Research on astrology 01.01-03.15.PL-R.Marzewski

Robert Marzewski

Research on astrology marked as:

01.01.15.PL-R.Marzewski (phase one)
01.02.15.PL-R.Marzewski (phase two)
01.03.15.PL-R.Marzewski (phase three)
for short respectively referred to as:


or phases, was carried out in the period from August 2014 till October 2015 under the supervision of Robert Marzewski, and with participation of the following astrologers: Mirosław Czylek, Karolina Eleganczyk, and Tomasz Kiedos, hereinafter respectively referred to as A, B, and C.
Charts used in the frames of 01.01, 01.02, 01.03 included the following elements: Sun, Moon, Mercury, Venus, Mars, Jupiter, Saturn, Uranus, Neptune, Pluton, moon nodes, Placidus house system. It was also recommended to include in the chart the following: Ceres, Proserpine, Lilith, Chiron, Point of Fortune. The above rule can be derogated from by resigning from and/or adding 2 up to 4 elements. Astrological analyses performed in the frames of the research are supposed to constitute description of personality matching respective birth chart. Length of each analysis – more than one but not more than two A4 pages. Font size – 12, line spacing – 1,5.
The research got carried out based on the following protocol:
Birth data of 14 persons got provided to the astrologers. Based on the data the astrologers performed analyses. In the next phase the analyses got randomly paired – 7 pairs were created. Pairs of analyses got marked with individual alphanumeric codes and sent to the respective research participants. The participants were supposed to get familiarised with the contents of the analyses and indicate the analysis they identified with to a bigger extent. The other person from each pair was expected to do the same.
The only difference between 01.01 and 01.02 consisted in the charts being paired intentionally instead of being paired randomly. 14 charts got intentionally paired creating 7 contrasting pairs.
In this case, similarly to 01.02 charts got paired intentionally, however this time 7 pairs were selected out of 28 charts. It means only 14 charts were used, while the other 14 got neglected. Such model of pairing aimed at putting even more emphasis on the differences between the charts set up in each pair.
After finishing phases 01.01. and 01.02 and before moving on to the next phase results achieved so far should be commonly discussed, and a special attention should be paid to analyses indicated by the participants erroneously. The aim of such discussion is to discover possible shortcomings but also strengths of the astrological analyses. Unfortunately, this intermediate step got neglected as researches wanted to begin and finish phase 01.02 as soon as possible in order to be able to publish results in the originally planned time. It was a mistake and it shouldn’t be done this way as it is not in line with the adopted research methodology. Such departure from the rules should not be repeated.
Research should be carried out in the shortest time possible, counting from the moment of the research participants being recruited. It suggests their response – as long as in phase 01.01. there was no problem with receiving responses from the participants, in phase 01.02. we experienced difficulties in this respect. All in all, we’ve managed to collect responses from all the participants. In phase 01.03., however, only 12 out of 14 participants responded, in some cases after being chased 2-3 times.
Results achieved in individual phases.
correct indications – 9, including A – 3, B – 6
incorrect indications – 5, including A – 5, B – 0
correct indications – 7, including A – 4, B – 1, C – 2
incorrect indications – 7, including A – 4, B – 1, C – 2
correct indications – 8, including A – 4, B – 2, C – 2
incorrect indications – 4, including A – 3, B – 0, C – 1
Ultimately, it would be ideal if the same number of astrologers took part in each phase of the research, which turned out impossible.
Results of phase 01.02 (7 correct out of 14) are in line with the expected random distribution (0,5), so this result cannot be considered statistically significant. The situation is slightly better in the case of 01.01. (9 correct out of 14):

The statistical significance threshold here is 0,793 (beginning of the blue field). In accordance with the adopted criteria, the achieved result of 0,643 is too small to consider phase 01.01. statistically significant.
The best result was achieved in phase 01.03 (8 correct out of 12). The result is 0,667, however with the adopted statistical significance threshold of 0,817 it is still too small to be considered statistically significant.

Apparently, in both cases the general assumed threshold of statistical significance of 0,95 was not achieved. It’s worth mentioning that the threshold is adopted according to custom and not because it can constitute an unquestionable indicator of value of the achieved results.
Irrespectively of how significant criterion it is, it’s worth paying attention to the fact that the total result of all 3 phases (24 correct out of 40) being 0,6 is still above the random average of 0,5 and nearly reaches the threshold of statistical significance of 0,649.

Apart from the above described results, the research was supposed to enable observing one more effect. Now, it was not an accident that the subsequent phases differed slightly from one another. Let me remind you that in phase 01.01. charts were paired randomly, in 01.02 intentionally based on 14 charts, and in 01.03 intentionally based on 28 charts. If we assume that astrology doesn’t work, results of each phase should be the same or at least very similar, and they should be near the expected value, i.e. 0,5. In other words, in each phase there should on the average be 7 correct and 7 incorrect answers. If, on the other hand, the astrology works, results of the subsequent phases should be better, because differences between paired charts were bigger in each phase. Let me remind you that in phase 01.01 the result was 0,643, in phase 01.02 the result was 0,5, and in phase 01.03 the result was 0,667. It is hard to say then, that the above-mentioned facts are in accordance with what was expected. The results do not get better in subsequent phases. It is also worth mentioning that the results of phase 01.03 are the best, which could be expected assuming that astrology works.
The truth is, the effect of growing value of the results of subsequent phases could be expected if in each phase the same number of the same astrologers participated, carrying out the same number of analyses. As we know, this condition was not met.
Therefore, we can take a look at individual results, divided into individual astrologers. It is also not exactly what we had in mind, because disproportions in the number of analyses carried out by each astrologer still affect the discussed effect of growing value of the results. However, despite the above impediments, it can still be observed. Still, due to big disproportion between the number of analyses drawn up by each astrologer in each phase, astrologer B (6 analyses in phase 01.01, 2 analyses in 01.02 and 2 analyses in 01.01) must be excluded from the considerations.
Let’s have a look at astrologers A and C, then. Astrologer A in each phase carried out the same number of analyses, i.e. 8. In phase 01.01 A achieved 3 correct and 5 incorrect answers, in 01.02 4 correct and 4 incorrect, and in 01.03 4 correct and 3 incorrect.
Astrologer C participated in two phases and achieved 2 correct and 2 incorrect answers in phase 01.02 and 2 correct and 1 incorrect answers in phase 01.03.
In the case of both astrologers the expected effect takes place. Even in the case of astrologer B, if we take into account phase 01.01 (1 correct and 1 incorrect answer) and 01.02 (2 correct answers) in which the same number of analyses was carried out, the expected effect also can be seen.
It cannot be said that results of the astrologer B turned out to be totally useless, it’s just the opposite. However, before I remind them to you, it is worth paying attention to one more methodological aspect of the research. Astrological analyses performed for the needs of the research are carried out by a team of astrologers. However, it cannot be expected that all the astrologers will achieve the same results. It should also be remembered that astrologers are playing the role of measuring device. If work of a measuring device is in line with its programme, i.e. in this case with rules being in accordance with theoretical bases of the research, and in particular with the rules of performing the analyses, it should be expected that results shown by a device will be fairly accurate.
However, we can notice disproportions between results of astrologers A and C and the astrologer B. Two options should be taken into account here. Option one – we can assume that astrologers A and C followed the above-mentioned rules (theoretical bases and rules of performing analyses) and their results reflect possibilities of astrology, and thus astrology does not work. If so, how should we explain results achieved by the astrologer B? What rules did she follow? Did she write her analyses as a result of some kind of a prophetic inspiration? Such explanation is not particularly tempting, if we refer to research.
We should rather lean towards a claim that all three astrologers tried to meet requirements outlined in the theoretical bases and rules of peforming analyses, but the effects they achieved were different. In other words, it seems reasonable to assume that results of the measuring device B are correct, while A and C should undergo calibration. It doesn’t mean, however, that astrologers A and C would write incorrect interpretations of charts. However, analyses carried out for the purpose of the research are a different thing and one should avoid habits of one’s everyday work with clients for whom „regular“ horoscopes are drawn up.
Such a view is favoured by the fact that all the astrologers participating in the research familiarised themselves with theoretical bases and rules of carrying out analyses, and they’ve accepted them. It can be expected, then, that they tried to follow the rules. If so, a suggestion according to which astrologer B did not do it and used undefined techniques or a gift of clairvoyance would be incorrect. It is much easier to assume that B followed the adopted recommendations and effectively, and at the same time impressively used them in practice.
This being said, astrologer B, who deserves being called by her name once again, achieved results at least satisfying. The total result achieved by Karolina Eleganczyk in all three phases shows 9 correct answers and only 1 incorrect, which means 90% effectiveness. It’s also worth mentioning that in phase 01.01 she was 100% effective with 6 correct and 0 incorrect answers. The total result achieved by Karolina in all three phases exceeds the assumed threshold of statistical significance, which is presented on the drawing below.

We should remember that results achieved by astrologers A and C had growing tendency. It means that calibration of the measuring device A and C in the form of analysis of results achieved in each phase (carried out after almost each phase) and detection of errors brought about good effects. It allows us to suspect that with time astrologers A and C can come near Karolina Eleganczyk’s results or even achieve the same or better results.
I don’t want to conclude whether the research has shown that astrology does work, and thus proved legitimacy of astrological rules applied by the astrologers, and in particular by astrologer B. Even if we assumed it was true based on the observed effect of growing accuracy of answers in subsequent phases achieved by astrologers A and C, and – or in particular – based on the total results achieved by astrologer B, the conclusion would need to be confirmed in the frames of further research of this kind. If tendencies that can be noticed in this research got observed or even strengthened in the following research, the level of conviction regarding scientific nature of astrology would increase significantly. At this point, though, it would not be legitimate to draw such conclusions. It was not the actual goal of the research. Robert Marzewski, author of the project and the research supervisor was aware right from the start that one research would obviously not be sufficient to demonstrate scientific nature of astrology, being a very complex and highly heterogeneous system. The actual goal was to demonstrate that research on astrology can be carried out and that it is worth it. However, in order to do it, the research should be adjusted to its subject, which doesn’t seem to be done in other research on astrology carried out so far.
Therefore, the conclusion is that the research is worth being carried out and thus such activities will be continued. Currently Institute of Research on Astrology „AstroLab“ is working on the next research project connected with legitimacy of assigning contents described by the natural language to compound terms, or even basic terms from the language of astrology. In the next few weeks yet another research project will be launched. Anyone willing to participate in the project should enrol. By participating in the research carried out by the „AstroLab“ Institute not only will you take part in a unique undertaking, but you will also develop your professional qualifications in a measurable manner, which I wish to all astrologers.
The above article constitutes part three of a series of articles discussing research carried out by the „AstroLab“ Institute and for the first time presented on October 17, 2015 during the Vth Conference of the Polish Astrological Society.

  1. Research on astrology of the „selection based on difference“ type – theoretical basis, methodology
  2. Research on astrology – astrological analysis
  3. Research on astrology 01.01-03.15.PL-R.Marzewski – results and description

Robert Marzewski