Longitudinal Modeling With Randomly and Systematically Missing Data : A Simulation of Ad Hoc, Maximum Likelihood, and Multiple Imputation Techniques  

Oleh:

Newman, Daniel A.

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: Organizational Research Methods vol. 6 no. 3 (2003), page 328-362.
Topik: modelling; organizational behaviour; comparative studies; mathematical models; monte carlo simulation; statistical analysis; effectiveness
Fulltext: 328.pdf (217.87KB)

Ketersediaan

Perpustakaan Pusat (Semanggi) Nomor Panggil: OO3.4 Non-tandon: 1 (dapat dipinjam: 0) Tandon: tidak ada

Lihat Detail Induk

Isi artikel

For organizational research on individual change, missing data can greatly reduce longitudinal sample size and potentially bias parameter estimates. Within the structural equatio modeling framework, this article compares six missing data techniques (MDTs) : listwise deletion, pairwise deletion, stochastic regression imputaiton, the expectation - maximization (EM) algorithm, full information maximization likelihood (FIML), and multiple imputation (MI). The rationale for each technique is reviewed, followed by monte carlo analysis based on a three wave simulation of organizational comment and turnover intentions. Parameter estimates and standard errors for each MDT are contrasted with complete data estimates, under three mechanisms of missingness (completely random, random and non random) and three levels of missingness (25%, 50%, and 75%, all monotone missing). Results support maximum likelihood and MI approaches, which particularly outperform listwise deletion for parameters involving many recouped cases. Better standard error estimates are derived from FIML and MI techniqiues. All MDTs perform worse when data are missing non randomly.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali