Modeling Longitudinal Count Data: Testing for Group Differences in Growth Trajectories Using Average Marginal Effects  

Oleh:

Mustillo, Sarah ; Landerman, Lawrence R. ; Land, Kenneth C.

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: Sociological Methods & Research (SMR) vol. 41 no. 03 (Aug. 2012), page 467-487.
Topik: Group Differences; Growth Curve Models; Marginal Effects; Count Data
Fulltext: S28 v41 n3 aug12 p467,win.pdf (147.44KB)

Ketersediaan

Perpustakaan PKPM Nomor Panggil: S28 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

To test for group differences in growth trajectories in mixed (fixed and random effects) models, researchers frequently interpret the coefficient of Group-by-Time product terms. While this practice is straightforward in linear mixed models, it is less so in generalized linear mixed models. Using both an empirical example and synthetic data, we show that the coefficient of Group-by-Time product terms in a specific class of mixed models—mixed Poisson models for count outcome variables—estimates the group difference in slope as the multiplicative change with respect to the baseline rates, not differences in the predicted rate of change between groups. The latter can be obtained from computing the marginal effect for the expected response with respect to time by group following model estimation. We propose and illustrate the use of marginal effects to test and interpret group differences in rate of change over time following estimation with mixed Poisson regression models.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali