Effective maintenance/improvement of sleep quality requires knowledge of how sleep quality is connected to quantitative features of sleep and arbitrarily selected habitual lifestyles, which naturally depend on the demographic characteristics of individuals. To fulfill these needs, a regression model of subjective sleep quality was constructed, whereby one might be able to design a practical strategy for achieving comfortable sleep adapted to individual conditions. Based on data obtained from our previous study, fundamental correlation profiles between day-to-day subjective and quantitative features of sleep were estimated. Obtained correlation profiles involving SRSs, quantitative features of sleep, and sleep habits across a week such as bedtime preference (chronotype), discrepancy between chronotype and social time cue (social jetlag), and habitual sleep-wake pattern (HSWP) were characterized specifically for each self-ratings of sleep quality (SRS) category through backward stepwise Linear Mixed Effect (LME) modeling. The LME model represented SRSs with acceptable accuracy, allowing identification of determinant factors for each category of SRS. The SRS is one possible option to clarify sleep status. In this study, we proposed a possible framework including model-based predictors of SRS in which self-awareness of sleep quality could be improved to facilitate healthy sleep practices.