One of the biggest mysteries in cosmology is Dark Energy, which is required to explain the accelerated expansion of the universe within the standard model. But maybe one can explain the observations without introducing new physics, by simply taking one step back and re-examining one of the basic concepts of cosmology, homogeneity. In standard cosmology, it is assumed that the universe is homogeneous, but this is not true at small scales (<200 Mpc). Since general relativity, which is the basis of modern cosmology, is a non-linear theory, one can expect some backreactions in the case of an inhomogeneous matter distribution. Estimates of the magnitude of these backreactions (feedback) range from insignificant to being perfectly able to explain the accelerated expansion of the universe. In the end, the only way to be sure is to test predictions of inhomogeneous cosmological theories, such as timescape cosmology, against observational data. If these theories provide a valid description of the universe, one expects aside other effects, that there is a dependence of the Hubble parameter on the line of sight matter distribution. The redshift of a galaxy, which is located at a certain distance, is expected to be smaller if the environment in the line of sight is mainly high density (clusters), rather than mainly low density environment (voids). Here we present a test for this prediction using redshifts and fundamental plane distances of elliptical galaxies obtained from SDSS DR8 data. In order to get solid statistics, which can handle the uncertainties in the distance estimate and the natural scatter due to peculiar motions, one has to systematically study a very large number of galaxies. Therefore, the SDSS forms a perfect basis for testing timescape cosmology and similar theories. The preliminary results of this cosmological test are shown in this contribution.

8 pages, 5 figures, PoS conference proceedings: VIII International Workshop on the Dark Side of the Universe, June 2012, Buzios, Brasil