Breeding bird communities were studied in eight habitat types in the Majella massif (central Italy) in two years (2007 and 2008) with two sampling methods (general survey and random point counts). At level of the general survey, our data confirm as bird communities are structured in response to vegetation structure and internal heterogeneity induced either by human activities (e.g., landscape patchiness) or natural limiting factors (e.g., linked to altitude). Richer habitat types correspond to mosaics of oak woods in both of years. In this habitat, the high level of habitat heterogeneity could increase the species richness at landscape level (i.e., γ-diversity). The low values in species richness observed in Pinus mugo formations and in primary pastures may be due to the altitudinal gradient of species richness (inverse correlation between altitudinal mid-point of each habitat type and number of species). Cluster analysis showed a first level of dissimilarity among faunal communities linked to the vegetation structure (forest versus open/shrubby habitat types), and a second one for open/shrubby habitats could be based either on spatial-heterogeneity and altitudinal effect. At level of species richness, a large proportion of the species (> 80%) has been sampled with random point count method when compared to general survey. Therefore, at species richness level, the method with lower research effort could furnish a reasonable description of the communities. At species composition level, data obtained by random sampling point count method showed different pattern between years for several habitat typology, turning out less exhaustive than general survey method. Therefore, in those habitat types a low research effort could expose the data to stochastic oscillations.