A study of the reliability of systems by considering the ability of different systems to maintain a required air flow rate over time is included in a subtask of IEA Annex 27 ＇Evaluation and Demonstration of Domestic Ventilation Systems＇. Measurements and calculations were performed to determine the variation in ventilation rates due to variation in climate and variation in performance of the ventilation system. Dwellings with passive stack, mechanical exhaust and mechanical exhaust-supply ventilation, representative of the Swedish housing stock, were studied. Diagnostic tests were carried out, to discover if the installed ventilation system was functioning as designed e.g. air flows in mechanical ventilation systems and to determine certain values e.g. airtightness. The continuous monitoring included tracer gas measurements in dwellings of overall and local (individual rooms) ventilation rates, and measurements of boundary conditions, during three different periods, each lasting 1-6 days. Predictions of air flows were made for the measuring periods using COMIS, a multi-zone network model. This article presents and discusses the measurements and the calculations and compares the two. The predicted and measured average total outdoor air ventilation rates agree reasonably well. The agreement is less good for individual rooms.