Search Results

You are looking at 1 - 10 of 139 items for :

  • "linear mixed model" x
Clear All
Restricted access

Samuel Ryan, Aaron J. Coutts, Joel Hocking and Thomas Kempton

Purpose:

To examine the influence of a range of individual player characteristics and match-related factors on activity profiles during professional Australian football matches.

Methods:

Global positioning system (GPS) profiles were collected from 34 professional Australian football players from the same club over 15 competition matches. GPS data were classified into relative total and high-speed running (HSR; >20 km/h) distances. Individual player aerobic fitness was determined from a 2-km time trial conducted during the preseason. Each match was classified according to match location, season phase, recovery length, opposition strength, and match outcome. The total number of stoppages during the match was obtained from a commercial statistics provider. A linear mixed model was constructed to examine the influence of player characteristics and match-related factors on both relative total and HSR outputs.

Results:

Player aerobic fitness had a large effect on relative total and HSR distances. Away matches and matches lost produced only small reductions in relative HSR distances, while the number of rotations also had a small positive effect. Matches won, more player rotations, and playing against strong opposition all resulted in small to moderate increases in relative total distance, while early season phase, increased number of stoppages, and away matches resulted in small to moderate reductions in relative total distance.

Conclusions:

There is a likely interplay of factors that influence running performance during Australian football matches. The results highlight the need to consider a variety of contextual factors when interpreting physical output from matches.

Restricted access

Berit Steenbock, Marvin N. Wright, Norman Wirsik and Mirko Brandes

filtering of the raw accelerometer data was conducted. Predictive models were created separately for each accelerometer and placement, resulting in 24 (4×6) models developed and tested. We built four predictive models: (1) a linear regression model, (2) a linear mixed model, (3) a random forest, and (4) an

Restricted access

Iñigo Mujika, Luis Villanueva, Marijke Welvaert and David B. Pyne

, SD, minimum, and maximum) were calculated for ΔPerf with respect to each swimmer’s Time sb-ls , rank, swimming stroke, distance, age, sex, and country for leading swimming nations. Statistical inference was derived from a general linear mixed model. The random structure incorporated the multilevel

Restricted access

Cassandra Sparks, Chris Lonsdale, James Dimmock and Ben Jackson

the effect of the teacher training on PE students’ perceptions of relatedness support, enjoyment, tripartite efficacy beliefs, self-determined motivation (i.e., RAI), 1 and amotivation, we conducted linear mixed models using SPSS 22.0 because of the hierarchical structure of the data (i.e., students

Restricted access

Ryland Morgans, Rocco Di Michele and Barry Drust

jump as high as possible. An Accupower force plate (AMTI, Watertown, MA, USA) was used for data collection. The CMJ height and peak power (PP) were taken as the outcomes of the CMJ test. Linear mixed models were used for data analysis, with random intercepts for individual players. First, for all

Restricted access

Farhan Juhari, Dean Ritchie, Fergus O’Connor, Nathan Pitchford, Matthew Weston, Heidi R. Thornton and Jonathan D. Bartlett

of whether participants competed in the Australian Football League or Victorian Football League competition, their planned weekly schedule in relation to training day type was the same. Statistical Analysis Linear mixed models were constructed to estimate session volume, intensity, and distribution

Restricted access

Zhiguang Zhang, Eduarda Sousa-Sá, João R. Pereira, Anthony D. Okely, Xiaoqi Feng and Rute Santos

, and stepping) was performed using linear mixed model, adjusted for center-level clustering effects, time sequence, and baseline demographic variables. No intervention effect was found ( Supplementary Table S1 [available online]). Descriptive characteristics, at baseline and follow-up, were presented

Restricted access

Janelle M. Wagnild and Tessa M. Pollard

or ≥2 hours per day. Participants provided demographic information about themselves on the study enrolment form. BMI (from approximately 8 weeks’ gestation) was extracted from medical records. Statistical Analyses Linear mixed models were used to examine the daily and hourly patterning of ST with

Restricted access

Carolina F. Wilke, Samuel P. Wanner, Weslley H.M. Santos, Eduardo M. Penna, Guilherme P. Ramos, Fabio Y. Nakamura and Rob Duffield

.41. Table 3 Linear Mixed-Model Parameter Estimates for the Effect of Recovery Profile, Training Load, Sleep, and Phase of the Microcycle on Total Quality of Recovery Scale Coefficient SE df t P Fixed effects  Intercept 12.30 0.93 107.30 13.29 <.01*  Cluster −0.28 0.21 16.13 −1.33 .20  TRIMP −2

Restricted access

Emily M. D’Agostino, Sophia E. Day, Kevin J. Konty, Michael Larkin, Subir Saha and Katarzyna Wyka

to determine the total variability in chronic absenteeism attributable to schools. The intraclass correlation coefficient = τ 2 /(τ 2  + 3.29) was computed based on an unconditional 3-level logistic generalized linear mixed model with random intercepts (observations nested in students, nested in