HUDM 6055 L. T. DeCarlo
Homework #3

An ‘illness’ path model presented in Kline is as follows:


1. How many unique elements are there in Σ? How many elements are there in θ? What are the
df for testing model fit?





2. Is the model identified? Exactly how did you determine this?







3. Fit the model in Lavaan using the standard deviations and correlations given below:

1.00
−.03 1.00
.39 .07 1.00
−.05 −.23 −.13 1.00
−.08 −.16 −.29 .34 1.00

SD’s: 66.5 38.0 36.8 67.0 62.48

The sample size is 373, the variables are exercise, hardy, fitness, stress, and illness.


How many parameters are in the baseline model, using Lavaan in R?



How does one get the degrees of freedom shown for the baseline model in Lavaan?





4. Get the standardized and normalized residuals. Which normalized residual is the largest? What
is it and which variables is it for?










What is the largest modification index? How might you modify the model in light of the MI and
residuals? Does your suggested modification introduce any problems?















5. If we were to use the alternative baseline model discussed in class and used in Mplus, then how
many parameters would there be in the baseline model and what would be the df for testing the
fit of the baseline model? Fit the alternative baseline model in Lavaan and report the chi-square
test of model fit and the df.





















































































学霸联盟