Systematics
The starlings belong to the superfamily Muscicapoidea, together with thrushes, flycatchers and chats, as well as dippers which are quite distant and Mimidae (thrashers and mockingbirds). The latter are apparently the Sturnidae's closest living relatives, replace them in the Americas, and have a rather similar but more solitary lifestyle. They are morphologically quite similar too—a partly albinistic specimen of a mimid, mislabelled as to suggest an Old World origin, was for many decades believed to represent an extinct starling (see Rodrigues Starling for details).
The oxpeckers are sometimes placed here as a subfamily, but the weight of evidence has shifted towards granting them full family status as a more basal member of the Sturnidae-Mimidae group, derived from an early expansion into Africa.
Usually the starlings are considered a family, as is done here. Sibley & Monroe (1990) included the mimids in the family and demoted the starlings to tribe rank, as Sturnini. This treatment was used by Zuccon et al. (2006). However, the grouping of Sibley & Monroe (besides leaving the subfamily rank vacant) is overly coarse due to methodological drawbacks of their DNA-DNA hybridization technique and most of their proposed revisions of taxonomic rank have not been accepted (see for example Ciconiiformes). The all-inclusive Sturnidae grouping is all but noninformative as regards biogeography, and obscures the evolutionary distinctness of the three lineages. Establishing a valid name for the clade consisting of Sibley/Monroe's "pan-Sturnidae" would nonetheless be desirable to contrast them with the other major lineages of Muscicapoidea.
Starlings probably originated in the general area of the East Asia, perhaps towards the southwestern Pacific, as evidenced by the number of plesiomorphic lineages to occur there. Expansion into Africa appears to have occurred later, as most derived forms are found there. An alternative scenario would be African origin for the entire "sturnoid" (as per Zuccon et al. 2006) group, with the oxpeckers representing an ancient relict and the mimids arriving in South America. This is contradicted by the North American distribution of the most basal Mimidae.(Cibois & Cracraft 2004, Zuccon et al. 2006)
As the fossil record is limited to quite Recent forms, the proposed Early Miocene (about 25–20 mya) divergence dates for the "sturnoids" lineages must be considered extremely tentative. Given the overall evidence for origin of most Passeri families in the first half of the Miocene, it appears to be not too far off the mark however.(Zuccon et al. 2006)
Recent studies (Cibois & Cracraft 2004, Zuccon et al. 2006) identified two major clades of this family, corresponding to the generally drab, often striped, largish "atypical mynas" and other mainly Asian-Pacific lineages, and the often smaller, sometimes highly apomorphic taxa which are most common in Africa and the Palearctic, usually have metallic coloration, and in a number of species also bright carotinoid plumage colors on the underside. Inside this latter group, there is a clade consisting of species which, again, are usually not too brightly-colored, and which consists of the "typical" myna-Sturnus assemblage.
Interestingly, the Philippine creepers, a single genus of three species of treecreeper-like birds appear to be highly apomorphic members of the more initial radiation of the Sturnidae (Zuccon et al. 2006). While this may seem odd at first glance, their placement has always been contentious. In addition, biogeography virtually rules out a close relationship of Philippine creepers and treecreepers, as neither the latter nor their close relatives seem have ever reached the Wallacea, let alone the Philippines. Nonetheless, their inclusion in the Sturnidae is not entirely final and eventually they may remain a separate family.
Genus sequence follows traditional treatments. This is apparently not entirely correct, with Scissirostrum closer to Aplonis than to Gracula for example, and Acridotheres among the most advanced genera. Too few taxa have already been studied as regards their relationships however, and thus a change in sequence has to wait.
The review by Lovette & Rubenstein (2008) is the most recent work on the phylogeny of the group.
Read more about this topic: Starling