Do scales for measuring morningness-eveningness exist all over the world?

Morningness-Eveningness has been theoricized as a continuum along which subjects can be arranged or classified into coarsely determined categories: owls (or Evening-types), intermediate and larks (or Morning-types). This trait is thought to be universal chiefly because it is biologically and genetically determined. The number of categories changes to 5 after adding the “Definitively Evening” and “Definitively Morning” extreme types. This procedure is justified for epidemiological, clinical and research purposes, is common in psychology but raises a number of concerns we shall discuss in this chapter. To proceed, one first needs some valid instrument: the Morningness-Eveningness Questionnaire is the most widely used scale though others exist that deserve attention. Reliability is a necessary but insufficient property on the road toward validity for a scale’s score. Next, beyond the fact that the common used methods for categorisation are not theoretically sound or equivocal (i.e. for what reasons should we use extreme deciles or extreme quartiles? how are the cut-off scores determined?), the most important criticism is that demographical variables such as age and gender are not taken into consideration. Some recent data emphasize that evening-types are not rare in middle-aged, an erroneous conclusion derived from the rigid categorization. To say it simpler, norms for Morningness scores have not been determined. Moreover, we as others have claimed that cross-cultural validity is a desirable property for scales intended to measure personality traits including Morningness-Eveningness. This property might be extended to the factors extracted in the studied cultures leading to the problem of factorial invariance. In this chapter, we review the existing scales and their general properties in different cultures before focusing on the problem of categorization with examples showing its limits (e.g. lack of coherence between types determined on different scales). We will eventually propose some remedies. Finally, we will address the issue of cross-gender and cross-cultural factorial invariance of these scales. We conclude that the way is long that will lead to between groups valid comparisons in terms of Morningness-Eveningness scores and/or frequencies of the corresponding types.