Proving is the method by which homeopathic remedies are "tested" for their effects. Throughout a proving, several people, usually about a dozen, will take the remedy and record their thoughts, feelings and even dreams. From these diaries, homeopaths will "discover" what the remedy is good for curing and can happily announce that their remedy has undergone a "trial" - although one barely recognizable as a trial to any qualified physician or researcher.
A proving records the experiences of the "provers", usually a dozen or so people who consume the remedy over a period of time. Their thoughts, feelings, dreams and habits are then recorded in what is often termed the "Materia Medica" (from the Latin meaning "collected body of knowledge") and analysed. An example from the homeopathic proving of positronium is outlined below:
01P 01 17.00 NS
- A sense of solidity.
01P 01 17.00 NS
- I moved my neck in a circle, it felt thick and as if my head was the same size - like a turtle.
02P 01 20.40 NS
- Speech feels slow and deliberate, and very exact in a quiet and weighty way.
02P 34 22.00 NS
- Mood heavy and drained.
02P 34 XX.XX NS
- Feeling OK till supper. After supper a tremendous feeling of exhaustion and heaviness. Legs feel like they're buckling under me. I can hardly walk. Head feels slow and heavy. Food in my digestive tract feels like a stone. Can hardly think, eyes falling. Can't feel anything - too exhausted to even drag myself along.
04P 99 XX.XX NS
- I've put on weight as I've been eating crap.
As can be seen, this is quite solid medical information and not at all subjective opinion. The numbers refer to the identity of the provers and time that they experienced these sensations. The letters are codes to describe the state of the symptom, whether new or old.
- NS A new symptom never before experienced.
- OS An old symptom previously experienced, but not in the preceding year.
- RS A recent symptom experienced within the last year.
- AS An altered symptom, one previously experienced but with at least one quality changed.
- CS A cured symptom, a symptom that was removed during the proving.
- IOS An old symptom that is felt with significantly greater intensity than before.
As the "symptoms" are most often described in the most abstract sense possible, it's difficult to see how they can be classed as new or old except in the most subjective of senses. Identifying gaining weight as a "new" symptom seems to imply that the prover has never, at all, ever, gained weight from eating too much - a highly unlikely, if not impossible, situation. It's for precisely this sort of reason that established medical practices is highly suspicious of such "self reported" data.
 Comparisons with clinical trials
In clinical trials it is important that the results are hardened against any possible forms of bias. This can include selection bias, interpretation and the placebo effect. Therefore, at the very least, clinical trials compare an active medicine to a previous treatment or placebo treatment, and are randomised, that is, participants are assigned placebo or treatment with a true random distribution. In addition to this, there is blinding, whereby participants and researchers remain unaware of who gets what until the end, so that their bias cannot influence their interpretation of the results. Proving, by contrast, features few, if any, of these methods of control.