Saturday, July 25, 2009

The Science of Ghost Hunting and Paranormal Investigation Part I: The Basics of the Scientific Approach

The word “science” or “scientific” has been thrown around in the field of ghost hunting and paranormal investigation for many years. What does science offer a field that is almost completely philosophic in nature, from the lack of insurmountable evidence to the loose theories that fuel our pursuits?

In my last blog I discussed the beginnings of the Spiritualist movement which was the catalyst behind Parapsychology and the scientific pursuit into ghosts and related phenomena. Psychical Research, now Parapsychology, helped separate the fakes and frauds of mediums, table tipping sessions and man-made spirit photography. Through all of this there have always been amateurs who have worked in the background of the professional organizations and individuals, at no other point since the beginning of the Spiritualism era has the amateur ranks played such an important part of ghost and paranormal research. The Society for Psychical Research (SPR, 1882) and the American Society for Psychical Research (ASPR, 1885) still exist as well as other professional arms of paranormal research (Parapsychological Association, Rhine Research Center, Institute of Noetic Science, etc.) though the amateur ranks outnumber the professional ranks as well as the overall exposure and interaction with spontaneous cases.

What separates professionals and amateurs, besides a degree from a University and and/or a professional organization is the interpretation and use of science. The raw form of scientific approach in the historical context of parapsychological study was mere case reports (data collection). This would merely be a collection of firsthand accounts of paranormal activity as well as possible witnesses. This was not enough to make this into a science until they were able to systematically study or replicate experiments with these reports.

This now draws us to the 21st Century and the amateur pursuit of this field using scientific tools. The use of tools in this field (EMF detector, compass, dowsing rods, ion detector, infra-red thermal imager, camera, camcorder, cassette and digital audio recorder, motion sensors, etc.) has seemingly clouded the minds of those who are attempting to verify anomalous phenomena through them. Many people think that the mere use of these tools is science and having anomalous readings with them serves as evidence of the paranormal.

There is no such thing as a ghost detector. No tool in use in the field of ghost research has the ability to determine a ghost is present. The mere use of these tools alone does not constitute a scientific endeavor. These preceding statements are not my opinions, these are fact. These statements have been made by parapsychologists and others in various scientific disciplines.

It seems that the objectivity of the use of these tools has been overlooked since many groups have stated they have collected data that they have detected the presence of ghosts. But the question is, how do they know it has detected a ghost and not another anomalous object or field of energy? If the detection with the instrument was accomplished in addition to an anomalous event happening at the same time then we have a possible legitimate reading on our hands (see Auerbach, (2004, p. 112). Then we have to ask if there were baseline readings done of this area in question prior to and after the anomalous reading (without altering the environment, i.e., turning off lights or power)? Were there witnesses to this event or any other devices that could back up the other? We could wonder whether those who set out to gather this evidence may have unintentionally created the anomalous event. Is this a repeatable event? To many ghost hunters this sounds like an impossible task, but it should to most since this is how real science works. Believe me, I am being generous in my examples here compared to those that would be required for other scientists to have a sliver of belief in what we are trying to accomplish.

The major piece of the puzzle that would come next would be for the group that encountered these readings to go through a formal scientific process and ultimately publishing their findings and have another group attempt to gather the same data (or finding discrepancies with the prior data collection methods) using similar or related instruments. I can already hear some people out there laughing at the premise of a group inviting another one out to confirm or deny their findings, though in fairness I know quite a few that would be happy to do this. The bottom line is that this is true science. Merely wielding tools of science is not scientific, even if you know how to use them, which is my whole point here.

Let me share with you an excerpt from the handbook I designed for the Ohio Paranormal Investigation Network in dealing with the scientific approach as well as steps in a scientific investigation.

The Scientific Approach to Behavior

There are three sets of interrelated goals to turning a hypothesis into scientifically accepted data; measurement and description, understanding and prediction, and application and control.

Measurement and Description- Before a scientist can explain why the world works in a certain way; they need to describe how it works. Science’s commitment to observation usually requires that an investigator figure out a way to measure the phenomenon under study. The goal here is to develop measurement techniques that make it possible to describe behavior clearly and precisely. The attempt is made by using gadgets such as EMF detectors, thermal imaging cameras, etc. The problem lies in the fact that ghost hunters are only concerned with the “what” instead of the “how”. The “how” is how the devices are detecting what they are detecting as well as how they know it is an apparition they are recording. This is also difficult, if not impossible, since most of what they record with these devices is spontaneous in nature as with all Psi phenomena.

Understanding and Prediction- Scientists believe that they understand events when they can explain the reasons for their occurrence. To evaluate their understanding, scientists make and test predictions about relationships between variables. This seems like an easy one, but we have little understanding of what is going on with ghosts. We have been able to predict certain behaviors, but we have yet to learn the causes that lead to these behaviors. Psi phenomena is the same, we do not yet know how (or where in the brain) it exists nor can we predict when it will occur.

Application and Control- Ultimately, Most scientists hope that the information they gather will be of some practical value in helping solve everyday problems. Once people understand a phenomenon, they often can exert more control over it. . Once we begin to understand how ghosts come into existence, or devise a practical means to communicate with them, the World as we know it will become a different place. The same holds true if we are able to utilize ESP or Psychokinesis on a daily basis at any degree.

Steps in a Scientific Investigation

The following example is based upon the “top-down method”. In a further blog I will describe this, as well as other examples of various methods described as “bottom-up” and “Integrated Approach” (Watt, 1994, p. 77), in further detail.

1. Formulate a Testable Hypothesis.

The first step is to translate a general idea into a testable hypothesis. A hypothesis is a tentative statement about the relationship between two or more variables. Hypotheses are generally expressed as predictions. They spell out how changes in one variable will be related to changes in another variable. To be testable, scientific hypotheses must be formulated precisely, and the variables under study must be clearly defined. Researchers achieve these clear formulations by providing operational definitions of the relevant variables. An operational definition describes the actions or operations that will be made to measure or control a variable. Operational definitions establish precisely what is meant by each variable in the context of a study. Most ghost groups come up with theories, which are based upon unproven ideas or speculation of how events may hold an answer. A theory can be thought of as merely a guess or assumption and has no application toward science, but can help lead to a more specific hypothesis.

2. Select the Research Method and Design the Study.

The second step in a scientific investigation is to figure out how to put the hypothesis to an empirical test. The research method chosen depends to a large degree on the nature of the question under study. The various methods- experiments, case studies, surveys, naturalistic observation - each have advantages and disadvantages. The researcher has to ponder the pros and cons and then select the strategy that appears to be the most appropriate and practical. Once researchers have chosen a general method, they must make detailed plans for executing their study.

3. Collect the Data.

According to their plans, researchers obtain their samples of subjects and conduct their study. Psychologists use a variety of data collection techniques, which are procedures for making empirical observations and measurements. Commonly used techniques include direct observation, questionnaires, interviews, psychological tests, physiological recordings, and examination of archival records. These methods are broken down below. Collecting research data often takes an enormous amount of time and work.

Direct observation - Observers are trained to watch and record behavior as objectively and precisely as possible. This is where the amateur movement is focused using their instrumentation as a guide (should be using their tools to back up events not merely be the only record or indication).
Questionnaire - Subjects are administered a series of written questions designed to obtain information about attitudes, opinions, and specific aspects of their behavior.
Interview - A face-to-face dialogue is conducted to obtain information about specific aspects of a subject’s behavior.
Psychological test - Subjects are administered a standardized measure to obtain a sample of their behavior. Tests are usually used to assess mental abilities or personality traits. This phase of study includes the use of Zener cards (5 symbols used in guessing experiments) in Parapsychology (which are no longer used) as well as random number generators (used in psychokinetic experiments).
Physiological recording - An instrument is used to monitor and record a specific physiological process in a subject. Examples include measures of blood pressure, heart rate, muscle tension, and brain activity.
Examination of archival records - The researcher analyzes existing institutional records (the archives), such as census, economic, medical, legal, educational, and business records. This is where Psychical research began by studying stories of those who had spontaneous events occur to them.

4. Analyze the Data and Draw Conclusions.

The observations made in a study are usually converted into numbers, which constitute the raw data of the study. Researchers use statistics to analyze their data and to decide whether their hypotheses have been supported. Thus, statistics play an essential role in the scientific enterprise. This is where many groups get lost. It’s not really one case that will bring you answers, it is the specific information gathered by the sum of many.

5. Report the Findings.

Scientific progress can be achieved only if researchers share their findings with one another and with the general public. Therefore, the final step in a scientific investigation is to write up a concise summary of the study and its findings. Typically, researchers prepare a report that is delivered to a journal for publication.

The process of publishing scientific studies allows other experts to evaluate and critique new research findings. Sometimes this process of critical evaluation discloses flaws in a study. If the flaws are serious enough, the results may be discounted or discarded. This evaluation process is a major strength of the scientific approach because it gradually weeds out erroneous findings. For this reason, the scientific enterprise is sometimes characterized as “self-correcting”.

This is why it is critical for ghost hunting and paranormal investigation groups to share and compare data. We are stuck making the same guesses over and over from one group to the next until the data is put out there for everyone to evaluate. Each group feels that they alone will produce a piece of ground-breaking evidence that will set the World on its ear, this is a belief founded in misunderstanding of how the scientific process operates. Groups must work together building upon work to come up with answers. It is possible that one group may eventually hold the key, but this group will be one that uses the scientific approach from work that eventually will be confirmed or denied by others.

Advantages/Disadvantages of the Scientific Approach

The Scientific approach offers clarity and precision. Common-sense notions tend to be vague and ambiguous. The major advantage of the scientific approach is its relative intolerance of error. While possibly not proving anything beyond argument, the scientific approach does tend to yield more accurate and dependable information than casual analyses and armchair speculation do. Knowledge of scientific data can thus provide a useful benchmark against which to judge claims and information from other kinds of sources.

The major disadvantage of the scientific approach (especially in the case of ghost research) is that experiments are often artificial. Experiments require great control over proceedings and researchers must construct simple contrived situations to these their hypothesis experimentally. It is practically impossible to simulate the environment of a haunted location in a laboratory setting, this is why there is no experimentation within the study of the paranormal- not only can we not duplicate this environment we could never possibly control it.

We must rely solely on descriptive research methods. These methods are used when the variables cannot be manipulated. In other words, these methods cannot be used to describe cause-and-effect relationships between variables. This understanding could come with time and understanding and after the use of the descriptive research. Descriptive methods permit investigators only to describe patterns of behavior and discover links or associations between variables. This is where Parapsychology has been stuck for over 130 years. Once variables are introduced the results usually seem to become inconclusive.

Descriptive research cannot demonstrate conclusively that two variables are causally related.

Descriptive research methods include:

Naturalistic Observation
A researcher engages in careful, usually prolonged, observation of behavior without intervening directly with the subjects. The problems with this approach are twofold. First, it is nearly impossible to observe ghost activity for a prolonged period of time, especially when dealing with a specific case. This lack of observation time usually proves the information gathered inconclusive. Secondly, it is nearly impossible to observe the ghost behavior without becoming directly involved with the subjects. We try hard to gather as much as possible and to go in to a location at the last possible second, but with this you risk not being able to observe any activity yourself. Once you are part of the environment you have changed the variables and altered the environment in which the events occur.

Case studies
These are an in-depth investigation of an individual subject. Data is collected for individual cases and compared to that of other cases to arrive at a common explanation. The major problem with this approach is that they are highly subjective. The information from several sources must be knit together in an impressionistic way. During this process one may focus on information that fits with their expectations, which usually reflect their theoretical slant. Thus, it is relatively easy for investigators to see what they expect to see in case study research.

Researchers use questionnaires or interviews to gather information about specific aspects of subjects’ behavior. Surveys are often used to obtain information on aspects of behavior that are difficult to observe directly. Surveys also make it relatively easy to collect data on attitudes and opinions from large samples of subjects. The major problem with surveys is that they depend of self-report data. Intentional deception and wishful thinking can distort subjects’ verbal reports about their behavior.

Scientific research is a more reliable source of information than casual observation or popular belief. However, it would be wrong to conclude that all published research is free of errors. Scientists are fallible human beings, and flawed studies do make their way into the body of scientific literature. This is why replication of a study (or observation) is important. Replication of a study may lead to contradicting results. Some inconsistency in results is to be expected, given science’s commitment to replication. Fortunately, one of the strengths of the empirical approach is that scientists work to reconcile or explain conflicting results. In fact, scientific advances often emerge out of efforts to explain contradictory findings. This is very important to keep in mind as we collect data and continually observe findings in the field. We must not always be too quick in search of the answer, or the truth may elude us. Looking at conflicting evidence is very healthy and may lead us to the answers.

Flaws in evaluation of research

A sample is the collection of subjects selected for observation in an empirical study. In contrast, the population is the much larger collection of animals or people (from which the sample is drawn) that researchers want to generalize about. Sampling bias exists when a sample is not representative of the population from which it was drawn. We must be careful to not drawn conclusions until we are able to deal with a diverse array of the public, which will give us a fair sample of the overall population which is encountering these occurrences.

Placebo effects occur when subjects’ expectations lead them to experience some change even though they receive empty, fake, or ineffectual treatment. Placebo, for our concern, may happen if we give them information on the subject, they may conform their observations on the new information given to them or alter what has happened in the past or jump to quick conclusions about natural experiences. The overall effect of the thought of a ghost is interacting with them causes some to distort previous experiences. We must be careful with what information we give them and at what time we give it.

Social desirability bias is a tendency to give socially approved answers to questions about oneself. This could also include conforming your observations by what is popularly known to be ghost activity. Example; someone feels a draft in their house and assumes it is a ghost, they may lump other unrelated experiences with it to draw the conclusion or convince others based on the current social trends of paranormal activity.

Experimenter bias occurs when a researcher’s expectations or preferences about the outcome of a study influence the results obtained. This is a common problem with ghost research on many different levels. The first that comes to mind is that we (like some who experience ghosts) jump to conclusions based on bits of information obtained during the case. We must learn to look at each piece of evidence as separate and not lump everything together immediately and assume every clue adds another piece of the puzzle and help confirms a ghost. Another lies with the popular orb photographs. In this we have a tendency to draw our conclusions from what we see or believe not what we can prove or study. We must not try to “see what we want to see” and look for viable evidence to deny any rational or natural explanations first and foremost.

We must take what the history of science has given us and use it to our advantage, it can only help us confirm or deny what we set out to find.


Auerbach, Loyd. (2004). Ghost Hunting: How to Investigate the Paranormal. Oakland: Ronin Publishing.

Irwin, H.J. (1994). An Introduction to Parapsychology, 2nd. Ed. Jefferson, NC. McFarland & Company.

Watt, Caroline A. (1994). Advances in Parapsychological Research, Volume 7. (Making the Most of Spontaneous Cases, p. 77-103). Jefferson, NC. McFarland & Company.

Weiten, Wayne. (1992). Psychology: Themes and Variations, 2nd Ed. Pacific Grove, California. Brooks/Cole Publishing Company.