In this paper we want to compare the working time of teachers with that of other professionals. However, the apparently simple question on working time is much more complex than it seems at first sight. The question how many hours a week one works involves several steps and each of these steps can lead to a wrong answer (Robinson, Chenu & Alvarez 2002). An accurate assessment of working time implies that all respondents interpret' work 'in the same way, memorize all work moments and make the correct sum of these moments of work. It also presupposes that the respondents feel comfortable in reporting the total duration to the interviewer and do not adapt the total duration to social norms and expectations, stereotypes and ideas that exist on the number of hours that a 'normal' or 'good’ worker should have to perform. This applies to all questions to measure the time spent on specific activities, but even more to socially desirable activities such as work. Survey questions on these topics generally lead to over-reporting of time spend on it (Bonke 2004, Frazis & Stewart 2004, Glorieux, Minnen & Van Thielen 2004, Glorieux & Vandeweyer 2002, Robinson & Bostrum 1994, Robinson, Chenu et al 2002). Especially for teachers it is difficult to estimate working time on the basis of surveys (Chenu 2002, Menter et al. 2006). Even when both the number of hours in school and the time spent on preparation of school-related activities is asked, the results are flawed (Drago et al. 1999). The actual number of hours worked can be underestimated as well as overestimated because of several reasons, e.g.: hours spent at school when one does not work, school hours lest less than 60 minutes, flexible schedules and homework that make it difficult to estimate the correct working time, … . In this paper we therefore want to calculate the working time of teachers in a different and a more accurate way, namely by means of time-use data.

In this paper we want to calculate the working time of teachers by means of two time-use surveys. The first one was carried out in 1999 in Flanders, the second one in 2004. Both studies are based on a representative sample. In time-use studies respondents provide detailed information regarding all of their activities during a certain period. These activities are registered in a pre-coded time-use diary. In the time-use studies that are used here, respondents were asked to register every activity during one week. On the basis of these datasets we will compare the working time of teachers with that of other employees. These results will be compared to results based on other time measurement methods. Next to that we will portray the daily rhythm of working and compare the working rhythm of teachers with that of other professionals.

The working time reported in surveys is much higher, for both teachers and other employees, than the working time that is calculated based on a time-use diary. Working time is also higher when the researcher makes clear that the respondent’s job is under consideration or when only working activities have to be registered. The most reliable results are therefore obtained from general time use research. Further it is found that there is a large gap between the contractual working time and the real working time. Yet, the working time of teachers is certainly not higher than the working time of other employees. However, their working rhythm differs from most employees. Teachers have a more rigid working structure during the day but their work and private live blend into each other at the evening and during the weekends.