under progressive plyometric loading (9, 16). Moreover, the findings support evidence from volleyball performance analyses showing that technical and tactical efficiency, rather than absolute power, best predicts match success (18). Hence, relative improvement in performance should be considered a key indicator of training effectiveness in both male and female athletes.

The longitudinal (pre–mid–post) analysis provided insight into the progression of performance adaptation. The Plyometric groups exhibited consistent improvement across all stages, while the Air Alert groups showed slower progression and mid-phase stagnation. This pattern reflects the expected transition from initial coordination-based gains to later structural and strength-related adaptations (5, 17).

The multiple regression model accounted for 86% of the variance in post-training performance (R² = 0.86, p < 0.001). Pre-test performance (β = 0.51) and training type (β = 0.47) were the strongest positive predictors. In contrast, the negative mid-test coefficient (β = −0.29) indicates that athletes who improved rapidly during the early phase tended to show slower gains afterward. This pattern aligns with known neuromuscular adaptation principles, where early improvements are primarily neural—enhanced motor-unit recruitment, coordination, and movement efficiency—before reaching a plateau. Further progress then requires slower structural adaptations such as hypertrophy or increased tendon stiffness. As such, high mid-phase scores may reflect an early peak combined with short-term fatigue or overreaching, ultimately reducing late-phase progression. These findings reinforce the principle of diminishing returns and highlight the need for periodic load adjustment and mid-phase monitoring to sustain improvement (12).

Despite the high explanatory power of the regression model (R² = 0.86), model robustness has not yet been verified. The present study did not apply cross-validation procedures such as k-fold validation, hold-out testing, or independent cohort validation. Incorporating these approaches in future research would help determine whether the predictive equations remain stable across different teams, performance levels, and demographic characteristics, thereby improving the practical utility of the model for long-term athlete monitoring.

Comparable analytical frameworks have been successfully applied to model tactical, technical, and scoring efficiency in volleyball, as demonstrated by recent Frontiers work on spiking efficiency across attack types, zones, and set-phase conditions (18), along with related analytical models in professional leagues (19, 20).

These findings provide several practical insights for coaches and practitioners: 1.

Plyometric training should be prioritized for vertical power development, with controlled progression and sufficient recovery to prevent overload.