The influence of habitat type upon reef fish assemblage structures has been extensively reported, but there is still a need to quantitatively demonstrate the factors driving reef fish assemblages within and among habitats. This mensurative study examined diversity and community composition of reef fish assemblages between five habitats (shallow kelp, deep kelp, sand, algal turf, and sponge fiats) within the Cape Rodney‐Okakari Point Marine Reserve, northeastern New Zealand. Two contrasting sampling techniques, point count and timed search were used to quantify a range of species with different life history strategies and characteristics. Diversity generally decreased from kelp to sand habitats. Different habitat types displayed significantly different assemblages irrespective of the sampling technique used, as tested using two different multivariate techniques: constrained canonical analysis of principal coordinates and analysis of similarities (CAP, ANOSIM). Differences in depth between defined habitat types explained a significant proportion of variability in fish assemblage data (17% and 28% for the point count and timed search data, respectively), although a large proportion of variability was unexplained. A core group of species were found to be best discriminators among habitats, including Meuschenia scaber, Parapercis colias, Upeneichthys lineatus, Cheilodactylus spectabilis, Notolabrus celidotus, Notoclinops segmentatus, Forsterygion lapillum, and F. varium. Sand habitat was characterised by a general absence of species. This study examined analytical methodology for characterising reef fish assemblages using existing statistical procedures and demonstrates the ability to detect significant differences in fish communities among different nearshore habitats along the northeastern coast of New Zealand. Further monitoring and measuring of other environmental parameters would allow the creation of an explanatory model for relative fish community abundances for the northeastern coast of New Zealand.