Measuring the strength of teachers' unions: An empirical application of the partial independence item response approach

The literature on teachers' unions is relatively silent about the role of union strength in affecting important outcomes, due in large part to the difficulty in measuring union strength. In this article, we illustrate a method for obtaining valid, reliable, and replicable measures of union strength through the use of a Partial Independence Item Response (PIIR) model. This method uses the individual regulations found within collective bargaining agreements (CBAs) and models their existence as a function of a contract-specific latent level of restrictiveness. This allows for the estimation of (a) a measure of the latent level of relative union strength and associated standard errors of measurement; (b) the reliability of the measure; (c) item severities for each item addressed in the contract, allowing for assessment of item restrictiveness; and (d) contract and item information.