Exercise

Absolute fit statistics

The absolute fit statistic reported in the CFA output is the likelihood ratio Chi-square test, which tests the hypothesis that the estimated model works as well as the null model. You've seen this fit statistic in the previous chapter, so you probably remember that the criteria for this test are very difficult to meet. Most models have a statistically significant result, which is not desirable here, since it means model fit is statistically worse.

After checking the absolute model fit statistics displayed by the summary() function, determine whether the likelihood ratio test is statistically significant.

Would the GFI for this model be considered acceptable?

Instructions 1/2

undefined XP
    1
    2
  • Set the fit.indices option to include several common fit indices.
  • View the summary output from the theory-based CFA and look for the likelihood ratio test.