You can also download a
of this lecture
Low Signal to Noise (S/N) is unfortunately a frequent situation in mm astronomy. Since S/N increases only as the square root of the integration time, it may not be feasible at all to go from a case of low S/N to a case of even ``decent'' S/N. Thus the astronomer has to worry about interpreting properly low S/N data. The risk of over-interpretation in such cases is to be considered seriously. This lecture will give you some hints, as well as point out some traps which must be avoided.