Correlation analysis: Spearman’s rank and Pearson’s coefficient in business astrology

Ever wondered if Mercury retrograde leads to fewer sales? Or Pluto in Capricorn empowers governments?

While some parallels between planetary movements and economic trends are evident, not all astrological claims can withstand the scrutiny of a scientific test.

In this article, you’ll see how we can use Pearson’s coefficient and Spearman’s rank to verify astrological prognoses with real-world data.

Applications of Spearman’s rank and Pearson’s correlation coefficient

Both Pearson’s test and Spearman’s rank can help us identify linear relationships between two independent variables. This allows us to compare astrological and business data sets and determine whether a positive or negative linear relationship between them exists.

For example, we can test if Mercury retrograde days and the number of cars sold are linearly correlated (SPOILER: they are not).

Spearman’s and Pearson’s tests can be used for trend forecasts, including predictions on price development, sales volume, and marketing metrics. Apart from economic predictions, both tests can help academic research into the accuracy of astrological prognoses.

Pearson’s correlation coefficient

Pearson’s correlation coefficient can be used to measure the strength of linear association between two variables, ranging from -1 (perfect negative correlation) to 1 (perfect positive correlation). Pearson’s coefficient produces the most accurate results when data is normally distributed.

Using Pearson’s correlation coefficient

We can use Pearson’s coefficient to test for a linear relationship between a businessvariable (e.g. sales, price, views, etc.) and the movement of a planet through the zodiac (using 0 to 360°) or through a specific sign (0° to 30°).

Case #1: Uranus in Taurus and electricity prices

An astrologer posits that Uranus’ entry and movement through Taurus is linearly correlated with the national electricity prices. They have managed to secure a data set containing bi-annual average energy prices for household consumers (in cents/kWh). With this is hand, they compare energy price to the exact position, measured by the number of degrees, of Uranus in Taurus.

Based on the data set above, they calculate Pearson’s r using the formula:

where x and y are the variables to be compared.

For this data set, the astrologer receives an r value of 0.87031418, which suggests that there is a strong linear correlation between Uranus in Taurus and electricity prices (in the chosen sample). And indeed, such a relationship can be observed when we map both variables on a scatter plot and take note of the upward sloping trend line.

The astrologer confirms the statistical significance by calculating the p-value. As p = 0.0001091, the results are statistically significant. However, with just 13 data points in this sample, the findings should be considered with caution.

Case #2: Pluto in Capricorn and the number of employees in public service

Pluto in Capricorn is traditionally interpreted as a period when the governments, authorities and institutions increase their influence and power. One way to measure state power is through the number of employees working for government institutions.

An astrologer therefore expects a positive, linear correlation between the degrees of Pluto in Capricorn and the number of people employed by the state.

The astrologer decides to verify this hypothesis using data on the number of people in public service in Germany during the most recent Pluto in Capricorn transit (2008-2024).

They calculate Pearson’s correlation coefficient and receive an r value of 0.922. This means that there is a very strong positive association between the number of state employees and the Pluto in Capricorn transit. With a p-value of 0.0000003734, the result can also be seen as statistically significant.

However, looking at the scatter graph we can see that the data is not perfectly linear, but may instead be monotonic – meaning that Spearman’s rank correlation may be a better correlation test.

Case #3: Pluto transiting the zodiac and the price of the S&P 500

A financial analyst believes that a change in stock market prices is linearly correlated to the movement of Pluto through the zodiac. To test this hypothesis, they ask an astrologer for Pluto’s position in the zodiac (0 to 360°) at the beginning of each year since 1900.

The analyst decides to compare the annual change in the position of Pluto to the price of the year opening price of the S&P 500.

Based on the following data, they calculate Pearson’s correlation coefficient.

The analyst receives an r value of 0.1126 and a p-value of 0.27048, meaning that there is a weak positive, linear correlation but that the results are not statistically significant. They confirm their findings by mapping data on a scatter graph – and indeed, there is no clear distribution.

Spearman’s rank correlation

While Pearson’s correlation coefficient only measures the strength and direction of linear relationships between two variables, Spearman’s rank works for all types of monotonic relationships. This means that a change in X does not need to have a proportionately equal change in Y, because the ranked order of the data points is considered rather than the data points themselves.

Spearman’s rank is equal to the Pearson correlation coefficient between the ranked variables. Using Spearman’s rank is therefore useful when you have a compelling reason to assume that a monotonic, but not necessarily linear, relationship exists between variables.

Case 4: New passenger car sales and Mercury retrograde days (per month)

A researcher and an astrologer disagree on the impact of Mercury on sales. The astrologer claims that retrograde Mercury slows down negotiations, leading to fewer sales during this phase. The researcher believes that Mercury’s movement should not impact sales figures.

To find out who is right, the researcher gathers monthly sales figures and tallies up the number of retrograde days in each month. But before they can conduct a Pearson’s test – the astrologer notes that Mercury retrograde may start in one month and end in the next. The effect may last a little longer than the precise days of the retrograde period, according to the astrologer, so one should not assume a linear relationship.

The researcher is disgruntled by this objection, but then recommends that they use a Spearman’s test to verify if a monotonic relationship between Mercury retrogrades and car sales exists.

The researcher crunches the numbers and finds that rs = 0.0105, meaning that no relationship exists between the days that Mercury is retrograde and number of cars sold each month. They confirm their findings by visualising the data on a scatter plot – and the astrologer begrudgingly agrees that no apparent relationship between both variables seems to exist.

Limitations of correlation analysis in astrology

While we have seen some possible applications of correlation analysis in astrology – the accuracy of findings may be limited by several factors.

Correlation analysis can only offer insights when fundamental assumptions hold true, including the monotonic distribution of data, and the availability of a sufficiently large pool of data.

Additionally, correlation coefficients are a tool for single factor analysis – meaning that they can never consider the full picture in a horoscope. As a result, the symbolism and interdependencies between planets and zodiac signs may fall by the way side.

In using correlation coefficients, researchers are looking only at a very small fragment. Just like a blind person can only feel (but not see) an elephant’s tail – they will not be able to see the synergies between multiple factors if they limit themselves to statistical tools and the belief in rationality.