Best practices

Paranormal or paranatural events are notoriously difficult to verify and analyze. Often phenomena appears so suddenly that the only "evidence" collected is witness testimony (which is unreliable at best). Although intriguing to hear, at Fortean Labs we don't analyze events based on testimony. Event substantiation and analysis suitability requires data that can be analyzed.

Video/Photographic Data

Video or photographic data requires at least two cameras unless context metrics are known (velocity, distance, etc.). The addition of visual confirmation is even better. Was it seen visually or just on video? For example, suppose you capture an "orb" in a video. Two cameras can rule out dust near the lens or lens flare. It's even more credible if the "orb" is confirmed visually as well as captured in video. It's all about verifiability and data is the key.

Reflections can be a problem if shooting through a window. Another camera located outside the window will verify data collected by the one behind the glass.

Security cameras running in night mode (infrared) can produce video artifacts as well as slow frame rates. Flying insects appear as "rods" instead of just flying bugs. Using at least two cameras can eliminate many misidentifications. Objects can appear to be far away when, in fact, they may be just close and out of focus. To harp on it again -- USE TWO CAMERAS!

Audio Data

Audio data collection should be treated similarly to video/photo data. Best practice requires collection by redundant devices to rule out contamination by misinterpreted sources. An EVP recorded should be verified by another spatially-separated device either on other persons present and/or another location at the site.

EVP (Electronic Voice Phenomenon) is highly, highly susceptible to pareidolia. Devices such as spirit boxes, word generators and other such devices are worse than worthless in our opinion for this reason. Listen to random words or syllables for long enough and you're bound to find something "meaningful." Our recommendation is stick to Raudive's methods (Raudive, K 1971) and record pure audio recordings. Get a firm grip on your goal, entertainment or science. Remember Occam's Razor -- the simplest explanation is usually the correct one.

Redundancy

You're probably detecting a trend by now -- device and data redundancy. I don't know how many "paranormal" events I've seen that could have been verified or debunked by the addition of a second recording device and visual confirmation.

Reject the NULL Hypothesis

Waving around an EMF meter or temperature probe is NOT science. Even tangibles such as video, photographs or audio must be vetted to be accepted as “proof” of a true anamoly. Science works through validation of a hypothesis by rejecting the "null" hypothesis. Gathering "evidence" with debunking in mind is essential -- always assume that the event is NOT paranormal. Take the approach of rejecting the null hypothesis, i.e.the event is NOT paranormal. Demonstrate that no normal interpretation explains the event. Only then can the null hypothesis (the event is not paranormal) be rejected. That's how science works.