Date of Award

Summer 2025

Document Type

Open Access Dissertation

Department

Physical Education

First Advisor

David F. Stodden

Abstract

This dissertation consists of three studies integrating frequentist and machine learning techniques to examine physical fitness trajectories and future physical military readiness. Study 1 consists of a cross-sectional examination of Physical Fitness Test (PFT) scores from four cohorts of first-year Cadets at a Senior Military College from 2019 to 2022. Study 2 consists of a longitudinal examination of PFT performance from one cohort of Cadets across a baseline test and three years of training at a Senior Military College. Study 3 examines predictive machine learning and growth curve modeling methods to project future physical military readiness based on motor competence and physical fitness data from youth and adolescents.

The primary aim of Study 1 was to determine if the physical fitness levels of incoming cohorts of Cadets at a Senior Military College vary over time from 2019 to 2022, which spans across the COVID-19 pandemic. Physical fitness is a critical determinant of military readiness, and declining fitness levels among U.S. young adults have negatively impacted eligibility for military service. The COVID-19 pandemic, beginning in the Spring of 2020, caused widespread disruptions to physical activity opportunities, potentially amplifying the ongoing military recruitment challenges in the U.S. Further research is needed to examine the impact of the COVID-19 pandemic and associated limitations to structured group physical activity and fitness opportunities on physical fitness levels. This study utilized baseline PFT scores for four cohorts of incoming Cadets from 2019 to 2022 during the Fall semester of their freshman year. The PFT includes three fitness assessments: push-ups (muscular endurance), a plank hold or crunches (core muscular endurance), and a 1.5-mile run (cardiovascular endurance). Each component was scored on a 0 to 100-point scale, and the scores were summed to calculate a total PFT score ranging from 0 to 300 points. We conducted four 4 x 2 factorial ANOVAs to examine differences in total PFT scores and individual component scores (push-ups, plank or crunches, and run time) across cohorts (2019, 2020, 2021, and 2022) and by sex. Post hoc pairwise comparisons were conducted to specify differences across cohorts using Tukey’s HSD. Initial analyses revealed a significant main effect of cohort on total PFT scores (F(3, 1874) = 41.1, p= < .001). However, this result was primarily driven by inconsistencies in the abdominal strength and endurance assessment across cohorts. The 2019 cohort, which completed a crunches assessment, had notably lower core muscular endurance component scores (M = 64.7 points) compared to the subsequent cohorts, which completed a plank assessment: 2020 (M = 81.1 points), 2021 (M = 89.0 points), 2022 (M = 90.1 points). Push-up points decreased significantly from 2020 (M= 97.5 points) to 2021 (M = 87.8 points), decreasing by an average of 9.73 points (p < .001). Push-up points began to rebound by 2022 (M = 89.5 points), with a mean increase of 1.7 points (p = .004), suggesting a temporary negative association of the pandemic on muscular strength. Although there was a significant effect of cohort on 1.5 mile run points, post hoc analyses revealed a mean difference of only 2.9 points between the 2021 and 2022 cohorts, and otherwise indicated stability in cardiovascular endurance of incoming Cadets during the pandemic (F(3, 1871) = 4.05, p = .007). There was no significant interaction between cohort and sex (F( (3,1871) = 0.01, p = .99), indicating that the differences in PFT scores across cohorts are consistent for both males and females. The stability in cardiovascular endurance suggests that activities like running and walking, which were less restricted during the pandemic, helped mitigate declines in this aspect of fitness. However, the temporary decline in push-up performance highlights the potential negative impact of the pandemic and associated physical activity and structured physical training opportunity limitations on muscular strength and endurance. These findings have important implications for military training programs, emphasizing the need to address gaps in training to sustain military readiness. Lastly, the absence of targeted strength and power assessments in the PFT limits the ability to fully assess physical military readiness. Implementing components from the Army Combat Fitness Test (ACFT), which was designed to test speed, power, strength, endurance, and functional capabilities, could improve the ability to assess military readiness.

Study 2 utilized the PFT of a single cohort of Cadets (N = 388, female = 50) across three years (2019 to 2022), encompassing four time points: baseline and three consecutive years of training. Training provided at Senior Military Colleges aims to improve physical fitness over time as it is a core aspect of physical military readiness. Decreasing trends among young adult fitness levels in the U.S., coupled with the COVID-19 pandemic interrupting normal modes of training, may have influenced physical fitness trajectories during the years surrounding the COVID-19 pandemic. This study examined changes in PFT performance among Cadets at a Senior Military College across three years of intensive physical readiness training. The PFT utilized includes measures of muscular strength/endurance (push-ups, plank hold) and cardiovascular endurance (1.5-mile run). We conducted three Split-Plot ANOVAs to examine longitudinal performance trajectories in push-up repetitions, plank hold time (s), and 1.5-mile run time (s) across three years, as well as determine if male and female fitness trajectories were different. There was a significant main effect of time on push-up performance (F(3, 1,543) = 326.4, p < .001, ηp² = .39), plank hold performance (F(2, 1,154) = 23.3, p < .001, ηp² = .04), and 1.5-mile run time (F(3, 1,543) = 10.3, p < .001, ηp² = .02, indicating year-to-year variations in performance. Post hoc pairwise comparisons among years revealed that Cadet’s performance trajectories were not linear, and performance may have been significantly impacted by the COVID-19 pandemic. In addition, there were significant differences between male and female performance trajectories for push-up performance (F(3, 1,543) = 20.5, p < .001, ηp² = .04) and run time (F(3, 1,543) = 7.78, p < .001, ηp² = .02). Post hoc analyses revealed that males experienced more significant declines in fitness performance than females in all PFT components during the pandemic. Cadets in this study met the minimum physical readiness standards of the Senior Military College; however, minimum standards on some tests decrease in difficulty, based on U.S. military standards as Cadets age from 18 to 22 years. While some improvements in physical fitness were observed, performance trajectories were not linear across the three years, indicating that fitness gains were neither uniform nor sustained year-to-year. Furthermore, the presence of significant sex differences, with males experiencing more pronounced declines in fitness test performance during the pandemic, suggests that training programs may not have fully mitigated the impacts of external disruptions. These findings highlight the need for more targeted and adaptive interventions to better support continuous physical fitness improvements across enrollment years rather than relying on minimum standards that may not adequately reflect optimal physical readiness.

The purpose of study 3 was to apply growth curve modeling coefficients from longitudinal motor competence (MC) data to predict future recruitment age physical fitness and potential Army fitness test performance. A secular decline in MC and physical fitness among U.S. youth has contributed to a national security crisis due to the alarming number of young adults unable to demonstrate adequate physical fitness levels to pass required military physical fitness tests. Previous research demonstrates moderate to strong associations among MC levels, physical fitness, and Army fitness test performance/failure rates in adults. As MC and fitness levels track across youth, a continued decline in youth MC levels may further exacerbate recruitment issues and future physical military readiness. Currently, there is no longitudinal MC data that spans beyond 7 years within the time from childhood through adolescence. The purpose of this study was to use longitudinal childhood MC and fitness test scores (ages 8-10) and apply a predictive method that includes growth curve modeling and machine learning to predict future recruitment age physical fitness levels and hypothetical Army fitness test performance. An initial sample of children (N = 262, female = 135, Mage = 8.8 years) completed MC (standing long jump distance, throw speed, & kick speed) and physical fitness (grip strength & push-ups) assessments across three years. Using a machine learning regression method, random forest regression, tracking coefficients were calculated for MC and physical fitness measures to predict MC and physical fitness levels at age 13 (β =.44 to 1.01). Using growth curve modeling, we further extrapolated MC and physical fitness levels to military recruitment age by applying longitudinal tracking coefficients for the same MC and physical fitness variables from a different longitudinal data set (ages 15 to 18 years - β = 0.34 to 0.68). The future predicted MC and physical fitness levels at age 18 years were compared to MC and physical fitness levels from a current sample of young adults in a pre-military setting (N = 306, Mage = 19.55) to calculate the initial sample’s predicted MC and physical fitness percentiles. The extrapolated growth curves revealed that 80% of females and 63% of males are predicted to be in the lowest 25%tile of MC and physical fitness at age 18. Approximately 75% of individuals at or below the 25%tile of MC are predicted to fail the current Army fitness test at age 18. This innovative predictive growth curve modeling application highlights the potential significance that the secular decline in MC, as a critical antecedent for physical fitness, will have on physical military readiness in future generations of military aged recruits.

Rights

© 2025, Giovanna Leone

Download

Share

COinS