Please note all this is a work in progress: little more than a series of memo’s in which Becky and I are developing ideas we want to explore…
OK last night I posted on a topic that interested me, and seemed to suggest that we forget anomalous experiences quite quickly. Andrew has raised the possibility that the more recent experiences may simply be made up: I admit that is possible, but wonder why people would claim the fictional experience was situated in the last twelve months, rather than long ago, or just today?
This morning I am going to look at the data again, but this time look at age at time of the experience rather than time elapsed between the experience and the report. The first chart shows the ages of our respondents – by category, as we do not possess precise ages for most respondents, and many said things like “when I was living with my parents” or “twelve years ago”.
Table 1. Age of Persons Who Responded
As you can see, they cluster around the 30′s - unsurprising given the method of collection, as most of the people who viewed the question were in that age category! However the age of percipients does set an upper limit for both how long ago an experience can have occurred, obviously enough (people can not have anomalous experiences before they were born), so I have reproduced the data here. It is a shame we did not ask for more precise ages!
Perhaps more interesting is the following chart, which shows how old people were when the claimed anomalous experience took place.
Table 2. Number of INCIDENTS reported by age category
The sharp decline after 30′s is simply owing to the age of our respondents as shown in the first chart, os it is not safe to assume any falling off in experience as we grow older, and the data set is really too small to allow for any meaningful statistical analysis (which is why the study used Grounded Theory methodology). The 1894 Census saw a peak around the age of 21: however there were methodological problems in that Census which may account for this. The Census of Hallucinations (1894) discounted experiences below the age of ten years – we have reported on them and included them in our dataset.
Bear in mind these are experiences, not people. As we shall see in future reports, some people reported many incidents of allegedly anomalous experiences. Also note that continuing and ongoing experiences were not included in these figures. 42 incidents could be placed at a certain age in the percipient’s life from the accounts submitted.
However, some experiences (mainly those in childhood) were placed at a precise age: so here are the break downs for the under 20′s…
Table 3. Incidents in the Under-20′s
The number of experiences at age 18 to 19 appears comparatively high: not quite sure why that should be. Maybe moving away from home?
Anyway these figures are not as interesting as the last set to my mind, but if anyone has comment or thoughts on all this we would love to hear them!
OK something mildly interesting tonight – very much a preliminary set of thoughts on something Becky and I are still working on. Before she started her PhD we undertook some simple research on reports of “paranormal experiences” together, using a novel new methodology which is actually quite close to that used by the SPR in the 1894 Census of Hallucinations. And something has already come up that I find fascinating!
I don’t know yet if Becky is going to develop said methodology for her PhD, so I won’t talk about it here, but the important thing is with the help of a number of friends, including but not limited to Yvette Nicole-Hall, Axel Johnston, Rupert Scott-Ward, Miranda Cardew, Lynn Cinderey, Emma and Paul Tudor, Thomas Nowell and others (the full list will appear in the final paper if we ever finish writing this up: please drop me a line if you took part and I have missed you off the list!) we collected sixty “paranormal” type experiences. We then coded them, using a Grounded Theory approach, and I’m still looking at the data.
The question that was posed in our “accidental census” (Becky developed the methodology quite by chance) was
“Have you ever – when believing yourself to be fully awake and unaffected by illness, drink or drugs — had a vivid experience of seeing something or someone, or of hearing a voice, when there is nothing there and no ordinary cause you can find?”
or one of three other minor variations of the same, as we were playing around with the wording, experimenting with the original SPR form, DJ West’s, the MAss Observation Survey question and finally Becky’s own version (above). The only difference noted was in number of responses,
So what did we find? Well one thing I mentioned the other day leapt out at me immediately. Remember I said in the piece on Thetford Priory and my own ghost experience that I thought experiences diminished rather the grew in the telling, and that many events that seemed quite “paranormal” at the time are quickly forgotten? Well sixty cases is not a lot to base anything on, but here is a quick chart I just knocked up in Open Office Calc.
Bit blurry as I’m not good with the Export function. Anyway at first glance, it seems to show pretty much what we might expect in terms of a fairly even distribution of our experiences. Ah I hear you cry – there are only 49 percipients (people who experienced the event) listed here. Yep, in some cases it was impossible to work out exactly WHEN the event occurred from the narratives we received, and we also omitted repeated phenomena (as in “this happens to me every day” and continuing phenomena, as in “and it’s still happening…”) from this chart. If they were included the effect would be stronger I think…
So what is puzzling me? The garishly (and with no regard to red/green colour blindness: I should have checked how to change the colours) bars do not represent equal periods of time. Two people reported their event the day it happened, the first bar. The second bar is those who told us of something in the last week, the third the last month. The first 5 bars represent events that were experienced in the last year… But as we get further down the chart, well the fifth bar is four years (from 12 months to 5 years), then we go up in 5 year blocks, then ten years.
Here is the important bit: in the last twelve months, twelve people claim to have had an experience of the type we are interested in – including witnessing an apparition, seeing an object move in what appears to be a “paranormal” fashion, hearing voices, being touched by invisible presences, etc, while well and not under the influence of drugs or drink. So lets assume that people are more likely to recall and report accurately events in the last twelve months. Using that 12 months as a baseline; if that rate was the average (assuming that Dave Williams was wrong earlier tonight when he joked “it looks like Zuul is coming to the West Midlands!) then we would expect sixty reports in each of the 5 year blocks; the mean is actually 4.6 reports per five years. So where are the missing 275+ reports?
Well the age of the population reporting is obviously crucial. We only asked by decade of age, and though I have some precise ages, for most people I only know if they were say in their thirties or forties. I created a chart to show the distribution here. With the limited data I have I would estimate the average age is somewhere around 33. It is hardly surprising then that many people do not report events forty years ago — they were not born.
This leave three main hypotheses to consider to account for the issue
1. The one the SPR Report On the Census of Hallucinations put forward in 1894, and I mentioned to Wiseman & Watt at the Science of Ghosts event in Edinburgh last April, which seemed to surprise them – – people rapidly forget anomalous experiences in the main.
2. People are reporting the most spectacular events they can think of they have experienced, and ignoring minor recent experiences that would meet the Census question.
3. People are making up their experiences, and claiming these false experiences happened recently. I rather doubt this one for various reasons, not least the mode of collection for the data, and the fact the same pattern is found in Sidgwick (1894) – and I suspect in Donald West’s three studies.
I will return to this issue tomorrow, and talk more about the preliminary findings. For the moment I welcome any comment, in particular suggestion of appropriate statistical measures to employ on the quantitative data as Becky has returned to Derby to work, and I’m fairly rubbish at this sort of thing. :)