Biased Ghost Research: Is Seeing Really Believing?

When ghost hunters attempt to research ghosts in a scientific way one of the first mistakes that they will make is insisting that although their methodology might look unscientific to the outside observer, this simply isn’t the case. Ghost hunters often conduct their investigations in the dark, use pieces of equipment that have no real use in ghost research (such as EMF meters, REM pods, Dictaphones and similar), may employ the use of psychics and mediums, and use spirit communication techniques because they claim that they are attempting to experience the occurence reported to them by the eye-witness. This is so that when they experience the activity for themselves they can then try to distinguish – scientifically – what the cause is.

The problem here, of course, is that unless the eye-witness experiences what they report while in the company of a medium, with the lights turned out, while using an EMF meter and holding some dowsing rods, then all the ghost hunter is actually doing is priming themselves to have a weird experience. This will then be erroneously linked to the original experience as evidence that something has occurred that may be considered as potential evidence of the paranormal. It isn’t always possible to replicate the exact conditions in which an experience happens, but ghost hunters often make a lot of effort to conduct their research in a manner that is as unlike the original conditions as possible.

This could be an indication that the ghost hunters in question are not really interested in figuring out the original cause of the reported activity and, instead, just want to experience something weird and thrilling. They offer up a scientific-sounding explanation for their unscientific behaviour because nobody likes to be accused of being stupid and it can be very rewarding to play scientist. When you reach a satisfying conclusion, if you have managed to convince yourself and others that you did so scientifically then it is even more satisfying.

However, by turning the lights off, “calling out” to ghosts, using phone apps or spirit communication methods, working with psychics or chasing EMF fluctuations for every single case, ghost hunters are simply subjecting themselves to a series of events brought about by expectation bias. In fact, there is a whole range of biases that ghost hunters will experience during the course of their investigations and the only way to avoid these is by learning what they are and how they can affect the work being conducted on a case.

Confirmation bias, for example, can cause an investigator to interpret results incorrectly as they’re unknowingly looking for information that fits with their belief that a location is haunted. Expectation bias can cause ghost investigators to allow their expectations of a ghost hunt to influence the outcome. Attentional bias can cause ghost investigators thinking about a haunted location or ghosts to pay more attention and put more emphasis on mundane occurrences (in the context of it being potential evidence) than they wouldn’t think twice about if they weren’t thinking about ghosts at the time.

Conclusions and field reports written by ghost hunters can often by subjective accounts of what took place in which the investigator has not taken steps to rule out personal biases, and for ghost hunters across the globe, there is often a problem with cherry-picked data. For example, ghost hunters who use a flashlight to allegedly communicate with ghosts (typically asking a ghost to turn the flashlight on and off in answer to questions), will note down the occasions when the flashlight turns on in response to a question, but they will not note those instances where the flashlight did not turn on in response to a question, or perhaps when it did turn on but not in response to a question. Because all data is important when evaluating research, the inclusion of only positive data means that the reports being written are not accurate, honest or useful.

A large part of paranormal research is about minimalising the influence that we, as researchers, can have upon our findings. We have to ensure that our observations are as objective as possible and not working alone on a case can help us to rule out misinterpretations of what happened. Even post-investigation peer review can help you to work out those occasions on when you may have made biased decisions through a colleague pointing it out to you. Acknowledging the limits to your ability to be an objective and rational investigator isn’t a bad thing – it means that you’re already half way there.

Published by

Hayley Stevens

Hayley is a ghost geek and started to blog in 2007. She uses scientific scepticism to investigate weird stuff and writes about it here while also speaking publicly about how to hunt ghosts as a skeptic.

4 thoughts on “Biased Ghost Research: Is Seeing Really Believing?”

  1. Paranormal phenomena of all types seem to resist performing on cue. The most impressive phenomena seem to occur when they are least expected and least sought-out. It’s the ubiquitous Trickster – everything is always just out of reach, just over the horizon, just on the liminal edge. Being impossible to pin down seems to be an actual characteristic of the phenomena. Efforts to actually produce the phenomena or pin them down “scientifically” are likely to do neither, but rather to open doors that the “researcher” will one day wish he hadn’t opened. The tactics you are describing strike me as far more dangerous than “simply subjecting themselves to a series of events brought about by expectation bias.” I have studied reports of these phenomena for decades and have had occasional spontaneous experiences of them (not ghosts, alas), and I would run as far and as fast as possible from this sort of ghost-hunting.

    1. Although I respect that people are entitled to opinions i genuinely don’t think there is anything to fear and that the most that the people using these techniques will experience is whatever their imagination summons up.

Leave a Reply

Your email address will not be published. Required fields are marked *