Belief Systems and Phobia Prevention

Is there a correlation between the incidence of pronounced coping strategies and adherence to a religious belief system?

I. Abstract

This paper describes an experiment to investigate the correlation between two groups as to the relative importance they attach to ritual as a means of preventing or avoiding fear. The first group was one that has sincere, Christian religious beliefs; the second group was composed of people who claimed to be atheist. The underlying assumption is that religious ritual might be an expression of obsessive-compulsive disorder (OCD). Thus the aim of the experiment is to examine the extent to which the practice of Christian religious rituals and atheist rituals compare in terms of this specific psychological function - the avoidance of fear.

The first step must be to offer a definition of OCD. I take that offered in DSM IV, which specifies several aspects that are critically important in its diagnosis, but nevertheless remain entirely subjective. These are:

1. Actions that are repetitive and formalised.
2. The subject feels driven to perform them.
3. The acts are performed to reduce distress, and are not an end in themselves.
4. The subject may recognise that the behaviour is unreasonable or unrealistic.
5. The subject finds the behaviour to be disturbing and attempts to resist or avoid situations where ritualising will become necessary.

This set of proposals would seem to necessarily include people of a strong religious disposition. OCD behaviour typically involves: washing, checking, ordering, concern with symmetry, counting, hoarding, repeating words silently or aloud. In religious ceremonies and practices, public and private, many of these features may be found.

II. Aims & Hypotheses

The aim of this investigation is discover whether a strong religious belief in an individual is indicative of pronounced tendency towards use of coping strategies.

First, it should be noted that the Literature has a marked tendency to refer to the strong term "obsessive compulsive disorder" (OCD). I prefer to use a weaker term that I have coined for this purpose: "pronounced coping strategy" (PCS). I suggest that the two are closely linked since both are used by the sufferer to prevent phobia. However, "OCD" has a precise clinical definition (see the reference to DMS IV in Section I, above), whereas PCS is a more generally applicable term that need not refer to a chronic condition. I therefore use the term "PCS" throughout as I do not wish to infer too strong a claim at this stage. This is particularly so because none of the subjects that participated in the experimental procedure I will describe were actually sufferers of OCD and it would be unethical to draw strong conclusions. However, I will continue to use "OCD" where I am referring to a work in the literature that explicitly uses the term.

Two hypotheses are offered. The first is a one-tailed alternative hypothesis that the Christians should score higher (i.e. show a stronger negative tendency to PCS) than atheists because they have a ritual based religion, which could possibly be viewed as a form of PCS or phobic presence. However, if the atheists score higher than the Christians this may be because they introduce some other form of ritual into their lives in order to compensate but without the group structure to support them and potentially mitigate the ritual. The interesting conclusion with regard to the latter would be that Christians are less prone to PCS because they draw some beneficial effect by participating in regular group ritual in an organised and controlled setting. Atheists on the other hand have no such group ritual to fall back and so draw on individual ritual without the controls that are implicit or explicit in the Christian ritual.

The second hypothesis in a null hypothesis. There will be no difference between Christians and atheists and that hence there is no link between religious ritual and PCS. If both groups score low then neither group has any discernible tendency towards PCS and Bowyer & Lienard's claim (2005, p.7) is supported. If both groups score high then both would be more prone to PCS or phobic presence. This would also support Bowyer & Lienard's claim (ibid.) Either of these conclusions would effectively discount religiosity as a factor in the development or externalisation of PCS.

The overall hypothesis is that the one-tailed alternative is more likely. Specifically that Christians will score higher due to the pervasive presence of Christian ritual in their personal lives. If this were to be the case then Smay's claim (2001, p.2) would be supported.

III. Review of Previous Research

It is clear from a literature search (principally conducted via the Internet, though with strict criteria concerning academic provenance) that there is very little in the academic corpus that approaches this particular topic. This impression is supported by at least one other researcher:

"There are substantial data exploring the link between religiousness and health, yet there is little consensus regarding the nature of the relationship between religiousness domains and mental health."
- (Salsman & Carlson, 2005, p.201)

The strongest empirical evidence is apparently provided by (Sica, 2002), but although this work is referenced it proved impossible to find other than in summary form, although it can be purchased over the Internet via the Journal of Behaviour Research & Therapy. However, some other researchers do make inferences that there is a relationship between the manifestation of OCD (Obsessive Compulsive Disorder) and religion.

"Obsessive-compulsive disorder (OCD) is a multi-faceted psychiatric disease, which is characterized by persistent and distress-causing cognitions (obsessions) that trigger repetitive behaviors or mental acts such as washing and counting in the majority of patients (compulsions). These two psychopathological clusters are usually functionally connected and typically deal with themes of aggression, contamination, symmetry, order and religion." - (Moritz & Mühlenen, 2005, p.118)

Consider also the following. It is a modest claim, but it addresses the theme of this paper, and it is the philosophical corrollary of what I shall seek to demonstrate (or otherwise) by experiment, which the latter part of this paper will describe:

"The occasions for ritualized behaviors also vary, either contingencies such as illness or misfortune, life-stages like birth, initiation and death, or recurrent occasions such as seasonal changes. Finally, the connections between rituals and religious concepts are crucial in some cases (e.g. ancestor worship, Islamic prayer), or only peripheral (e.g. anti-witchcraft divination), or just absent (in “secular” rituals). It might seem imprudent to make any general statements about such a disparate set. Note, however, that our aim here is not to account for all these social occasions, for the reasons why they exist and why people find them of interest. Our aim is to account for the psychological salience of a particular feature they share, namely the performance of what we call ritualized actions." - (Boyer & Lienard, 2005 p.7)

However, some authors in addition what has been reported of Sica (ibid.) have made strong claims. For example, Portelli suggests that:

"The ritualistic attempted solutions can be 'repairing' or ‘preventive‘. This means that we were able to pin point two different types of obsessive compulsive rituals; the first is carried out to intervene and repair after a feared event has taken place, so as not to feel in danger, thus it is oriented towards the past; the second is focused on anticipating the frightening situation or event to propitiate the best outcome or to avoid the worst. However, recent empirical-experimental results revealed that there exists 2 main variants of preventive rituals: Rational-preventive and propitiatory magical thinking rituals. Rational-preventive rituals are specific actions put into action that arise from an irrational belief that doing so, the subject would prevent certain situations he/she fears, such as contamination, lose of control, lose of energy and so forth. The propitiatory rituals are a form of magical thinking highly linked to fatalistic religious beliefs, superstitious convictions, confidence in extraordinary powers or in faith and so forth." - (Portelli, 2004, p.43)

Regrettably, Portelli offers no experimental data or other empirical evidence with which to support his claim although he does repeatedly refer to empirical evidence being available. Others too have remarked on this fact (Arezzo, 2004, p.92.). One is driven to assume that Portelli's work is a preliminary paper and not a full account of his research.

Smay on the other hand presents a thoughtful philosophical piece which makes reference to extensive data that have been gathered regarding OCD as a pathology (Smay, 2001, p.4.) and leads on to discuss physiological structures and their role in OCD in particular and ritualised behaviour in general. Although it is not relevant here she refers to Greist's reporting of methods of psycho surgery, which involve severing connections between the Basal Ganglia and Frontal Cortex (Greist, 1992, 1998)

Smay's clear belief is that the origins of OCD and ritualised behaviour in general are related and she makes explicit link between OCD and religion:

"In an influential paper, Fiske and Halsam (1997) begin with a description of a man in an unfamiliar country. We observe him to be dressed all in red in a red doorway, washing his hands six times in six different basins that have been arranged meticulously. His eyebrows are plucked bare, and as he washes, he repeats the same phrase, occasionally tapping his earlobe with his right index finger. Their question to the hypothetical observer is: Is this man a priest performing a sanctified ritual? Or is he afflicted with obsessive compulsive disorder? Is he normal, or mad?" - (Smay, 2001, p.2, et passim)

However, there is a claim that participation in a group ritual, such as a religious service is significantly different:

"While individual ritualists (especially patients) may feel great anxiety at the prospect of not going through the ritual sequence, participants in a cultural ritual are likely to participate (among other reasons) to the extent that the particular sequence meets a minimal threshold of relevance." - (Portelli, 2005, p.43)

It should be noted that this casts doubt on whether religious ritual has any significant connection with OCD:

"The idea of 'scaling' would also predict all sorts of interesting phenomena that are simply not observed, for instance that people who become more religious would tend to become more obsessive, or that OCD patients would tend to be more religious than controls, that children during early childhood should be more interested in religious ritual than at other stages of development, etc. Although there are connections between certain forms of religious practice and obsessionality … they fail to support these general conjectures." - (ibid., 2005, p.43)

But this fails to recognise that there is a difference between the public and private execution of ritual in that the underlying psychological needs may not be the same and the intensity of the ritual may not be the same. One sort of ritual might not "suit" a given individual, whereas it might another. In fact, the OCD subject commits themselves to a form of therapy only when the escalation between the phobic perception and prevention of the compulsive rituals has driven them to lead an impossible life. Before this stage, they live believing that the ritual is a good strategy with which to allay their fears (Portelli, 2004, p.90).

Little substantive empirical research appears to have been conducted in this area, although several researchers, as noted above, allude to it, However, recent research by the University of Parma, in Italy, has indicated, though not conclusively, that there is a strong link between religiosity and vulnerability to OCD (Sica, 2002).

Therefore I would further justify the pursuit of this research based on the argument presented by Douglas (1982). He points out that many rituals seem to focus around such themes as: pollution and purification, danger and protection, the possible danger of intrusion from other people, the use of particular colors or specific numbers, the construction of an ordered environment (Dulaney & Fiske, 1994). A ritual space or instruments are described as “pure” or “safe” (or on the contrary as the locus of concentrated “pollution”) or the point of the ritual is to “purify” people or objects, to “cleanse” mind or body, etc.. In cultural rituals, this concern with pollution and cleansing is so prevalent that it has been considered a foundation of religious ritual. This claim alone emphasises the relevance of my proposal to survey religious and atheist groups in order to shed more light on the issue.

In concluding this review of the literature one must draw the conclusion that there has been little research done in the area that I propose to investigate. Certainly, in spite of allusions to the contrary, there is no substantial of evidence of any empirical research - although absence of evidence is not evidence of absence.

Therefore there is little for me to call on by way of substantial evidence from previous research, although the claims by Bowyer and Smay will prove to be of relevance when describing the hypotheses and assessing the results of the experiment.


IV Experimental Method & Design

1 Method
This research employed conventional survey techniques to gather raw data which were then be analysed using established statistical methods, including measures of standard deviation and correlation analysis. Thus:

· There were be two subject groups.
· The one group was be Christian, the other was atheist.
· A survey questionnaire was designed with its focus on the concept of ritual.
· Care was be taken in wording the questionnaire for ethical reasons and to avoid possible linguistic bias.
· A small pilot test was be conducted using a small sample to test the overall rigor of the method and design.
· The results of the pilot test informed necessary changes to the method and design, including modifications to the survey questions and the structure of the questionnaire.
· The primary survey was conducted in carefully controlled environments.
· The collected data was collated and analysed.
· Relevant conclusions were drawn and substantiated.

2 Design
The design of this experiment uses independent measures because two different subject groups are used, each with different beliefs and being tested under different conditions.

The questionnaire to be used in the experiment's primary survey is given in Appendix A. The scoring values for each question are shown in situ. The corresponding questionnaire for the pilot survey and its scoring values are given in Appendix B.

The questionnaire was divided into seven categories of questions, respectively covering the subject's:

· Past Experience Q1-4
· Present Experience Q5-8
· Habits Q9-11
· Daily Routines Q12-15
· Daily Rituals Q16-21 & Q27
· Anxieties Q22-26
· Future Q28-29

The categorisation system was deliberately not made explicit on the questionnaire. Note that Q8 and Q26 are the same question, but worded differently, as a partial test for the subject giving untruthful or inconsistent information. Q23 was a qualitative question designed to ease the subject into thinking about their fears, hence Q23 was an unscored question.

In the primary survey the maximum possible score for a given subject was 50, the minimum 0. An arbitrary scoring system breaks down as follows:

· 0-12 there is very little likelihood of the subject having PCS
· 13-25 there is a negative tendency towards PCS
· 26-38 there is a positive tendency towards PCS
· 39-50 there is a strong likelihood of PCS.

For the pilot survey however, a random set of questions was drawn from the main survey question set. The maximum score for any given subject in the pilot test was 10, the minimum again being 0. The arbitrary scoring system for the pilot test, thus breaks down as follows:

· 0-2 there is very little likelihood of the subject having PCS
· 3-5 there is a negative tendency towards PCS
· 6-8 there is a positive tendency towards PCS
· 9-10 there is a strong likelihood of PCS.

In both the pilot and primary surveys each subject was assigned a unique identifying number. Those subjects who are atheist were assigned a three-digit number beginning with "0" (001, 002, etc.). The Christian subjects were assigned a three-digit number beginning with "1" (101, 102, etc.). The individual subject numbers were assigned in the order in which subjects from each group presented to be interviewed. At no point was the subject number disclosed to the individual subject. Once the data had been collected and collated, there was no mechanism for re-associating subjects, with their subject numbers. The age and gender of each subject was also recorded on the questionnaire. There was also provision to make ad hoc notes.

The process for interviewing Christians and atheists was the same, although the locations were different. The Christians were interviewed on the church premises and the atheists were interviewed at Weymouth College.

Prior to the actual interview, each subject was given a standard consent form to sign (Appendix C). They retained one copy for their own records, and one was retained by the interviewer. On the subject's copy were contact details whereby they could discover more about the experiment, or confirm the interviewer's bona fides.

On completion of the interview, each subject was told the purpose of the research. It was felt unlikely that any subject would be upset by the interview. None the less, it was assumed that if any Christian subject were to be upset then they would be able derive support from within their congregation. The atheists from were given cards advising them of counselling services within Weymouth College.

The materials used in the experiment were entirely conventional: pens, paper, desk and computer. Microsoft Word was used to write all research related correspondence and other documents. Microsoft Excel was used to create the necessary tables and charts and perform the statistical analyses. Adobe Illustrator was used to layout the questionnaires.

The independent variables that could affect the experiment are "Christians" belonging to a specific congregation, "atheists" belonging to a specific college community, "gender" and "age".

The dependent variable is the measure of the "tendency towards" PCS, determined by a point score for each subject for overall comparison between Christians and atheists. A high score indicating a strong negative tendency towards PCS, a low score indicating the converse.

The confounding variables are the subject's "honesty", "amiability", "personal background", "psychological mood", "environmental factors" (which include the facilities designated as the test environments), the "willingness" of the relevant authorities to allow the experiment to take place on their premises and the "availability" of the required subjects.

Another confounding variable, over which I had no control, was how religious a given subject was..

The primary controls in place was the use of standardised instruction for each subject, the use of consent forms, a debriefing procedure, the sequence in which questions were asked and types of questions that were asked.

There was no control group, though time permitting an entirely randomly selected group could have been interviewed and their results compared with the principal subject groups.

Ideally, the subject groups would have been representative of the population overall, but given the very small sample sizes (16 in each group) this was not likely to have been the case.

3 Evaluation of Design
It was hoped that most of the adverse environmental factors could be negated through the use of controlled environments. However, such control was harder to achieve for the atheists than for the Christians because the former, drawn from a student community, was much harder to gather together at a single date and time. It is possible that an adverse environmental effect was therefore experienced with the student group, as it was not possible to use the same facilities for all.

The use of the pilot survey was intended to assist in reducing practice effects on the part of the researcher, as well as helping to assess the overall validity of the method and design.

The risk of fatigue effects was minimal since the questionnaire was be short enough to be completed twenty minutes. Indeed, most interviews lasted a little over ten minutes.

Problems were expected to arise if particular subjects were sensitive to ethical or bias factors. For example, an atheist interviewer might inadvertently signal their scepticism to a Christian subject. It was intended that objectivity and professional detachment would reduce such risk to a minimum. The converse is of course true and to be equally guarded against. It was decided to cloak the purpose of the research in such a way as to deny any possible inference that might be drawn by a subject that this research somehow seeks to detract from the integrity of the Christian faith. In fact, relatively little cloaking was required.

There was always a risk that a particular subject might guess that the focus of the experiment was research into phobias, thereby giving rise to demand characteristics. However, every attempt was been made to avoid this situation by not referring to terms such as "phobia", "PCS", or "religion" throughout the survey. Instead the less weighted terms "routine", "ritual", "anxiety" and "coping" were used.

4. Pilot Survey
A pilot survey was conducted in accordance with the broad method and designed detailed above. The completed questionnaires being given in Appendix D. The subject group was comprised of three Christians and three atheists that were know to the researchers and so were easy to access at short notice. They were each interviewed in accordance with the method set out above. The results obtained are given in Table 1.


Table 1. Pilot Test Individual Scores

Subject Score

Number Age Gender 1 2 3 4 5 6 Total

001 17 F 1 1 0 0 0 0 2
002 56 M 0 1 1 2 1 1 6
003 39 F 1 1 0 3 3 1 9
101 51 M 0 0 1 1 0 1 3
102 25 F 0 0 1 3 1 1 6
103 39 F 1 0 0 2 3 0 6


Table 2 shows the overall scores for the pilot survey, with the mean, median and mode scores for each group. A simple bar chart illustrating these data is given in Figure 1.

Table 2. Pilot Survey Scores

Group Score Mean Median Mode

Atheists 17 5.7 6 6
Christians 15 5.0 6 6


The results are on a small scale to the actual experiment, and use only a small section of the questions. Therefore no substantive conclusions can or should be drawn. The principal aims of the pilot survey was only to test the overall design and methodology and to obtain a snapshot of what the final results might look like. Therefore it is only for a sense of completeness at this stage that it can be seen that the atheists scored higher than the Christians, which according to the hypotheses should indicate a higher tendency towards PCS. However, this is a very crude analysis and in terms of mean, median and mode, the atheists and Christians appear very similar.

5. Experimental Procedure

5.1. Recruiting of Subjects
A number of churches in the Weymouth area were contacted with a view to obtaining Christian subjects. St Aldhelms was the first to respond. The vicar was very enthusiastic and an appointment was made to visit him to give more information, discuss the provision of facilities and recruiting volunteers and devising a suitable test schedule. The vicar was given an introductory letter (Appendix E), and shown the full questionnaire, but was given no indication as to the scoring mechanism and was not allowed to retain a copy. He was asked to sign a consent form (also Appendix E) and shown the consent form that subjects from his Church would be required to sign. He was most enthusiastic and readily agreed to help. He recommended that we contact his curate to ask that we might attend his service on 10am service on Sunday 12 March. He also sent an email to his curate informing him of the proposal. I also sent the curate a letter (also Appendix E) and arranged subsequent details through him.

In order to attract atheist subjects from within Weymouth College, recruitment posters (Appendix F) were put up detailing the date, time and location for the interviews. A change was made to the original intention that atheist subjects should first contact me to arrange interviews. Instead they were simply asked to present themselves at the required time and place.

5.2. Christian Group
On the morning of 12 March, the curate of St Aldhelms was shown and agreed the consent form and the questionnaire. He made available two private rooms (the Vestry and the Prayer Room) in which the interviews were to be conducted. Both were already equipped with suitable tables and chairs. The curate proposed telling the congregation of the need for volunteers towards the end of the service. He had already prepared a slide to display for the purpose of making the announcement. He checked the details of the consent form and questionnaire. As an aside, he asked if I would send him an email offering a personal (as opposed to psychological) view of the service I was about to witness.

Towards the end of the service the curate made the promised announcement, together with a slide bearing the question, "Do Atheists have a means of compensating for a lack of religious belief?" Later he admitted that this was probably not the best way to present the matter. In a future experiment I would request to see any similar notice before it was displayed to a potential group of subjects. With some encouragement from the curate, the required sixteen subjects volunteered, but it was a close run thing. But this is probably not so much because of a lack of willingness to take part as confusion. The Reverend told us later that several people had told him that they thought his slide was offering a subject on which they should meditate in the week.

When the service finished, I moved to the front of the church, and the subjects were arbitrarily divided into two groups of eight. I was assisted in conducting the interviews by a friend who had agreed to help and we both interviewed one group of eight each.

As each subject entered the interview room to which they had been allocated, they were invited to take a seat and were told that the experiment was designed to compare the coping strategies used by Christians with those used by atheists, and that the questions would mainly require "Yes/No" answers. They were then shown the consent form and invited to read it and then sign and date, it if they wished to proceed. It was made clear to them that their identity was wholly confidential and that they were free to leave, should they wish, at any point during the interview. Having signed the consent form, the subject were given a copy of it and were told how they could contact myself or Weymouth College for further details, should they require them.

The actual interview was then performed, the researcher asking the questions in turn from 1 - 29 and entering the subject's responses in the appropriate boxes on the questionnaire. Most subjects were quickly responsive with their answers. However, a few asked for clarification, some gave quite elaborate answers. Once all the questions had been asked, the subjects were given an explanation that the experiment was not only based on coping strategies, but about the incidence of pronounced coping strategies and adherence to a religious belief system. All subjects took this information with considerable intrigue and excitement. None appeared perturbed, indeed most seemed very content.

The subject was then thanked and discharged, and the next subject was brought forward and the process repeated until all subjects had been interviewed and discharged. Each interview took around 8-10 minutes and the entire interview process took a little over an hour. There followed a final conversation with the curate. It was agreed that he should be informed as to the results and he was offered a copy of the final paper.

5.3. Atheist Group
The interview area was prepared in good time and the allocated rooms were well sign posted to allow volunteers to find the general area easily and to also find each individual room. Refreshments were provided in the designated waiting room. In order to ensure the required numbers were achieved we toured the campus asking individuals if they were atheist and if they were inviting them to take part in the research and giving them directions as to where they should go and when.

Subjects began to arrive promptly and were called in individually. The overall interview process was as per the Christian group described earlier.

In all ten atheists were interviewed. This was short of the required number and so six additional atheists were found, three within the college and three from outside. These were interviewed on an ad hoc basis, but always in a quiet, relaxed environment, which mirrored as closely as possible the type of setting used for the Christian interviews and for the main group of atheists.

In general the atheists were compliant, readily understood the questions and answered quickly. Very few observational notes were made, or needed to be made.

V. Results

Information concerning the various forms of statistical analyses was taken from Cardwell, et al. (2004).

1. Descriptive Statistics
The overall data sorted in the order of ascending scores is given in Table 3. The full data set is given in Appendix G. This is simplified to show only the subject information, their individual scores as a raw value, and the normal distribution value of their scores. The data is sorted in ascending order of each subject's individual score.

To assist in visual interpretation of the data, the atheists' scores are shown in red, whilst the Christian scores are given in blue. It is apparent at a glance that the atheist subjects have a marked tendency to a low score, whilst the Christian subjects have a marked tendency to a high score. The standard distribution curve derived from the normal distribution column data is given in Appendix H.

The greatest difference between the two groups was shown in the scores obtained for Question 17. It was decided to perform a correlation analysis on this question alone. Accordingly, Table 4 shows the scores gained on Question 17 in the primary survey for each subject in each group sorted in ascending order.


Table 3: Subject Scores in Ascending Order

Subject Score

Number Age Gender Total N.Dist

011 29 F 14 0.04
007 17 M 17 0.05
004 16 F 12 0.02
105 81 M 22 0.06
001 49 M 17 0.05
015 51 M 11 0.02
106 74 M 15 0.04
009 16 F 8 0.01
107 63 F 20 0.06
003 49 F 17 0.05
006 17 F 20 0.06
014 47 F 14 0.04
002 17 F 20 0.06
012 21 F 28 0.03
016 16 F 28 0.03
113 65 M 35 0.01
110 69 M 15 0.04
112 66 F 31 0.02
114 65 F 13 0.03
005 17 F 23 0.06
101 31 F 26 0.04
109 72 F 14 0.04
104 60 F 29 0.03
108 68 M 21 0.06
008 18 F 29 0.03
013 23 F 27 0.04
116 31 M 23 0.06
102 65 M 21 0.06
103 64 M 30 0.02
111 64 F 20 0.06
115 65 M 21 0.06
010 31 M 28 0.03


It is clear that the atheists score heavily at the lower end of the scale, whilst the converse is true of the Christians. The derived correlation coefficient is very low at 0.41. This would seem to indicate that Christians assign a greater significance to ritual than do atheists,.

2. Inferential Statistics
The experimental used independent measures, which examined the ordinal data level using the Mann-Whitney Test, as shown in Table 5. This compares the ranks the individual subject scores in each of the subject groups (i.e. atheists and Christians). If one group has a rank that exceeds the rank of the other group by a significant margin this would tend to indicate that the results obtained were due to chance alone.

Table 5: Mann-Whitney Table and Analysis

Atheists:

Subject score 8 11 12 14 14 15 17 17 17 20 20 20 22 28 28 35
Subject rank 1 2 3 4.5 4.5 6 8 8 8 11 11 11 13 14.5 14.5 16

Total Rank score = 136

Christians:

Subject score 13 14 15 20 21 21 21 23 23 26 27 28 29 29 30 31
Subject rank 1 2 3 4 6 6 6 8.5 8.5 10 11 12 13.5 13.5 15 16

Total Rank score = 136

U1 = 16*16 + 16 (16+1) / 2 – 136 = 256

U2 = 16*16 – 256 = 0

Observed Value = 0

Critical Value = 75

The U value is lower than the critical value, so the difference is significant at the 5% level for a two-tailed test.

Test Proportion Observed Critical Significant/
Value Value not Significant

Mann-Whitney 5% 0 75 Significant


From these results the one-tailed hypothesis (that Christians should score higher than atheists because they have a ritual based religion) is accepted. The null hypothesis (that there is no difference between atheists and Christians and that hence there is no link between religious based ritual and PCS) is rejected. Thus, Smay's theory (2001) would seem to the confirmed, whilst Bowyer's (2005) is not.


VI. Discussion

Even a cursory examination of the results and the derived statistics show clearly that the Christians were scoring higher and with a greater degree of consistency on the key questions. The atheists score consistently lower. This would tend to support my hypotheses and suggests that Smay's (2001) linkage between OCD and Christian ritual and belief systems is valid. The Mann-Whitney test supports both this contention and my alternative hypothesis. The null hypothesis, however, that there would be no difference between Christian and atheist coping strategies is not supported. This experiment has seemingly yielded significant results.

Whether this is indeed as Smay (ibid.) claims or is in accord with the implication of Portelli's (2005.p.43) more modest claim that Christians are bound to score higher since ritual is a strong part of their culture is a matter for future conjecture. This experiment does not offer sufficient light. Yet Bowyer & Leinard's (2005. p. 7) claim supports the notion that there are strong indications that the Christians tend towards PCS, and that this is incorporated into a significant part of their Christian life.

The limitations discovered before and during the experiment were principally in finding enough representative candidates for the atheists. There is no validation of the individual subject's atheism or otherwise. I am reliant upon their individual honesty. Although one can be reasonably confident that the Christians were indeed, it may be that they as a group are not representative of Christianity as a whole, or indeed manifest the same psychological traits as other religions or congregations. A larger, more varied and better-validated group of subjects might reveal different data.

Largely for this reason the strong use of the term "OCD" was avoided in the context of this experiment, hence the adoption of the weaker term "PCS". A revision of this experiment would almost certainly have offered more challenges if the subject groups included numbers who were diagnosed OCD sufferers, either within both groups or as a separate group in their own right. I would suggest that in such an experiment, the actual OCD sufferers would score highest, followed by the Christians, and then the atheists.

Better control of the experimental environment might also affect the results. There might also be a difference in scores between life-long Christians, recent converts, and so on. It is also possible that atheists who have lapsed out of a religion might retain elements of ritual at some level. Information about the different subject's lifestyles and backgrounds might also have a bearing. An experiment that considered the response from members of different religions might also yield interesting results. However the majority of these proposals would require more complex designs and more control that I was able to exert in this limited investigation.

Bibliography

Bowyer, P & Lienard, P; 2005; "Why Ritualized Behavior? Precaution Systems and Action Parsing in Developmental, Pathological and Cultural Rituals"; Monograph, Department of Psychology, Washington University in St Louis.

Cardwell, M, Clark, L & Meldrum, C; 2004; Psychology for A2 Level; Harper Collins, London.

Fiske, AP & Haslam, N; 1997; "Is obsessive-compulsive disorder a pathology of the human disposition to perform socially meaningful rituals? Evidence of similar content." Journal of Nervous & Mental Disease, 185(4), pp. 211-222.

Greist, JH; 1992; "An integrated approach to treatment of obsessive compulsive disorder"; J. of Clinical Psychology; 35, pp. 38-41

Greist, JH; 1998; "The comparative effectiveness of treatments of obsessive compulsive disorder"; Bulletin of Menninger Clinic; 62, pp.197-207.

Moritz, S & Mühlenen, A; 2005; "Inhibition of return in patients with obsessive-compulsive disorder"; Journal of Anxiety Disorders, Elsevier, 19, pp.117-126

Portelli, C; 2004; "Advanced Brief Strategic Therapy for Obsessive-Compulsive Disorders"; Monograph found at www.centroditerapiastrategica.org/journal_english_1/Articoli_Inglese/portelli.pdf

Salsman, JM & Carlson, CR; 2005; "Religious Orientation, Mature Faith and Psychological Distress: Elements of Positive and Negative Associations"; J. for the Scientific Study of Religion, vol. 44, pp. 201, June 2005

Sica, C, et al.; 2002; "Obsessive-compulsive Disorder Linked to Piety"; Behaviour Research and Therapy, 40, pp.813.

Smay, D; 2001; "The Disease of Ritual: Obsessive Compulsive Disorder as an Outgrowth of Normal Behaviour"; Department of Anthropology, Emory University; Presentation to Myth and Ritual in American Life, 7 February, 2001