Abstract

Twenty-four female subjects were asked to recognize as true or untrue complex personality characteristics describing themselves and to select one of three personality profiles as their own; personality information had been derived by "blinded" astrologers from natal charts representing the moment of birth. Three different experiments varied as to the complexity of the astrologically derived personality characteristics, method of test material administration, and subjects' knowledge of the astrological basis for personality information. Overall results for the three experiments evaluated using cumulative binomial distribution were significantly non-random, with p<.001 for 15 valid trials and p<.01 for all 24 trials including nine found non-eligible for inclusion. These results supported the validity of astrology's capability to generate unique personality descriptors that subjects affirm by selection as representative of their own personalities.

Historical Perspective

Early civilizations, including the Babylonian, Egyptian, Chinese, Aztec, and Greek, all had established systems of astrology in order to study the influences of the sun, moon, stars, and planets on such earthly events as the growing season, birth, and death, and to determine the best times to perform rituals. In the West, even up to the time of Copernicus, Galileo and Newton, the exploration of astrology was an accepted practice.

The Age of Reason, however, introduced doubt into the acceptance of apparent mystical or mythical phenomena with the then-emerging scientific techniques available to challenge and explore every formerly mystifying natural phenomenon. Measurements of the awesome distances to the planets and the small amounts of gravitation, magnetism, and radiation exerted by celestial bodies on earthly life forms resulted in the apparent lack of a physical explanation for the influence of astrology; thus, any possibility that planetary bodies exert determining forces on a baby at the moment of its birth seemed doubtful (Bok, 1975).

Since the time of the Lockeian era with its assertion of man's independence and free will, conflict has raged regarding the question of the influence of planetary forces on mundane events or on the human personality. Astrologers, on the one hand, have been willing to continue to trust the "principle of synchronicity,, (Jung, 1960), the connection and balance between all life forms in the universe, while scientists, on the other hand, have been trained in the most rigorous scientific method to not accept any phenomenon as a fact without repeated, replicable testing using agreed-upon assumptions of proof.

Despite widespread popular acceptance and belief, astrology has been perpetually denounced by some scientists as a foolish, magical, irrational and pretentious pseudoscience without scientific foundation (The Humanist, 1975). Despite the disbelief of these vocal scientists, thirty-two million adult Americans, or one in five, express a belief in astrology, according to a Gallup Poll (1975). Nine persons in ten can name their sun sign. A poll by Wuthnow (1978) revealed that 48% of college graduates believe in astrology. This is a considerably higher percentage than the overall national average reported by the Gallup Poll. Astrologers view this as a vindication of the controversial profession, while many scientists think the renewed interest in astrology is an alarming indication of retreat into irrationalism by the public. Unfortunately, the controversy has involved little in the way of reasoned and constructive dialogue between astrologers and their critics. The design of this study is the result of one of a very few such dialogues.

Following The Humanist magazine articles against astrology, Dr. George 0. Abell and Neil Marbell debated the merits of astrology on WBBM-TV (CBS Chicago). They agreed to develop a definitive test satisfactory to both skeptic and astrologer and to conduct the study. After enlisting the help of David Kanouse, a senior social scientist at the Rand Corporation, the agreed upon procedures to be followed were submitted for review to the Social and Developmental Psychology Program of the National Science Foundation; NSF staff conducted both an informal and a formal 'inside' review, and had the methods reviewed by an 'outside' panel of experts. In addition, reviews were conducted by astrological panels and interdisciplinary panels of scientists. A special review was also conducted by the Hon. Jerome Ambro of the Science and Technology Committee, U. S. House of Representatives. Prominent astrological practitioners and professional astrological organizations were also requested to review and approve the methodology. All reviewers and panels concluded that the method to be employed was adequate for an impartial test of the correspondence between astrology and personality.

While it is understood that any single study cannot be conclusive, a great part of the truth of that maxim rests with the possibility of differing views from different quarters. A lack of consensus (a) of the interpretation of the results, or (b) on the methods by which the results were obtained could be at issue. The participation and consensus of the opposing factions and disinterested scientists as well are a feature of this study. While the following letters were obviously not intended for presentation in this report, their presence here is to pre-empt any possibly recalcitrant participants and establish clearly for the reader who might be skeptical, the extent, of the participation of any or all concerned. For this reason the letters are reproduced here. The letters are as follows: Letters A & B show the approval and endorsements of two major astrological organizations. Letters C & D show the approval of two prominent astrologers. Letter E shows the involvement of the Committee to Investigate the Claims of the Paranormal (CSICOP). Letter F shows the involvement of Dr. Abell. Letter G shows the involvement of Dr. Kanouse. Letter H shows the opinion of the National Science Foundationpeer reviewers on methodology. Letter I shows the approval of Dr. Teger, the Director of the Personality and Developmental Psychology program at NSF. Letters J & K, while addressed to the National Science Foundation, must suffice to reflect the positive results of review process. More special information is obtainable directly from them.

Rationale

It is important to state the reasons and motivations underlying this unusually cooperative project beyond those of 'proving a point.' The great value of a validated astrology is as an amazingly useful tool for understanding our human species.

Beyond even the great practical potential is the significant intellectual consideration. A proven astrology links the two primary intellectual pursuits of humankind, the physical sciences and the life sciences, into a unified system. These two systems are distinctly separate in the minds of many. The life sciences relate to experience by quality, aesthetics and natural response; the subtle difference between two master violinists is an example of an application of the principles of the life sciences. On the other hand, the physical sciences relate to experience by measurement, probability, and natural law. For example, the precise difference in centimeters of two lengths is an application of the principles of the physical sciences. In most cases each of these two cultural endeavors are not confirmable by the standards of judgment of the other.

Life science psychologists cannot precisely describe the personality of an individual without essentially asking him or his peers. A physical science geologist could not conceive of merely asking a rock its weight; he would weigh it and report the data in exact terms. For the physical scientists, the following project demonstrates that the personal elusive and individual natures of humankind may be directly related to the natural physical ' world in independently measurable ways satisfactory to their Quantifying instruments. For the life scientists we demonstrate that the Universe reflects our reality and provides pattern, purpose and meaning, satisfactory to their range of responsive understanding.

If pursued, the system established here will provide for the physical scientist a new science to explore: the components and nature of time. To the evolving culture of our society, this system offers perspective, understanding and personal dignity. These qualities have the power to restore social conscience, add richness to the life experience, and provide a new holistic image of man by correcting the illusion of chaos held by contemporary society. The benefit available to broad segments of our population are sufficient to encourage serious consideration of the findings presented here.

Definition of Astrology

Astrology is a coherent mathematical system which posits that the individual's personality is defined and influenced by the position of the sun, moon, and eight planets along the zodiac at the moment of birth. The zodiac is an area of the sky 32 degrees wide that extends in a circle around the earth roughly parallel to the equator. The zodiac is divided into twelve equal portions, each assigned a particular sign. The position of the sun in the zodiac at the moment of birth determines the individual's sun sign. Although popular conceptions regarding astrology are primarily based on sun sign alone, the natal chart, (or "birth chart,' or "horoscope') is a more complex, highly individualized 'map' of the major planets in the solar system at the precise moment of birth, from the perspective of the birth place. These planets are delineated according to all twelve signs into a structure of twelve houses, with geometric relationships among planets ("aspects') also defined. An astrologer who works with an individual client uses this systematic map and all elements combined to study that particular personality; the astrologer might find, for instance, that other planetary forces restrict, amplify or modify the expression of the sun sign in that individual.

Review of Pertinent Research

Research regarding ancient astrology's reputed ability to predict specific events has been inconclusive at best (Dean, 1978). Far more importantly, however, studies exploring the accuracy of astrological explanations of the human personality have developed in response to modern psychology's attempts to understand complex and often unpredictable human personalities. Rudhyar (1970) has attempted to combine holistic Eastern thought with Jungian concepts regarding the importance of archetypal symbology in his development of 'humanistic astrology.' The field of astrological counseling, which uses astrological interpretation of an individual's birth chart to diagnose and counsel, has gained substantial use as a complement to more traditional therapies in assisting the individual to understand limitations and accept potentialities of the personality (Arroyo, 1975).

Most attempts to determine if any scientific basis exists for astrological interpretation of the personality, however, have focused on single unitary aspects, either of personality or of astrology; these attempts have primarily used psychological personality inventories to compare astrological indices with traditionally standardized measures or descriptions of personality One common method has been to attempt to determine the relationship between the sun sign and various personality f actors. For instance, whether an individual's sun sign corresponds to predicted astrological patterns of introversion and extroversion (first sign Aries extrovert, second sign Taurus introvert, continuing in an alternating pattern) has been the subject of several studies using various personality inventories (Foriano & Ehrlich, 1941; Mayo, White, & Eysenck, 1978; Smithers & Cooper, 1978; Veno & Pamment, 1979); some of these studies resulted in evidence supporting astrologers' predicted pattern exactly, and others found no significant correspondence.

Pellegrini (1973,1975) found a strikingly significant relationship between the sun sign of a person's birth and scores on the Femininity scale of the California Psychological Inventory, although he found no significant relationships for scores on the Communality, Socialization, and Flexibility scales. Standen (1975),attempting to replicate Pellegrini's finding, discovered an alternative pattern of correspondence between high Femininity scale scores and the sun sign (persons born between January 21 and July 23 had lower femininity scores than persons born in the other half of the year). Illingworth and Syme (1977) found sex effects but no birth effects in a similar study; Tyson (1977) and Mayes and Klugh (1978) in replicative studies.

Silverman (1971) used only sun signs to determine if undergraduates ranked the certain adjectives associated with their specific sun sign higher on the Rokeach Value Survey. He also applied sun sign to marriage and divorce records. Neither technique supported the supposed influence of the sun sign. Silverman and Whitmer (1974) tested whether personality characteristics attributed to the sun, moon, and ascendant (the sign on the eastern horizon at the moment of birth) were related to subjects' and subjects' friends' rankings of eight attributes on nine-point scales and scores on the Eysenck Personality Inventory. They found no relationship between the astrologically derived characteristics and personality rankings.

Gauquelin's work (1974), as well as that of Gauquelin and Gauquelin (1970-71 1972-77), has explored extensively the relationship between career choice and the location of important planets in the birth chart, rather than employing the sun sign as the only basis of astrological interpretation. They found that a disproportionate number of military leaders were born when Jupiter was near the eastern horizon or at upper culmination; they also found statistically significant planetary positionings in the birth chart of famous athletes (Mars), scientists (Saturn), and persons who have gained renown in the literary field (Moon). Tigg (1976) found a strong correlation between the astrological 'strength" of Mars and Buss-Durkee Inventory Scores for hostility.

Very few researchers have attempted to focus on one characteristic of life or personality (such as marriage, divorce, introversion, femininity, career) or to use the entire birth chart (with its complex of sun, moon, eight planets, twelve houses, twelve signs, and relationships between the planets) to address the personality in a more holistic manner.

Clark (1970) devised three experiments testing the accuracy of astrological interpretation of the entire birth chart. In the first test astrologers attempted to match ten birth charts with ten descriptions of occupation. In the second, the astrologers were given ten case histories and ten pairs of charts; in each case they were to select from the pair the chart which matched the case history. In the third test, the astrologers were again given ten pairs of charts which they had to match correctly; one was for a person of high intelligence but otherwise normal and one was for a person who had cerebral palsy.

The first two tests were blind, the last double blind; in each test a control group of twenty psychologists and social workers with no astrological knowledge were given the same test. The control group performed at exactly chance level; the astrologers selected at an extremely high level of statistical significance in tests. Supporting evidence for the validity of astrological interpretation was gained in replication studies by Astrology Now (1975,1976), Dean and Edwards (1976), Joseph (1975), and Press (1977)

Another study which used several components of the entire chart, although isolating them from one another, was Hume and Goldstein's 1977 study. They attempted to determine if positioning of any of the ten planets in any of the twelve houses or twelve signs, the ascendant (point rising on the eastern horizon at the time of birth), or five aspects (geometric relationships between two planets) corresponded with extreme scores on various scales of the MMPI and the Interpersonal Check List. Of 632 chi-squares tests performed, 23 were significant, fewer than the chance level of 32

A few more recent studies have addressed the question of whether individuals accept astrological explanations of conditions that acceptance operates. Snyder (1974), for instance, gave subjects the identical generalized personality interpretation but randomly varied assigned subjects to one of three specificity conditions in which they were told the profile was: (a)generally true of people; (b) based on month and year of their birth; or (c) based on day, month, and year of their birth. He found that the I more specific the birth time referent, the more the subjects accepted the interpretation as an accurate description of their personality (i.e., the "Barnum effect"). Delaney and Woodyard (1974) gave high school students astrological personality descriptions that differed on elements of dominance and change; afterwards, subjects completed domination and change sub-scales on a personality inventory. Results indicated that self-reports were influenced by the astrological personality descriptions. Silverman (I 971) found undergraduates more likely to agree with personality descriptions if they thought their sun sign was the source of the description; but if the astrological basis of the information and subjects' affiliation with a sign was not specified, subjects did not choose the personality description fining their sun sign. Snyder, Larse and Bloom (1976) found subjects more likely to accept as accurate a profile that they were told was based on either psychological, graphological, or astrological procedures than when told the profile was "generally true of people." there were no significant differences in acceptance of any of the three specific information sources.

A major methodological difficulty with self -report of personality description, then, is the evidence of a strong acquiescence bias, particularly when the response required is a simple yes or no. However, acquiescence has questionable influence on subjects' selection of personality descriptions. For instance Hampson, Gilmour, and Harris (1978) found that subjects did not gullibly accept statements that referred to characteristics shared by most people. Subjects did identify astrological statements of personality as true of most people, but in describing themselves they accepted or rejected characteristics specifically on the basis of 'truth' (results from a personality test), rather than on the basis of generality or social desirability. Furthermore, they found the astrological statements lacking in social desirability, even though 'true of most people.' Hampson et al. found these results supportive of the theory that subjects possess self insight, contrary to mixed evidence regarding gullibility. Another methodological difficulty with this group of self-report studies is the assumption that high school students and college sophomores are as self-perceptive, self-aware and/or gullible as the general population.

Almost no studies attempting to determine the validity of astrological interpretation of the personality, then, have examined descriptions of multiple factors of the personality that are generated from a birth chart (for instance, whether some other planetary influence substantially modifies expression of the sun sign), which is the method most similar to that which an astrological counselor uses in a session with a single client. In addition, since the discovery of acquiescence bias and the Barnum effect, little is known about sex or age differences in these effects. Almost no studies have been designed to control for acquiescence bias and the Barnum effect regarding specificity of information source such as allowing subjects the opportunity to select the most fitting astrological personality description from among several.

The three experiments reported here were designed to contribute to the determination of whether astrology does indeed have any empirical basis by using a holistic and integrated method of chart interpretation rather than a singular astrological factor such as sun sign, and controlling for both acquiescence bias and subject knowledge of astrology as the source of personality information.

General Method

Three experiments were conducted which varied in control setting, method of test material administration, and subjects' knowledge that personality information was based on astrologers' interpretation of birth charts.

Subjects

Subjects were all mature females, ages 31 to 55; both age and sex were selected to minimize variability in self-insight and defensiveness (Maccoby & Jacklin, 1974). All were 'white collar' office employees who were contacted by their personnel director for willingness to take a "personality test."

Test Materials

Test materials for all three experiments were personality profiles generated by a team of astrologers and derived solely from birth information. Specific time and place of birth were used to construct a 'birth chart,' which, according to the principles of Ptolemaic astrology, combines ten planets, twelve signs, and twelve houses into a highly individualized 'map" of the solar system at the precise time of birth (Parker& Parker, 1975). Engrams, which are linear constructs that integrate these various elements in a structured format, including the planetary associations of signs on house cusps, formed the basis for very specific and highly personalized descriptions of different personality features (outward appearances, value system, communication style, selfimage, philosophy, relationship style, etc.). The sample profile in Figure 1 indicates the complexity and specificity of the personality descriptions utilized here, which differed significantly from common and simplistic 'sun- sign' astrology (eg., 'You're a Taurus, so you must be stubborn, patient, and sensual.'). Since this very personalized birth chart information is unknown to the general public, subjects' prior knowledge of their sun sign characteristics was minimized as an influence. All astrologers used birth chart information alone in developing the personality descriptions; all were blind to the subjects.

A two-page profile for each subject presented numbered statements dealing with topics of personal reactions or personality features. Experiment One profiles contained 17 topics, Experiment Two 14 topics, and Experiment Three 13 topics. Each subject was given three of these profiles and asked to mark "yes," "uncertain,"or "no" to each of the topic statements in each of the three profiles. Secondly, a 'profile' score was obtained by subtracting the number of "no" responses from the number of "yes" responses. Lastly, each subject had to select one profile from the three that most nearly matched his own personality.

The other two profiles in each set of three were generated from birth data of two other members of the test population. Thus, asocial desirability response bias was controlled since any general tendency to select or reject a given profile would be balanced in each set of three subjects; additionally, topic statements were thus based on actual people rather than being abstract generalizations (Forer, 1949), or extreme statements of opposite characteristics (Hampson et al., 1978) as used in previous research. Primacy or recency bias was also controlled by stacking the three profiles in the same order for all three subjects in a set.

Experiment One

Method

This was the best controlled of the three experiments. Two complete mutually exclusive sets of three subjects, or six in all, were tested in a laboratory setting. The second set contained two subjects with the same sun sign. Personality files, composed by individual astrologers from birth data alone, were revised by a committee of five astrologers in order guarantee consensus, uniformity in style, content and overall presentation.

Results

All six of the subjects tested selected the profile that matched their individual astrological factors, rejecting the two that matched the other members of three-subject set. However, one of these subjects was disqualified because it was discovered her birth information was incorrect.

The 100 percent correct profile selected by all five qualified subjects contrasts with the 33-1/3 percent that would expected by chance. The binomial probability that all five subjects would by chance select the correct profile is .004.

Conclusion

The following comments come from an earlier draft and not the final peer reviewed paper published in NCGR:

In the experiment, nine of the subjects were disqualified by the monitors for various reasons: incorrect birth data, failing to return tests within a reasonable time span and completing the tests at their desks with their regular work load.

Of the remaining 15 subjects, there were 12 successes. "The best overall data for consideration is that 80% of the subjects selected the preassigned profile correctly where only 33 1/3% would be expected by chance." This results in a p value of .001. "Thus highly confident results in verification of the hypothesis can be reported."

... for a copy of the full article with tables contact the NCGR Research Journal