I believe in ghosts (or specters, spirits, apparitions, or whatever else you choose to call them). I believe in ghosts because I have had personal experience with ghosts and I am personally acquainted with others who have also. So you'd think that with my personal experience I would be all over these ghost shows clogging up the airwaves, wouldn't you? Well, you'd be wrong. As a matter of fact, most of the ghost shows on TV are a complete waste of time.

I'm specifically talking about the ghost hunting shows; Ghost Adventures, Ghost Hunters /International, Paranormal State, etc. These shows drive me crazy because, well frankly, they're doing it wrong! That and, seriously, could they possibly be more obnoxious?

If you're going to go ghost hunting here's how you should do it:

1. Find a group of people who are calm, intelligent, and not likely to freak out thinking it's a ghost if someone farts (Most Haunted I'm looking at you).
2. Be prepared to investigate over an extended period of time; I'm talking 6 months or more, not just 1 or 2 nights. Anything can happen once, if you only investigate 1 night then you cannot claim that the phenomenon is due to something unexplainable!
3. For 1/3 of the length of the investigation you should keep people strictly out of the investigation area; only measuring and recording equipment should be present (this will allow you to determine if the phenomena are human-presence related or truly "unexplainable"). For the 2nd third, people may enter the area but not "interact" with anything they might come across (this allows for the possibility that spirits prefer to act when there is an audience). For the final third, the investigators may interact; this means calling out or responding to a "presence" (just try not to be a dork about it, à la Ghost Adventures and Most Haunted - I mean, seriously...). By doing the investigation this way you decrease the chance of your data being seen as total trash because you didn't bother to cancel out confounding factors in your experiment.
4. For each third of the experiment the time should be split between day-time and night-time investigation. In other words, don't just research night-time activity, examine day-time activity, too. You might be surprised.
5. When analyzing data it should be independently reviewed by a third party. This removes the inherent experimenter bias. More than 1 person should review and each reviewer should be segregated during the review, writing down their findings. Only after all reviewers have performed separate analyses should  the results be collated. This ensures that no observer bias creeps into the study clouding the conclusion.
6. And finally: As with all good experiments it should be repeatable by others. Therefore, your process should be well documented so that others may repeat the procedure, as exactly as possible, if needed.

If you really want to "prove" to the non-believer that ghosts exist you have got to stop being stupid about it. Those who have had experiences (like me), or want to believe, always will; the ones who haven't experienced it and don't believe...are going to take a lot more convincing, so isn't it about time you did it right?