Values of travel time savings are often used in cost-benefit analysis of transport projects and policies, and also to compute generalised travel costs. There has been considerable debate as to whether different research methods (e.g. stated versus revealed preference) will lead to different values of travel time savings, and which segmentations (e.g. by income or mode) are most important to capture the heterogeneity in these values. In addition there are many countries where no specific valuation studies have been done. In this paper new equations are estimated on the outcomes of value of travel time savings studies from various countries. In the data set, several countries appear more than once, which is taken into account by estimating random effects panel models. The meta-analysis sheds some new light on the variation of the value of travel time savings by income, country, travel purpose, mode, distance and by survey method. Furthermore, the resulting meta-models are applied to produce new values of travel time savings for business travel, commuting and for other purposes in passenger transport, for 25 European Union Member states. Similar methods could be used to statistically analyse studies carried out on other non-monetary effects, both for transport and non-transport projects, and for inclusion in cost-benefit analysis.