The p-value for the null hypothesis that the level of commenters would not be associated with the rate of Yule I increase is less than .01 (p<0.00635). As this is below all standard levels of significance, we reject the null hypothesis in favour of the alternative hypothesis that there is an association between comments and a Yule I increase. The result of the experiment produces this premise:
There is a positive relationship between the number of comments in a thread and the rate at which the Yule’s I characteristic increases.
We can also use this equation to draw a line predicting the meaning generated in a thread using the coefficient of the X Variable1 (0.02) and the intercept (-9.09).
\(\begin{equation}{Yul}e^{{}^{\prime}}s\ I\ Score=\ -9.09+\text{Comments\ }\times\ 0.02\nonumber \\ \end{equation}\)
This means what with each extra comment in the threads studied, the Yule’s I score increased by 0.02.
Why?
Therefore, the Polemic Science issues on Reddit studied were sites of the Complexity Loop. Consensus does not arise, but this is a rich environment for the generation of new ideas.

Conclusion

Yule's I indicator provides a powerful, automated method to analyse online conversations.  Using  the Simplicity Loop and the Complexity Loop one can determine if new ideas are being developed or if a coherent, hegemonic narrative is developing.  Using Luhmann's definition of meaning,  the product of the different choices that a system makes to deal with complexity, this tool allows us to see if those choices are accumulating or simplifying.
Looking at Polemic science issues in reddit, I found that many new ideas were generated- issues became more complex. However, this was just a pilot study opening up a new area of study. Yule's I and feedback loops could allow for analysis of more threads- even the entirety of Reddit. Furthermore, the tool would be far more useful if the Yule I scores were compared across different websites. I hope, in the future, to research the feedback loops at work on social media sites like Twitter, VK and Facebook, and also online news sources like The Huffington Post, Business Insider, and Breitbart.
Feedback loops are a useful tool to understand how people decide what things mean online, and I hope that this paper opens the door to more research.