KOINOS MAGAZINE #21 (1999/1)

Youthful Sexual Experience and Well-being

Important Conference in Rotterdam

Bob Ferguson

Since the middle of the 1970s it has become clear that sexual abuse of children occurs on a much larger scale than people had suspected until then. The justified concern about this phenomenon, which appears to take place both in and outside the family, has led however to the general view taking root that the majority of mental disturbances among adults - if not all of them - must be explained as a consequence of youthful sexual experience, and that such experience practically always leads to serious harm later in life. Recent analyses of scientific research in this field, though, indicate that a number of these assumptions are untenable, or at the very least must be nuanced.

On December 18, 1998, at the Pauluskerk in Rotterdam, The Netherlands, a study conference took place, the chief component of which was a presentation of the results of research into the assumed characteristics of sexual abuse of children, based on meta-analyses of scientific research which had appeared previously. A trio of publications on these investigations by Bruce Rind, Robert Bauserman and Philip Tromovitch, from the United States, were already discussed in earlier issues of Koinos. Bauserman and Tromovitch were present in Rotterdam to present the results for an audience which included, among others, representatives from the fields of Dutch sexology and social sciences.

The conference was organized by the Dutch Foundation for Church Social Work (KSA), led by The Rev. Hans Visser. Rev. Visser displayed a large degree of courage in once again raising a topic about which, over the last few years, opinions have apparently come to be rigidly intrenched, to reexamine the issue in a somewhat more nuanced way. The title of the conference therefore read ‘The Other Side of the Coin’, to indicate that there was more to be said about the topic than has been customarily heard in recent years. To accompany the conference a book appeared in Dutch under the same title, with contributions from various authors, published by the KSA at its own expense. The reporting in the Dutch press could be called pathetic. If any attention was bestowed on the conference at all, it was to dismiss it as marginal, and no one dealt with the essence of the matter. The importance of the conference therefore lies more in the possible follow-up initiatives which may arise in academic circles.

 Assumptions

In the world of mental health care, politics, police and the courts, but also in the media and among the public at large, a number of assumptions have become widespread regarding sexual interactions with children and youth in which coercion or a significant difference in age is present, interactions which in the literature are generally termed Child Sexual Abuse (CSA). According to these assumptions, the following characteristics are generally true for CSA: 1) it causes harm; 2) this harm is pervasive; 3) this harm is usually intense; 4) it is equally strong and equally negative for boys and girls. It is further assumed that that which is known about this subject from clinical populations (patients who are in therapy) is generally valid across the board. Rind, Bauserman and Tromovitch asked the question of whether such sexual interactions, for those who had them in their youth, indeed led to intense psychological damage on such a great scale, and which was present in equal measure for both sexes and in both clinical and non-clinical populations.

 Qualitative literature investigation

Over the last years a large number of research reports have appeared on the correlation of psychological symptoms with youthful sexual experiences, irrespective of whether coercion or significant difference in age was a factor. Although the majority of these research reports share the assumptions regarding causality, intensity, pervasiveness and equal effect for boys and girls, they are not always unanimous in their conclusions. There are in fact few conclusions which can be drawn from qualitative literature investigations, because they do not appear to be consistent, in general are burdened by the fact that their research samples were not representative, and moreover display a vulnerability to distortions because of subjectivity or inaccuracy. In qualitative literature investigation, one assembles the judgements of previous research reports and summarizes in words what appears to be being said there. In general, the authors of such investigations have for the most part concluded that CSA must be associated with a sizeable number of psychological problems, including depression, anxiety, relational problems and low selfesteem. It is not unusual for them to unquestioningly conclude that CSA is the cause of the problems found and that most people with CSA-related experiences will face such problems in later life. Some even emphasized in this that it is a myth that boys are less effected by CSA than are girls. Not all the investigators, however, agreed with these conclusions. A number of them cast serious doubt on the assumed causality because the psychological symptoms found could equally well be explained by other problems in the family situations in which the persons involved had grown up, and in which experiences of sexual abuse were only one part of a complex of possible causes for poor adjustment in later life. The most serious problem with the interpretation of this research is, however, the fact that it is almost exclusively based on clinical or juridical populations, which can be assumed to be anything but representative of the general population.

The subjectivity of the researchers also appears to be an important factor in producing a distorted image. Researchers who are convinced that CSA always causes serious damage are inclined to seek examples that will confirm this image. Qualitative investigations are by nature subjective and therefore sensitive to this sort of distortion. It is also striking that rather often they deal with the interpretation of connections based on statistical evidence in a slipshod manner. While the research results refer to something statistically significant but weakly connected, the researchers talk about demonstrable serious consequences, without introducing the necessary statistical nuances when doing so.

 Meta-analysis

In order to avoid the problems which qualitative literature investigations present, a number of quantitative literature investigations, or meta-analyses, can be done, based on statistical techniques which make it possible to compare the results of different research projects with each other numerically, and to combine them into more general conclusions. From quantitative investigations in the field of child sexual abuse and youthful sexual experiences one can arrive at an image of the ‘effect size’ of such youthful experiences. The effect size tells us how large the difference is in psychological adjustment between persons with and without CSA experiences. With both people with CSA experiences and with control groups without such experiences, one can of course measure a wide variation in factors which are connected with psychological adjustment. In both groups one finds persons for whom things are going fine, persons who are doing less well, and persons who are in a downright bad way. If the relationship between the two groups (those with and without CSA experiences) clearly differs, one says there is an important, or statistically significant, difference. With that, however, the question is still not answered of to what degree the CSA experiences can be held responsible for this. It is, after all, possible that other circumstances, which possibly often appear in combination with CSA experiences, could be partially responsible. By involving the results of comparable research into psychological adjustment - investigations in which not CSA experiences but other differentiating factors play a role - one can quantify the influence of these different factors. The effect size is generally expressed as a percentage. Thus one can state that CSA has a very strong effect on psychological or social adjustment if this factor is responsible for at least 50% of the measured variation in adjustment. A percentage of at least 25% would be termed a strong effect, a percentage of between 10% and 25% a moderate effect, and a lower percentage than that is termed a small effect.

It now appears that the effect size with regard to adjustment for CSA experiences is never strong or very strong, and moreover depends on the sort of population which is being investigated. In a meta-analysis in 1995 Jumper came to the conclusion that CSA was only responsible for 0,8% of the variation in adjustment in student populations, for 2,25% in general population samples, and for 7,3% in clinical populations. A comparable meta-analysis by another group of researchers in 1996 arrived at an effect size of 1,4% for non-clinical populations and 3,6% for clinical populations. In other words, the differences between clinical and more general populations are considerable, and the effect size is small in general populations. With this, it is empirically demonstrated that the assertion that CSA causes large scale, serious damage is grossly exaggerated and that that which is known about this subject from clinical populations is not universally valid. One can certainly assert that a statistical correlation between CSA and maladjustment is confirmed. Quantitative investigations, however, bring such a judgement back to proper proportions, and give a considerably more nuanced image of the situation.

Still, there remain a number of important drawbacks to the two meta-analyses mentioned. For instance, they almost exclusively examined female populations, and did not investigate to what degree the connection demonstrated between CSA and maladjustment was caused by CSA, or by other factors such as problematic family situations. Equally little answer was given to the question of what percentage of people with CSA experiences were reporting the effects mentioned. For Rind, Bauserman and Tromovitch these drawbacks were the reason to perform two meta-analyses of their own. Because it had already been demonstrated that conclusions from clinical populations could not be generalized, they focused on more general populations. In issues 17 and 20 of Koinos these meta-analyses have already been discussed more extensively.

 National investigations

A good way of testing the assumption that sexual abuse is a special threat to public mental health is to take a representative sample of the population and try to draw conclusions from that. Such research has indeed been carried out in various countries. Bauserman and Rind have examined the results of all English-language national investigations in this field and compared them with one another. As a very first step it is important to determine what percentage of the population report CSA experiences. Because rather different definitions of the concept of sexual abuse were employed, these percentages are obviously going to vary somewhat.

Yet the general picture arises that about 11% of the male population were confronted with such experiences, as opposed to about 19% of the female population. Three investigations were based on self-reporting and in part provided an image of the effect of CSA experiences in the long term. It appears that a minority of the boys reported negative consequences in the short term, as opposed to a majority of the girls. Continuing harm, however, appeared to be rare, and there were also positive consequences reported, more frequently by boys than by girls. Five investigations contained objective measures of psychological or sexual adjustment, and permitted the calculation of the effect size (i.e., the degree to which these experiences determined the differences measured in adjustment) of the CSA experiences on that adjustment. For males this appeared to average 0,5% and by women 1%. In other words, about 99% of the adjustment variability must be ascribed to factors other than CSA experiences.

Thus it is demonstrable that youthful sexual experiences which are experiences as unwanted can have harmful consequences. The assertions that the damage is generally intense and continuing, is pervasive, and is equal for boys and girls, must be seriously cast into doubt.

 Student populations

The national samples proved to be of great use in testing the prevailing opinions about CSA. There are, however, few investigations of this sort, and they provide insufficient information to make a judgement about the assumed causality of the harm. For this reason a second meta-analysis was carried out on a group of investigations in this field which had used non-clinical populations. They chose investigations among college populations, because there were a relatively large number of them (54 such investigations were included in the meta-analysis), and because these generally provide results which can easily be compared. Moreover, a number of these investigations also provided information about the reaction of the research subjects to consensual youthful sexual experiences, and about the factors that influenced the seriousness of the harm reported. One can, of course, argue that these investigations do not provide a picture that can be generalized, any more than investigations among clinical populations do. The divergences, however, appear to be significantly less. There were comparable percentages of CSA experiences reported, and the distribution of serious and less serious forms of abuse agreed with that in the general population. The same effect size of CSA experiences was calculated with student populations as with the national populations. Therefore it can be assumed that for obtaining general insight into the effect of CSA experiences, college samples provide substantially better data than clinical samples.

Because a distinction was made in a number of the college investigations between wanted and unwanted youthful sexual experiences, the results could be split up in this regard. When wanted experiences were taken into consideration too, it appeared that the effect size on later adjustment for men was no longer statistically significant. Among women it remained so, and the effect size was about 0,6%. These investigations also allowed them to work out to what degree certain contextual factors influence the effect of youthful sexual experiences. Other than is generally assumed, reactions in cases with a higher frequency of contacts, longer-lasting relationships or with the presence of penetration were not more negative, and the consequences were not more serious. This certainly was the case, however, for relationships involving incest, involving violence, coercion, or without consent. With regard to the differences in reactions between boys and girls, several authors who conducted the college investigations remarked that boys seemed more inclined to regard the sexual episodes as an adventure that satisfied their curiosity, while girls often experienced them as a violation of their body and as something morally reprehensible. It is also significant in this connection that women reported incestuous experiences more frequently than men, that their CSA experiences took place at younger ages, and that they were significantly more often accompanied by coercion or violence.

 Conclusions

Meta-analyses show a statistically significant relation between CSA experiences and poor adjustment. However, this link is small, and a causal connection of the magnitude that is generally assumed at the moment is not demonstrated. Other causes apparently are at least as important a factor. Lumping together experiences which involve coercion with consensual contacts results in a distorted picture. The general assumption that youthful sexual experiences are harmful, and that this damage is pervasive and generally intense, is not confirmed by this research. The assumption that the harm-effect for boys and girls and for clinical and non-clinical populations is comparable, does not appear tenable. In short, a nuancing of what is generally accepted as the truth in this field is called for.

Other speakers

Several Dutch speakers, such as the retired psychiatrist Wijnand Sengers, clinical psychologist Lex van Naerssen, and of course Rev. Hans Visser, also made important contributions to the Rotterdam conference.

Sexologist Gerda van Dijk went into the way in which sexuality plays a role in the world of human experience, in which society, emotions and the body are important factors. She pointed to the fact that it is precisely in their attitudes toward these three factors that men and women can differ considerably, so that an essentially different experience of sexuality almost goes without saying. While for men natural sexual and relational communication leads to intimacy via arousal, for women the path is generally the reverse, and intimacy comes before arousal. It is also generally true that people do not necessarily experience the same forms of sexual contact in the same ways. Van Dijk listed four essential points as norms for good sexual relationships: mutual respect, taking into account each other's boundaries and developmental phase, and, lastly, communication. Where things do go wrong, one often finds sexual violence, physical violence and emotional neglect going hand in hand. These phenomena generally appear to be inseparably linked.

Van Naerssen observed that many of his male patients experience sexual problems because they were shown too little physical affection in their youth, not because they were maltreated or abused. Coming himself from a country in which affectionate mutual physical contact between children is very normal, on his arrival in The Netherlands it struck him that physical contact among Western boys chiefly takes place in the form of hitting and kicking. Paying attention to the sociology and psychology of the environment in which children grow up is more important than morals legislation that focuses on moralism, and the misconception that everything can be regulated by law.

Gert Hekma, instructor in the Homostudies department at the University of Amsterdam, labeled the way in which we provide children with information about sexuality too biological and too negative. Rather than beginning with the traffic code, we first teach them how the motor works. We will have to acknowledge that children are sexual beings and we will have to make an active investment in their maturity. Along with the responsibility the government has to invest in sports and cultural education, there is a similar responsibility to stimulate a positive sexual infrastructure that is based on living, and not on concealment and shame.

Bauserman, R. and Rind, B., Psychological Correlates of Male Child and Adolescent Sexual Experiences with Adults: A Review of the Nonclinical Literature. In: Archives of Sexual Behavior, Vol. 26/2, 1997, pp. 105-141

 Rind, B. and Tromovitch, P., A Meta-Analytic Review of Findings from National Samples on Psychological Correlates of Child Sexual Abuse. In: Journal of Sex Research, Vol. 34/3, 1997, pp. 237-255

Rind, B., Tromovitch, P. and Bauserman, R., A Meta-Analytic Examination of Assumed Properties of Child Sexual Abuse Using College Samples. In: Psychological Bulletin, Vol. 124/1, 1998, pp. 22-53

 Visser, Hans (ed.), De andere kant van de medaille. Stichting KSA (Walenburgerweg 55, NL-3039 AD ROTTERDAM, The Netherlands), 1998. ISBN 90-5782-016-1