Daylight offers visual comfort and a healthy environment inside buildings. However, factors like changing sky luminance patterns, occupant behaviour influenced by age and eyesight, daily activity schedule, building location and orientation, building plan, interior design and other socioeconomic factors influence the requirement of artificial illumination during daytime. Therefore to understand the causes in greater detail, a primary survey of 707 households belonging to different housing typologies like bungalows, duplexes, twin houses, row houses and apartments of the linear and angular forms was carried out in the study site. To calculate the hours of low natural illumination within a room, LOWLUX, a MATLAB based code has been developed. Sky classification for the site was done based upon the all weather model' by Perez. As revealed from the primary survey, factors identified as important in influencing artificial lighting during daytime are sky conditions, age, work hours, education, income and housing typology. A binary logistic regression applied to the database to predict whether people would switch on a light at daytime in the living room, revealed that the need was least in the Angular apartments and highest in the Bungalows. A similar model for kitchen revealed highest daytime artificial illumination requirement for duplexes with lowest for angular apartments. (C) 2015 Elsevier Ltd. All rights reserved.