Examining Movement Behaviours in Preschool-Aged Children: Novel Measurement and Data Analysis Techniques

• Author / Creator

  Kuzik, Nicholas Oliver Corey

• Movement behaviour patterns (e.g., more sleep, less sedentary behaviour, and more physical activity) in isolation have demonstrated benefits to preschool-aged children’s development. However, the integrated nature of movement behaviours is a relatively unexplored area. This scarcity of evidence presents an opportunity to sequentially build a foundation of high-quality evidence. The overall objective of this dissertation was to systematically advance the area of movement behaviours in preschool-aged children using novel measurement and data analyses techniques.
  Three manuscripts were written to address this objective. Data were collected from July-November 2018 on a sample of children aged 3-5 years and a parent. Parents/guardians were recruited from Edmonton, Canada and surrounding areas through a local division of Sportball, a program that aims to teach children fundamental sport skills through play. In total, 131 parents/guardians agreed to participate. Children’s and parents’ movement behaviours were measured with waist-worn ActiGraph WGT3X-BT accelerometers.
  The objective of the first manuscript was to create a sleep (i.e., night and nap) classification technique. A total of 1,091,232,000 accelerometer observations in 30 Hz epochs were used to calculate 144 features (e.g., fast Fourier transformations, axis specific offset angles, kurtosis) aggregated to 1-minute epochs. Ground truth estimates of sleep were classified using visual inspection techniques. Random forest models were trained and tested using leave one subject out cross-validation, followed by temporally smoothing predictions with Hidden Markov Modeling. Additionally, a simplified prediction formula was created using 10 features with the highest mean decrease in Gini index during training of Random Forests, and temporally smoothed with rolling median calculations. Findings demonstrated that machine learning techniques could distinguish between sleep and wake with 96% accuracy, while the simplified formula reached 94% accuracy. Though, significant differences were found between machine learning and ground truth behaviour predictions for participant-level daily summaries, whereas non-significant differences were found between the simplified formulas and ground truth predictions.
  The objective of the second manuscript was to examine the relationships between accelerometer-derived movement behaviours and indicators of physical (i.e., motor skills, adiposity, and growth), cognitive (i.e., response inhibition, visual-spatial working memory, and vocabulary) and social-emotional (i.e., sociability, externalizing, internalizing, prosocial behaviour, and self-regulation [i.e., cognitive, emotional, and behavioural]) development using compositional analyses. Compositional linear regression models and compositional substitution models were conducted to examine the associations between movement behaviours and indicators of development. Findings confirmed the importance of moderate- to vigorous-intensity physical activity (MVPA) for physical development, while stationary time results were mixed for cognitive development outcomes.
  The objective of the third manuscript was to examine the associations of parental movement behaviours, parent-child proximity behaviours, and proximity movement behaviours with children’s movement behaviours using Bluetooth-enabled accelerometers. Child, parent, and proximity detection accelerometer files were merged and children’s movement behaviour variables were categorized as no proximity (NP), proximity but mismatching movement behaviours (Close), and proximity with matching movement behaviours (Co). Compositional and non-compositional analyses were utilized to examine patterns in children’s movement behaviours based on these contextual parent-child variables. Findings indicated parent-child movement behaviours were not associated, however close proximity was positively associated with children’s light-intensity physical activity (LPA), and NP-MVPA was positively associated with children’s MVPA.
  The findings within this dissertation make important contributions to movement behaviour research in preschool-aged children. Methods were presented to accurately classify daytime and nighttime sleep in preschool-aged children. However, future studies should replicate these findings using other ground-truth estimates of sleep (e.g., polysomnography). For relationships between movement behaviours and health/developmental outcomes, findings supported evidence of a favourable association between MVPA and physical development. Additionally, the associations between stationary time and cognitive development were mixed, so future research should examine sedentary behaviours (e.g., sitting, reading) and cognitive development to explain this heterogeneity. For relationships between correlates and movement behaviours, findings indicated that parent-child close proximity was associated with children’s LPA, while children’s NP-MVPA was associated with higher levels of total MVPA. Future research should measure the whole family unit to better understand the dynamics of the household that are associated with children’s movement behaviours. Taken together, it may be advantageous to promote independent MVPA in preschool-aged children. However, this dissertation used a cross-sectional study design from a convenience sample, so future research should test these findings with longitudinal or experimental studies in larger and more generalizable samples.

• Subjects / Keywords

  • Sleep
  • Sedentary Behaviour
  • Preschool
  • Physical Activity
  • Child

• Graduation date

  Fall 2020

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-30cn-eq23

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.