HLM Example-Simple Example

Example from, Woltman, Feldstain, MacKay, Rocchi (2012). An introduction to hierarchical linear modeling. Tutorial in Quantitative Methods for Psychology, 8(1), 52-69.

30 basketball teams, 10 players per team
Three variables = Life satisfaction (outcome), shots, coach years of experience

Software (free student version):
http://www.ssicentral.com/hlm/downloads.html

SPSS Data:
Level 1
Level 2

Data Condition

 

Unconstrained (null) Model 

One-way ANOVA

Results:
Confirms that variability in outcome, by level 2 group, is different from 0.0.
ICC = 14.96/(14.96+14.61) = .51 (51% of the variance in outcome is at the group level (quite large)

 

Random Intercepts Model

Note: Group center Shots_on & click on u1

Results:

Effect size (r2) = (14.61-4.6)/14.61 = .715 [note that 14.61 came from the null model]
Confirms that variability in outcome and slope, by level 2 group, is different from 0.0.

Means as Outcomes Model

Note: COACH_EX is grand-mean centered

 

Results:

Effect size (r2) = (14.61-1.68)/14.61 = .888 [note that 14.61 came from the null model]
Coaching experience explains 88.8% between variance in life satisfaction

Random Intercepts and Slopes Model