Talk:Latent variable model

I don't think that Latent variable model and Latent variable should be merged - they're different concepts, and neither is obviously a subcategory of the other. The articles currently contain quite different information, so this isn't a case where there is duplication of effort happening. I believe (thought it is not my field of mathematics) that the use of the term latent variable is much wider than its article implies, and includes situations quite unrelated to latent variable models. Chrisjohnson 01:38, 8 April 2007 (UTC)[reply]

I agree that the articles should not be merged. The definition of Latent variables is commpletly unrelated to how you chose to use them. Therefore it may be misleading to link the definition to a particular class of models.74.60.48.250 22:22, 27 June 2007 (UTC)graciata[reply]

I just don't agree. These are 2 different things. —Preceding unsigned comment added by 70.55.147.246 (talk) 19:43, 22 November 2007 (UTC)[reply]

While they are two different things they are so closely linked that they should be one article, as is the case in the Encyclopedia of Statistical Science. All the description needed of a latent variable is that it is unobserved, everything else relates to the models, as without a model there can't be any manifest variables. Kjbeath (talk) 03:58, 22 December 2007 (UTC)[reply]

I agree with Kjbeath that the two articles should clearly be merged. A latent variable model simply is a model which includes latent (unobserved) variables. It doesn't make sense to talk of latent variable models outside of the context of latent variables. Similarly, the latent variables only exist in the context of a model. It is nonsensical to talk of a latent variable without reference to the model of which it is a part. A wikipedia reader who understands the concepts would never decide to read an article on latent variable models but also decide that latent variables are not of interest, or vice versa. The two concepts are inextricably linked.Satyr9 (talk) 23:39, 16 April 2014 (UTC)[reply]

The trouble with merging Latent variable model and Latent variable is that the concept of a 'Latent variable' is used in many different branches of science and engineering, but the term 'Latent variable model' is only used by social scientists and this article reflects that - it only considers item response theory and other psychometric theories. Nothing about quantum mechanics (see Hidden variable theory and Hidden variable), hidden Markov models, machine learning (e.g, Probabilistic latent semantic analysis), bioinformatics, etc. --Mark viking (talk) 03:21, 27 October 2014 (UTC)[reply]

Use in psychology

The articles says "Therefore there may be a psychometrical latent variable, but not a psychological psychometric variable". Is there an example to make this clearer? 92.0.230.198 (talk) 15:09, 27 June 2015 (UTC)[reply]