Web-based technologies are changing the culture of universities; challenging long held traditions of university teaching, students' attendance patterns and ways of learning. Combine this with the relatively recent worldwide trend that more students are finding it necessary to engage in paid work while studying and for longer hours, is it any wonder that class attendance is dropping as students increasingly rely on web resources to learn content? This paper seeks to quantify this phenomenon and the impact of class attendance on academic performance using statistical modelling, based on data from 158 business/commerce undergraduates studying a 2nd year business mathematics course at Griffith University in Brisbane. After controlling for maths ability with a pretest, the results showed that reduced class attendance had a strong negative effect on academic performance. Associated with these findings were the different patterns of participation in paid work exhibited by males and females as well as by domestic and international students, and the different attendance behaviour of these groups.