# 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.]