Research studies in this area have typically asked job holders at a single point in time to evaluate their current situation in relation to their recalled previous expectations, rather than longitudinally recording expectations in advance of later reality. For example, Taris et. al. (2005) obtained from employees direct assessments of 15 job characteristics in comparison with the level they had expected (from “much worse than expected” to “much better than expected”). Met expectations based on recall were on average correlated at +0.31 with overall job satisfaction, and the association remained significant despite controls for levels of the job features themselves and for individuals’ rating of the importance of those features.