The aim of this review article is to give a comprehensive description
of the scaling properties detected for the distribution of cosmic stru
ctures, like galaxies and galaxy clusters. I will also discuss the mor
e popular theoretical models, which have been proposed to account for
the huge body of observational data. Due to the great variety of stati
stical methods, developed in the last twenty years to statistically de
scribe the large-scale structure of the Universe, I will mainly concen
trate on those methods which reveal remarkable regularities and scalin
g in the structure of the Universe. Although in most cases I prefer no
t to enter into the technical aspects of how implementing such methods
, more details will be furnished about the description of galaxy clust
ering in terms of fractal concepts. Statistical methods based on fract
al analyses have been recently employed in cosmological context. Despi
te recent claims for a Universe, which behaves like a fractal at arbit
rarily large scales, I will show that the fractal language can be usef
ully employed to disprove this picture. The emerging scenario is that
of a Universe, which behaves like a self-similar structure at small sc
ales, where fractality is dynamically generated by non-linear gravitat
ional clustering, while preserving large-scale homogeneity. Neverthele
ss, even at scales greater than or similar to 10h(-1) Mpc, where gravi
ty still acts linearly, the distribution of galaxy clusters shows rema
rkable scale-invariant features, which could give precise hints about
the initial conditions for the evolution of the large-scale structure
of the Universe.