We use data about job search and work preferences, typically collected in a Labour Force Survey, in order to construct an indicator of .choosiness. of the supply of job-seekers. The method for obtaining the indicator, first at individual level and then at aggregate levels, is based on results from multiple correspondence analysis. We investigate the informational value of the indicator by examining its stability over time and its predictive power on labour force transitions. Empirical analyses of cross-section and panel samples of job-seekers from the Italian quarterly Labour Force Survey clarify the potentials (and limitations) of the approach.