SWUTC Research Project Description
Title of Project: A Study of the Inter-relationship between Parent’s and Children’s Physical Activity Participation Frequency
Project Number: 161020
Principal Investigator:
Chandra R. Bhat
(512) 471-4535
P.I. Affiliation: University of Texas at Austin
Project Monitor:
Johanna Zmud
NuStats LLC Corporate Office
206 Wild Basin Road
Building A, Suite 300
Austin, TX 78746
(512) 306-9065
Project Status: Active
Date Started: 9/1/09
Estimation Completion Date: 8/31/10
Estimated Cost - Current Fiscal: $39,000
Estimated Cost - Total Planned: $39,000
Project Summary:
Project Abstract:
In this research, we will study the question of how the daily frequency of physical activity participations of parents and children are inter-related, and how they are affected by urban form design and the built environment. The results from this research will aid policy makers in developing informed and effective policies that increase physical activity participations of adults and children (especially the children). In addition, an understanding of the dependency between parents’ and children’s activity-travel patterns will be important precursor to developing a good overall activity-based travel modeling system. Finally, the research will shed light on how we might be able to engineer our urban environment to encourage non-motorized modes of travel.
Project Objectives:
In this research, we focus on the family as a “cluster unit” when modeling the physical activity levels of individuals. In this regard, and because earlier physical activity studies have focused only on adults or only on children, our emphasis is on analyzing physical activity levels of families with one or more parents and children in the household. That is, we examine the determinants of physical activity in the context of family households with children. In doing so, we explicitly accommodate family-level observed and unobserved effects that may influence the physical activity levels of all individuals in the family. From a methodological standpoint, the daily number of physical activity episodes of each individual will be represented using an ordered response structure. The jointness between the episodes of different members of the same family will be generated by common household demographic and location variables, as well as through dependency among the stochastic error terms of the random latent variables assumed to be underlying the observed discrete number of physical activity episodes. In the past, several studies have considered such clustered ordered response systems by employing a normal mixing distribution to introduce dependencies among the error terms of the underlying propensities, but such an assumption assumes linear and symmetric error dependency patterns. The objective of our study will be to generalize the structure of the dependency by allowing non-linear and asymmetric error dependencies using a copula structure, which is essentially a multivariate functional form for the joint distribution of random variables derived purely from pre-specified parametric marginal distributions of each random variable. Adopting such a flexible dependency form does not entail any more computational burden than the typical multivariate normal ordered-response system.
Task Descriptions:
TASK 1: Literature review
In this task, we will first examine earlier studies in the area of understanding the determinants of physical activity participation, which will highlight the importance of the study and will contribute to the empirical analysis of our study. Then, we will identify the methodologies to deal with unobserved interactions due to cluster effects in an ordered response context, which will mainly contribute to the development of our methodological framework.
TASK 2: Development of conceptual framework
This task will involve the conceptual development of the model framework to accommodate the dependence in physical activity propensity among family members. To achieve this, the copula approach in general, and several copula types in use in particular, will be extensively examined. At the end of this task, the researchers will have all the background information needed to formulate the copula-based clustered ordered response model structure.
TASK 3: Formulation of the copula-based clustered ordered response model structure
The formulation of a comprehensive analytic structure to accommodate the dependence in physical activity propensity among family members will be a critical component of this project. Using the information gathered in the previous task, a copula approach will be formulated to generate a joint distribution of the daily number of physical activity episodes of each individual in a family. As stated before, the jointness between the episodes of different members of the same family will be generated by common household demographic and location variables, as well as through dependency among the stochastic error terms of the random latent variables assumed to be underlying the observed discrete number of physical activity episodes. This approach will allow testing of a variety of parametric marginal distributions for individual members in a cluster and will preserve these marginal distributions when developing the joint probability distribution of the cluster. The copula approach will separate the marginal distributions from the dependence structure, so that the dependence structure is entirely unaffected by the marginal distributions assumed. Thus, rank measures of the intra-cluster dependence of the underlying physical activity propensities for members of a family are independent of the marginal distributions used, facilitating a clear interpretation of the dependence structure regardless of the marginal distribution assumed. Under the copula approach, the assumption of a multivariate normal structure for the random terms is one of a suite of different types of couplings that can and will be tested.
TASK 4: Preparation of data and sample for analysis
The research will use the data drawn from the 2000 San Francisco Bay Area Household Travel Survey (BATS). This survey collected data on activity type of each episode at a fairly high degree of resolution, from which we will identify the total number of physically active episode participations during the day for children and parents in family households. The designation of an episode as physically active or physically passive will be based on the nature of the episode and the location type at which it is pursued, as reported in the survey. A number of variables will be constructed and considered as independent variables to explain the physical activity participations of children and their parents. These include household and individual demographic variables, household structure variables, work and school characteristics of individuals, and a whole set of built environment measures in the vicinity of the household residence. The last set of built environment variables will be developed using land-use/demographic coverage data, a Geographic Information System (GIS) layer of bicycle facilities, a GIS layer of highways and local roadways, and GIS layers of businesses as obtained from the InfoUSA business directory.
TASK 5: Model estimation and specification analysis
This task will involve the estimation of the model developed in Task 3 by using the data assembled in Task 4. Several different variables within the three broad variable categories of individual factors, physical environment correlates, and social environment determinants will be considered in our model specifications. The final model specification will be based on a systematic process of eliminating variables found to be statistically insignificant, intuitive considerations, parsimony in specification, and results from earlier studies. Several different variable specifications, functional forms of variables as well as interaction variables will be examined.
TASK 6: Asses policy implications of the results
In this final task, we will conduct a policy-based simulation to examine the magnitude of the influence of variables on the number of physical activity episode participations.
TASK 7: Prepare and submit final report
This task will prepare and deliver a final research report documenting the research performed, including the literature synthesis, methodology, data analysis, results, and recommendations.
Index Terms: