Multilevel Modeling

4. Multilevel Methods and Analyses

Exercise 2

Goal of this exercise is to engage reader in activity identifying the importance of hierarchy.

Below are a series of statements on the possible data structures, some are hierarchical and some are not – match the most appropriate category to each statement.

Categories:

  • Single Level Data Structure
  • Two Level Data Structure
  • Three Level Data Structure
  • Non Hierarchical, multiple membership
  • Cross Classified Model

Statements:

  • Cross-sectional data on a random sample of individuals.
  • Cross-sectional data on a random sample of individuals from a random sample of households from a random sample of neighborhoods.
  • Cross-sectional data on a random sample of individuals from a random sample of neighborhoods.
  • Cross-sectional data on a multiple responses for a random sample of children in a single classroom.
  • Cross-sectional data on a random sample of children belonging to a random sample of teachers, with the possibility of children being taught by multiple teachers.
  • Cross-sectional data a random sample of individuals drawn from a random sample of neighborhoods and worksites.
  • Data on repeated measures for an individual.

Answers:
Cross-sectional data on a random sample of individuals.
[Single Level data structure]

Cross-sectional data on a random sample of individuals from a random sample of households from a random sample of neighborhoods.
[Three level data structure of individuals at level-1 nested within households at level-2 nested with neighborhoods at level-3.]

Cross-sectional data on a random sample of individuals from a random sample of neighborhoods.
[Two level data structure of individuals at level-1 nested within neighborhoods at level-2.]

Cross-sectional data on a multiple responses for a random sample of children in a single classroom.
[Two level data structure of multiple responses at level-1 nested within individuals at level-2]

Cross-sectional data on a random sample of children belonging to a random sample of teachers, with the possibility of children being taught by multiple teachers.
[Two-level non hierarchical multiple membership data structure of children at level-1 nested within teachers at level-2 with the nesting being weighted if a child is taught by multiple teachers. The weight can be the time spent by the child with different teachers.]

Cross-sectional data a random sample of individuals drawn from a random sample of neighborhoods and worksites.
[Two-level cross-classified model with individuals at level-1 nested with neighborhoods AND worksites at level-2, such that workers from a single neighborhood can be working in multiple worksites, and workers in a given worksite can be coming from multiple neighborhoods.]

Data on repeated measures for an individual.
[Two-level data structure of repeated measures or time (at level-1) nested within individuals at level-2.]