--- - |2 Abstract It has long been noticed that older people tend to be more religious than younger people. However, it is still disputable whether this fact should be attributed to people generally becoming more religious with age per se (age effect), or to the process of secularization, wherein earlier cohorts (to which the now older people belong) used to be more religious than those that appeared later, younger cohorts (cohort effect). We try to distinguish between these two effects using a multifactor model applied to World Values Survey data (1981–2014) and find that at least in the developed countries the age effect strongly prevails over the cohort effect. This finding has important implications, e.g., that population aging in OECD countries can possibly slow down the transition from religious to secular values. This effect is already visible in some countries, such as Japan. - 'Journal for the Scientific Study of Religion, Volume 58, Issue 3, Page 591-603, September 2019. '