Tea is a widely consumed drink among elderly people in Chinese population. The consumption is growing in different ages because of the founding of many handy-tea-drink stores in Taiwan. However, determination of tea quality is still mainly based on sensory evaluation, which lacks of objectivity in science. Near infrared (NIR) spectroscopy, a rapid nondestructive inspection method, has been widely applied for evaluation of internal quality of agricultural products. Because an NIR spectrum of a mixture on first approximation is the linear addition of individual spectra of the constituents in the mixture, such a spectrum thus can be regarded as an assembly of ‘blind sources’ as the proportion of constituents in the samples remains unknown. Independent component analysis (ICA), a multiuse statistical approach originally used to implement ‘blind source separation’ in signal processing, is capable of disassembling the mixture’s signals with only a small loss of information and does not require any additional information from the source. To date, ICA has not been applied to analysis of the internal quality of tea. Therefore, the objective of the current study was to examine internal quality of tea in terms of quantitative approaches using NIR spectroscopy combined with ICA technique. The internal quality-related indices of tea, including tea polyphenols (TP), free amino acids (FAA), water, and pH value were evaluated simultaneously. The results show that ICA with NIR has the potential to be adopted as an effective method for evaluating internal quality of tea.