Factorial Structure of the Morningness-Eveningness-Stability-Scale (MESSi) and Sex and Age Invariance.

Assessing morningness-eveningness preferences (chronotype), an individual characteristic that is mirrored in daily mental and physiological fluctuations, is crucial given their overarching influence in a variety of domains. The current work aimed to investigate the best factor structure of an instrument recently presented to asses this characteristic: the Morningness-Eveningness-Stability-Scale improved (MESSi). For the first time, the originally proposed three-factor structure was pitched against a uni- and a two-factor solution. Another novelty was to establish that the best-fitting model would be invariant in relation to sex and age, two variables that influence chronotype. A Confirmatory Factor Analyses on the data obtained from a sample of 2096 German adults (age: 18-76; = 25.5, = 7.64) revealed that the originally proposed three-factor structure of the MESSi - Morning Affect, Eveningness, and Distinctness - was the only one to achieve acceptable fit indicators. Furthermore, each scale obtained good internal consistency. In order to assess age invariance, following the literature on development and chronotype, our sample was divided into three age groups: 18-21 years, 22-31 years, and 32 years or older. Full measurement invariance of the three-factor model was found for sex and age. Regarding differences between sexes, females did not differ significantly from males in Morning Affect, but scored significantly lower on Eveningness and higher on Distinctness; this last result has been consistent across validation studies of the MESSi. With respect to age differences, the oldest group scored lower on Eveningness and Distinctness in comparison with the other two age-groups; the intermediate group (age: 22-31) scored lower on Morning Affect when compared to both the younger and older age groups. Additionally, both Eveningness and Distinctness were negatively correlated with age. This latter relation has been consistently reported in other validation studies. Our results reinforce the idea that the MESSi assesses three different components of chronotype in a reliable manner and that this instrument can be used to explore sex and age differences.