Stars are organised into galaxies which in turn appear to form clusters and superclusters, separated by voidss. Prior to 1989 it was commonly assumed that the superclusters were the largest structures in existence, and that they were distributed more-or-less uniformly throughout the universe in every direction. However, in 1989, Margaret Geller and John Huchra discovered the "Great Wall", a sheet of galaxies more than 500 million light years long and 200 million wide, but only 15 million light years thick. The existence of this structure escaped notice for so long because it requires locating the position of galaxies in three dimensions which involves combining location information about the galaxies with distance information from redshift.

In more recent studies the universe appears as a collection of giant bubble-like voids separated by sheets and filaments of galaxies, with the superclusters appearing as occasional relatively dense nodes.

At the centre of the local supercluster there is a gravitational anomaly, known as the Great Attractor, which is drawing in galaxies over a region hundreds of millions of light years across. These galaxies are all redshifted, in accordance with the 'Hubble flow', as if they were receding from us and from each other, but the variations in their redshift are sufficient to reveal the existence of a concentration of mass equivalent to tens of thousands of galaxies.

The Great Attractor, discovered in 1986, lies at a distance of between 150 million and 250 million light years (250 million is the most recent estimate), in the direction of the Hydra and Centaurus constellations. In its vicinity there is a preponderance of large old galaxies, many of which are colliding with their neighbours, and/or radiating large amounts of radio waves.

Another indicator of large-scale structure is the 'Lyman alpha forest'. This is a collection of absorption lines which appear in the spectral lines of light from quasars, which are interpreted as indicating the existence of huge thin sheets of intergalactic (mostly Hydrogen) gas. These sheets appear to be associated with the formation of new galaxies.

Finally, there is evidence of quantisation of redshift. There have been numerous studies investigating this phenomenon, but it is not universally accepted as real, and is the subject of considerable controversy.

Some caution is required in describing structures on a cosmic scale because things are not always as they appear to be. Bending of light by gravitation can result in images which appear to originate in a different direction from their real source. This kind of optical illusion can obscure the actual processes taking place. Another possible optical illusion is where a galaxy cluster will contain galaxies with some random motion. When these random motions are converted to redshifts, the cluster will appear elongated, and this creates what is known as a finger of God; the illusion of a long chain of galaxies pointed at Earth.

There is much work in cosmology which attempts to model the large-scale structure of the universe. Using the big bang model and assumptions about the type of matter that makes up the universe, it is possible to predict the expected distribution of matter, and by comparison with observation work backward to support and refute certain cosmological theories. Currently, observations indicate that most of the universe must consist of cold dark matter. Models which assume hot dark matter or baryonic dark matter do not provide a good fit with observations.