Status : Verified
| Personal Name | Sanita, Ryan Ferdinand G. |
|---|---|
| Resource Title | Analyzing key factors affecting running performance: a statistical study of race results data from running events in the Philippines in 2024 |
| Date Issued | June 2025 |
| Abstract | Long-distance running remains to be one of the most participated sports in the world, as evidenced by the sheer increase in running event participation in the past years. In our aim to understand running performance better, we used a publicly available race results database and adopted a data-driven approach. This study examined the combined effects of demographic attributes (age group and sex) and participation metrics (specific race frequency and overall race frequency) on running performance. The dataset included 67,371 finishers (Male = 39,051; Female = 28,320). As a result of the Exploratory Data Analysis (EDA), a dataset with outliers removed (final N = 66,845) was used for the correlation. The statistical model focused on runners with ≥2 race entries only (N = 42,164). Average pace (sec/km) was fitted as a function of age group, sex, and frequency of race as fixed effects using Generalized Additive Mixed Models (GAMMs), adding smooth random effects for Runner ID and Race ID. The key findings of the main analysis were: (1) 25–29 (β = 11.17, p < 0.05), 30–34 (β = 9.18, p < 0.05), and 60+ (β = 25.64, p < 0.05) runners were all significantly slower than the reference group of 18–24-year-old runners; (2) females averaged 52.49 sec/km more slowly than males (p < 0.05); and (3) each additional race event enhanced pace by 1.40 sec/km (p = 0.003). These findings provide us with empirical evidence of the effects of the demographic attributes of age group and sex on running performance while also deepening our understanding on the relationship of race participation frequency on specific and general running performance. Furthermore, this study demonstrates the value of applying data science and advanced statistical modeling, such as GAMMs, in advancing research in sports performance. |
| Degree Course | Bachelor of Sports Science |
| Language | English |
| Keyword | long-distance running; performance; mixed models; data-driven |
| Material Type | Thesis/Dissertation |
Preliminary Pages
1.53 Mb
Category : F - Regular work, i.e., it has no patentable invention or creation, the author does not wish for personal publication, there is no confidential information.
Access Permission : Open Access