In this paper, we develop scale to measure the need for restoration of office employees in field studies that is not based on introspection; in particular a behavior based non-evaluative self-report scale (see Haans, Kaiser, & de Kort (2007) for a comparable application). The need for restoration scale is based on what Kaiser, Byrka and Hartig (2008) call Campbell's paradigm. It measures people's propensity to engage in restorative activities. The scale consists of multiple behaviors, which office employees can engage in, as means to fulfill their need for restoration. The probability of engaging in a restorative activity will depend on the person’s need for restoration and the efforts it costs to perform the behavior. Research has shown that office lighting can have an influence on the well-being and performance of employees (Knez, 1995; Boyce, 2003). In this research lighting effects on office employees are used as means to validate the need for restoration scale. In total 271 employees of eight offices with different lighting conditions filled out the questionnaire. A Rasch analyses was performed to calibrate the need for restoration scale. The convergent validity of the need for restoration scale was tested with existing established measures, like the fatigue assessment scale (FAS; Michielsen, de Vries, & van Heck, 2003) and Groningen sleep quality scale (GSKS; Meijman, de Vries- Griever, de Vries, & Kampman, 1985). The predictive validity of the need for restoration scale was tested by comparing the effect of the office lighting on the need for restoration, fatigue and sleep quality with multiple regression analyses. The results of the Rasch analyses showed that the items of the need for restoration scale could be described as a one-dimensional scale. The reliability of the scale was .77. The correlations between the need for restoration scale and the fatigue assessment scale (r = .68) and the Groningen sleep quality scale (r = .31) indicated that the need for restoration scale measures a comparable construct as the fatigue assessment scale and the sleep quality scale. Multiple regression analyses with personal characteristics, evaluation of the work situation and lighting quality as predictors showed that the lighting had an effect on the need for restoration, but not on the fatigue or sleep quality of office employees. Thus the need for restoration scale was more sensitive to lighting effects.