Sorry for this huge c&p:
Meta-analyses, in which large groups of studies are analysed and conclusions drawn based on the results as a whole, have been used to evaluate the effectiveness of homeopathy. Early meta-analyses investigating homeopathic remedies showed slightly positive results among the studies examined, but such studies have warned that it was impossible to draw firm conclusions due to low methodological quality and difficulty in controlling for publication bias in the studies reviewed.[12][13][19] One of the positive meta-analyses, by Linde, et al.,[13] was later qualified by the authors, who wrote:
The evidence of bias [in homeopathic trials] weakens the findings of our original meta-analysis. Since we completed our literature search in 1995, a considerable number of new homeopathy trials have been published. The fact that a number of the new high-quality trials...have negative results, and a recent update of our review for the most "original" subtype of homeopathy (classical or individualized homeopathy), seem to confirm the finding that more rigorous trials have less-promising results. It seems, therefore, likely that our meta-analysis at least overestimated the effects of homeopathic treatments.[14]
In 2001, a meta-analysis of clinical trials on the effectiveness of homeopathy concluded that earlier clinical trials showed signs of major weakness in methodology and reporting, and that homeopathy trials were less randomized and reported less on dropouts than other types of trials.[19]
In 2002, a review of systematic reviews found that higher-quality trials tended to have less positive results, to the point that those results were clinically irrelevant. Also, when taking collectively all the systematic reviews, there was no convincing evidence that any homeopathic remedy had better effects than placebo, and current evidence did not allow to recommend its usage in clinical treatment.[2]
In 2005, a systematic review of the representation of homeopathy in the medical literature suggested that mainstream journals had a publication bias against clinical trials of homeopathy that showed positive results, and the opposite was the case for complementary and alternative medicine journals. The authors suggested that this could be due to an involuntary bias, or otherwise a submission bias, in which positive trials tend to be sent to CAM journals and negatives ones to mainstream journals.[17] Reviews in all journals approached the subject in an apparently impartial manner, though most of the reviews published in CAM journals made no mention of the plausibility of homeopathy, whereas 9 out of 10 reviews in mainstream journals mentioned a lack of plausibility of homeopathy in the introduction.[17]
In 2005, The Lancet medical journal published a meta-analysis of 110 placebo-controlled homeopathy trials and 110 matched medical trials based upon the Swiss government's Program for Evaluating Complementary Medicine, or PEK. The study concluded that its findings were compatible with the notion that the clinical effects of homeopathy are nothing more than placebo effects.[6]
From: here where you can get all the links to the references.