Thanks for the kind words Shivam!

Hey Ko,

Thanks for commenting! 🙂 If you want, I can send you the sheet that parses the exported data from the SleepasAndroid app into a structured datatable? That way, you can track your own sleep using their app, and analyze the data on your own.

Let me know!

I know it is a difficult requirement so feel free to refuse me.

Thank you!!

Thanks for taking the time to comment! 🙂

I don’t know why I should remove the weekends per se for this analysis? It could result in different results but I’d just be focusing in on a specific type of day. This analysis is part 1 in a series, and therefore acts as an overall introduction really. The next step is to dive deeper, and as you say, I’d like to compare workdays vs. days off, or waking up with an alarm vs naturally waking up.

Trying to model my happiness based on the knowledge of every single happiness factor (love, work, sports etc..) sounds compelling, but I think it’s practically impossible. It’d still be very interesting to try though! 🙂

Thanks, Anthony! I really appreciate the nice words.

I guess there are a lot of different resources that have varying observations (as is with just about anything on the internet lol). I agree with you that your brain functioning is impacted by a sleepless night. A commenter on Hackernews was able to explain my sentiment much better than I ever could, so I’ll just quote him here:

“If you have a day of bad sleep (i.e. non-optimally consolidating the day’s events), followed by a day of good sleep, it’s not like your “health bar” is permanently lowered.

What’s lost forever by that poor sleep is the _opportunity_ to properly synthesize those 20 pages just on that one night. Not your “health”, and you don’t get a permanently increased level of cortisol. After a good night of sleep, you do physiologically recover and you’re back to normal.”

There is still so much about sleep (deprivation) that we don’t yet know for sure. That makes it extra interesting for me to continue to analyze this data! 🙂

Good question, but I’d like to first mention that a small correlation does not directly mean that it’s insignificant. Given the fact that my happiness is influenced by a virtually endless list of other factors, I’d say it’s pretty impressive that my sleep still has a distinctive influence, especially on the bad days.

With that said, you are right. My lack of sleep could actually be a result of unhappiness. This is one of the typical cases of correlation <> causation, which is a recurring theme in my happiness studies! 😉

Thanks for taking the time to comment!

Try a weekend free analysis.

Furthermore you should consider cross-correlation and auto-correlation — you’re dealing with timeseries data that is affected by a weekly cycle. Cross correlation and auto-correlation will show you lag effects.

Furthermore what you should do is try to model your happiness based on a these variables to better understand the factors.

You can also look at the difference or trend in happiness.