Compatibility of People — The Horoscope vs Data Science!
I have never believed in the predictions of the horoscope, however, in more than one occasion I have enjoyed reading about this idea that the future of a person can be foreseen in the stars. This has guided me to learn a little about the movement of the planets, the Ecliptic, the names of some stars and constellations, the zodiac and a good amount of Greco-Roman mythology. Notably, an intriguing concept for me is the one about compatibility of people according to their zodiac sign, the one I will briefly explain you next!
The 12 signs of the zodiac
The zodiac shows the position of the Earth, the Sun and some constellations at a point in time. You can imagine it as a giant clock where the Earth is the center of the clock, the Sun is the handle and the numbers are the stars. Here, the units of measurement are not the minutes, the hours, they days, nor the months, but the zodiac signs. These are divided into 12 and can be adapted to out calendar according the Image 1. Therefore, you could say that something happens at the end of Leo instead of saying that it happens on August 22nd. You would also know that the Sun is set on the constellation of Leo, which means that you wouldn’t be able to see Leo at night. In summary, the zodiac is a clock or calendar, but also a map of the stars.
The compatibility of zodiac signs
The horoscope is the birth of someone or something represented in the zodiac, for example, your birthday. According to astrology, the horoscope of people has a great impact on their personality. And since some personalities are more compatible between them than others, then the horoscope could be used as a tool to predict the compatibility between people. The image 2 shows some of the zodiac signs couples with the best and worst compatibility that I found on the internet. I want to be clear that the compatibility of zodiac signs is not the only tool that astrologist use to predict the effectiveness of couples, but it is one of the most popular.
Data science with zodiac signs
Previously, I had wondered how certain could the compatibility of zodiac signs be when applied to a large number of people, however, I never searched statistical information about it. Many years later, it came to my mind that I could use my knowledge as a beginner in data science to explore how real this was. In order to do this, I had to know 2 things:
1. The zodiac signs of a large number of people and their respective partners.
2. Know how good their relation works.
I thought that I could explore both points by downloading a public database of marriages and divorces that included the birthday of the spouses to assign them their zodiac sign. And that’s exactly what I did with 15 years of data in the city of Xalapa, Mexico. There, they registered approximately 24000 marriages and 5000 divorces from 2000 to 2015. Fortunately for me, I was able to automatically assign all the zodiac signs with a short program.
Finally, I evaluated the effectiveness between zodiac signs couples, or in this case, the ineffectiveness between couples as the number of divorces divided by the number marriages for a specific couple. It is important to note that the registered divorces do not necessarily correspond to the marriages registered on the same period of time. This is because I only have data for 2000–2015, and it could be that a divorce registered on 2005 corresponds to a couple that got married on the 90’s. I know it is not the ideal way to conduct this analysis but at least it gives me an idea.
Data analysis
The first thing I did was to build a table with the quantity of marriages for each zodiac sign couple. You can see the larger numbers with deeper colors so you can identify them faster. This way you can see that the couple with the greatest quantity of marriages are men of Virgo with women of Leo with a total of 209 marriages, and that the lowest quantity of marriages happens for men of Capricorn with women of Aries. Also, you can see that couples at the bottom right corner seem to have a lower quantity of marriages than the other couples. Could this be a revelation about the compatibility of people according to their zodiac sign?
Well, I don’t think so. What I think is that this difference in the quantity of marriages is attributed to the availability of people for a particular zodiac sign. In other words, there must be less people with a zodiac sign between Scorpio and Pisces. You can see this in the next graph, which shows that the number of births for women and men is below the average between the months of November and April. This is not strange, since sometimes the birth rate can be affected by events such as holidays or weather seasons.
Now is the time to explore the divorce/marriage ratio for each of the proposed zodiac sign couples. Here, we get low ineffectiveness if a couple has few divorces and lots of marriages. Similarly, we get high ineffectiveness if a couple has lots of divorces and few marriages. These values are shown in the table, where values of ineffectiveness above the average are marked in red (> 0.178), and ineffectiveness below the average is marked in blue (< 0.178).
We are looking for zodiac compatible couples to have low ineffectiveness values. This is because compatible couples should result in better and long lasting relationships without divorces. On the other hand, non-compatible couples should eventually have relationship problems that might guide them to a divorce, and therefore, increase their ineffectiveness value. With these results we can now create a table to count in how many of the proposed zodiac couples agree with the calculation of ineffectiveness. This means that ineffectiveness is correct when it is below the average for compatible signs, and above the average for non-compatible signs. Otherwise, ineffectiveness is incorrect. Here, there is a total count of 42 because each of the 21 zodiac couples are evaluated for man — woman, and woman — man couples.
Final thoughts
As I said at the beginning, I do not believe in the predictions of the horoscope, now even less after writing this article, however, I understand that some of the analyzed data is not ideal. First, I would have enjoyed to have millions of marriages and divorces registered all over the world for a larger period of time. This way, the statistical part would be more meaningful. Another aspect that could be improved is the calculation of the ineffectiveness, if possible at all.
Finally, did you know that the zodiac sign dates need to be adjusted as time passes similarly to the way we adapt leap years? This has created the existence of the tropical zodiac, and the sidereal zodiac. Did you know that the time the Sun sets on each of the constellations can be significantly different? As an example, the Sun sets in Scorpio less than a week, yet the period for the zodiac sign of Scorpio is about a month. Did you know that the constellations of Ophiuchus and Cetus are also in the Ecliptica? This would result in a total of 14 zodiac signs if they are ever considered. Also, we normally focus on the Sun to evaluate the zodiac sign but, did you know that the Moon can also be used to evaluate another zodiac sign? This moon sign is more difficult to evaluate since it requires your birthday down to the hours, minutes and seconds as well as the place you were born, but according to the astrologists, the moon sign influences heavily on the compatibility of people too. What I am trying to say is that to increase the validity of this study, it would be helpful to get more statistical data and to increase the astrological considerations, which are still intriguing to me.
