Local homogeneity in latent trait models. A characterization of the homogeneous monotone IRT model (Q1309413)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Local homogeneity in latent trait models. A characterization of the homogeneous monotone IRT model
scientific article

    Statements

    Local homogeneity in latent trait models. A characterization of the homogeneous monotone IRT model (English)
    0 references
    0 references
    18 September 1994
    0 references
    The authors consider latent trait models that have the following four properties: (1) unidimensionality, (2) local independence, (3) monotonity, (4) local homogeneity. In the case of binary items assumptions (1)--(4) are called the homogeneous monotone item response theory (IRT) model. It is shown that the property (4) is equivalent to subpopulation invariance of the model. This condition is studied for the monotone IRT model [\textit{P. W. Holland}, ibid. 55, 577-601 (1990; Zbl 0736.62094)]. The main result is the following characterization theorem: The homogeneous monotone IRT model holds for a finite or countable pool of binary items if and only if the pool is experimentally independent and pairwise nonnegative association holds in every positive subpopulation. In the conclusion the authors discuss a property of local homogeneity for different situations.
    0 references
    0 references
    homogeneous monotone item response theory model
    0 references
    latent trait models
    0 references
    unidimensional
    0 references
    local independence
    0 references
    monotonity
    0 references
    local homogeneity
    0 references
    binary items
    0 references
    subpopulation invariance
    0 references
    pairwise nonnegative association
    0 references
    0 references