|
|
|
|
|
Jichuan Wang, PhD, Harvey Siegal, PhD, Russel Falck, MA, and Robert G. Carlson, PhD. Community Health, Wright State University, 3640 Colonel Glenn Hwy., Dayton, OH 45435, 937 775-2084, jichuan.wang@wright.edu
Although the latent growth modeling (LGM) has been increasingly used to study individual growth trajectories, applications of the LGM have usually focused on assessing the rate of outcome change (i.e., the latent slope growth factor), as well as the determinants of the variation in the rate of outcome change. In this study, LGM is applied to assess the mean outcome change during specific time periods within the entire observation period under study. Both the delta method and bootstrapping were used to estimate the standard errors of, thus test the significance for, the predicted sub-period-specific mean outcome changes via both single outcome LGMs and parallel LGMs. The results of the study show that the rate of outcome change may be statistically significant but a sub-period-specific mean outcome change may not be significant. In addition, individual characteristics, such as ethnicity, may have a significant effect on the rate of outcome change, but the ethnic difference in mean outcome change during a specific sub-period may not be significant. The study suggests that only assessing rate of outcome change and its variation in LGM is not sufficient to fully understand the growth processes of outcomes of interest. Sub-period-specific outcome changes are an important issue that needs to be addressed in latent growth modeling.
Learning Objectives: At the conclusion of the presentation, participants (learner) would be able to know
Keywords: Statistics, Outcomes Research
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.