To my knowledge it was Paul Lazarsfeld who popularized the terminology of latent and manifest variables in measurement theory. The more interesting variables in our disciplines are always latent and we are never able to observe them in a direct way. Nobody has ever seen intelligence, knowledge of biology, or achievement motivation as such, and nobody ever will. What we have at hand is - at most - a collection of scattered behaviors or linguistic utterances expressing feelings or mental reactions to some situation, which in themselves are of no interest, but might be conceived of as allusions to the more interesting latent qualities we assume to be present.
However, simply because we talk about latent variables - and all textbooks in education or psychology are full of them - we may not assume that they exist, that they are just out there, and that we only have to collect data to measure them. Latent variables are no empirical quantities, but hypotheses about reality. The only way to get to know them is through a model that formulates hypothetical relations between the latent variables and certain observables. Social and behavioral scientists live in a hypothetical or latent space, and the only point of reference they have is a man-made model. We know that models can never be fully true, but that to some extent they are always idealizations, abstract schemes to represent reality.
At the recent AERA Annual Meeting in San Francisco a debate was held on "IRT in the 1990s: Which Models Work Best." I had the pleasure of participating as a discussant and enjoyed doing so - although I very much missed a tone of mutual respect among the participants. C'est le ton qui fait la musique! In the debate, I took the position that the first lesson to be learned from measurement theory is that we should be humble and not pretend to know more than we do. But I immediately added that the choice of model is as important as the choice of an Archimedean point, and that the Rasch model offers simplicity, interpretability, and plenty of statistical procedures with proven properties. Other models lack these features. I finally concluded that I might even use one of these other models, if I had to accept tests and data collected by others, but that I would do so suspicious of the results and knowing that a better world is possible.
I have been elected Chair of the SIG as successor to David Andrich, who finished his term at the Business Meeting in San Francisco. It is just fitting to thank him for the dedicated way in which he exercised his office. I am sure all of you will miss his insights as Chair, and so will I.
Latent variables, W van der Linden Rasch Measurement Transactions, 1992, 6:2 p. 213
Forum | Rasch Measurement Forum to discuss any Rasch-related topic |
Go to Top of Page
Go to index of all Rasch Measurement Transactions
AERA members: Join the Rasch Measurement SIG and receive the printed version of RMT
Some back issues of RMT are available as bound volumes
Subscribe to Journal of Applied Measurement
Go to Institute for Objective Measurement Home Page. The Rasch Measurement SIG (AERA) thanks the Institute for Objective Measurement for inviting the publication of Rasch Measurement Transactions on the Institute's website, www.rasch.org.
Coming Rasch-related Events | |
---|---|
Oct. 4 - Nov. 8, 2024, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
Jan. 17 - Feb. 21, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
May 16 - June 20, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
June 20 - July 18, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Further Topics (E. Smith, Facets), www.statistics.com |
Oct. 3 - Nov. 7, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
The URL of this page is www.rasch.org/rmt/rmt62l.htm
Website: www.rasch.org/rmt/contents.htm