|
Important Characteristics and Features |
Home  Research Areas Intensive Longitudinal Data Important Characteristics and Features
- Data have a multilevel structure. For example, occasions are nested within individuals, who may be nested within families, or classrooms; and families are nested within community.
- Covariates are either time-invariant or time-varying. A covariate is called time-varying if its value changes across time, e.g., mood changes over time; daily stressors vary from day to day; levels of air pollutants are different from day to day. A covariate is called time-invariant if it always takes the same value during a study. For instance, gender, race and social economic status are time-invariant.
- Not only the values of the covariates, but also the nature of their relationship to the response, change over time, i.e., there exist time-varying effects. For example, consider the relationship between mood and ad lib smoking. The strength and even the direction of this relationship may change across time.
- Effects vary not only across time but also between individuals. Different people may be affected differently by certain situations. For example, the relationship between mood and daily smoking may change over time and vary across individual, since some people smoke while having fun with their friends but others smoke because when are lonely or frustrated.
- The structure of the error process changes over time. The error variance may dramatically change over time, making the classic linear modeling assumption of a constant variance term untenable.
- The population may be a combination of several discrete latent subpopulations. This is related to property 4. For example, for some smokers, mood may have strong effect on urge to smoke, while it may have only weak effect for some other smokers. If there are qualitatively different groups in a population, then this should be taken into account in the model in order to avoid glossing over important relationships.
|
