The Effect Of Sleep On Happiness [Complete Analysis] - Happiness Through Sleep: Part 1


It is widely known that sleep influences our happiness. Many studies have shown that a continued lack of sleep (sleep deprivation) has negative results on not only the ability to be happy but also the immune system, brain functioning and blood pressure levels. It cannot be argued that sleep is one of the most important factors in our lives. If we don't sleep well, we are likely unable to function properly.

Yet, a lot of people don't pay attention to their sleep habits.

In March 2015, I took the decision to focus more on what my sleep habits were. I started tracking my sleep. Since then, I have recorded nearly 1.000 days of sleep.

I want to show you exactly what sleep does for me, and how it influences my happiness.

What am I looking to find out?

As usual, there's a couple of things that I would like to find out for myself. The most important question I want to be answered is:

  • Is there a positive correlation between my sleep and happiness? Let me rephrase that: Am I happier when I have more sleep?
  • In addition, I want to find out how much sleep I need in order to maintain my happiness. What minimum level of sleep do I require before it starts to affect me?

Tracking my sleep?

This site is all about tracking happiness. I track my happiness and want to inspire others to do the same by showing the benefits and results that I've gathered over the years.

In addition to tracking my happiness, I have also been tracking my sleep. This is a bit different than tracking my happiness.

There are multiple methods a person can use to track their sleep. I know of people that do it by hand, in a bullet journal or a simple notebook. I myself like to do things digitally. Therefore, I have been using an app on my smartphone for sleep tracking.

This app - Sleep as Android - is great. There are multiple apps out there that can track sleep, but I have not come across one with the ease of use and great features this has.

This app starts measuring my sleep once I turn it on every night. It not only tracks the start and end time but also tracks the movement and sounds of my (mis)adventures in dreamland. You can only imagine what kind of data this results in! I have only used a part of this data in this first analysis. I'll get to the data later.

When did I start tracking my sleep?

At the start of 2015, I spent a 5 week period working on a huge project in Kuwait. It was a very challenging period for me, and my happiness ratings were quite low at the time. I experienced one of my worst days ever during this time.

"5 weeks? That's NOTHING!".

I won't blame you if this thought came to your mind. 5 weeks isn't really that long of a period. Yet, I still managed to get completely burned out at work due to a complete lack of sleep.

You see, I worked about 80 hours a week. After 12 hour days on the project, I felt like I still wanted to do things I actually liked and enjoyed. So instead of going to bed at a decent time, I watched movies, exercised and Skyped with my girlfriend until late in the night. Even though my alarm went at 6:00 AM every single morning, I rarely went to bed before midnight. I was living on about 5 hours of sleep per day, while continuously working LONG days.

Why I started tracking my sleep?

These 5 short weeks lasted a lifetime. It was a difficult period, purely because I completely mismanaged my sleep duration on a daily basis. This period would have been a lot easier had I focused more on my sleep.

So I decided to do just that. I wanted to know more about my time spent in dreamland.

I also knew I was going to spend more time on challenging projects abroad in the future, so I wanted to be fully prepared when the time came.

What data did I collect?

I started to sleep with my smartphone next to my pillow, constantly collecting data about my sleep habits. So after having tracked my happiness before going to bed, I would turn on this app, and let it run in the background.  Sleep as Android collected all my sounds and movements, which were simultaneously backed up to the cloud for future reference. After waking up the next morning, I stopped the app from tracking and rated the way I was feeling. Easy stuff!

The data collected by my sleep tracking app

This obviously results in a lot of data, which is extremely interesting to analyse. However, I will only use the start and end time of my sleep for this first part in the "Happiness through sleep" series. No matter what this analysis determines, there will be lots of additional possibilities for me to further analyse this set of data!

Let's not waste any more time on this intro, and look at the shiny data this app has collected for me.

Processing sleep data

I am only interested in my daily amount of sleep for now. This is quite easy for me to calculate, as the application can export every recorded sequence of sleep to a single file. The only thing left for me to do now is to sum the duration of all sequences per day. It is possible that a single day contains more than one sleep sequence (think of a power nap).

An important detail here is that I've counted the duration based on the end date of the sleep sequence. Say, I slept from 23:00 on Friday, until 6:00 on Saturday, then the total duration of 7 hours will be counted for Saturday.

Daily amount of sleep

Before showing you the complete set of durations, I first want to zoom in on a smaller interval. The chart below shows the daily sleep durations for the months November and December 2016.

There are a few things I want to highlight here. It's immediately clear to me that I sleep below average on weekdays (Monday to Friday) and above average during the weekends (Saturday and Sunday).

Also, the average amount of sleep within this interval is 7,31 hours. According to the National Sleep Foundation, that is an acceptable amount for the majority of the adult population

Now, I'm going to make a huge assumption here. I'm assuming that my average sleep duration is equal to my minimum required sleep.

Yes, let that sink in.

I make that bold assumption based on the following lines of thinking: I have been a functioning human being, and have lived a happy life so far. I've experienced my fair share of sleep-deprived days, in which my happiness was definitely influenced (my period in Kuwait springs to mind). However, I have always recovered from those periods by catching up on sleep. This is included in the average sleep duration.

You could say that I might be sleeping too much and that I can still be a functioning and happy human being on less sleep. To that I say: you might be right, and that I simply don't know. It's one of the things I want to determine by analysing this entire set of data. I want to find out what minimum level of sleep I require before it starts to affect me.

Anyway, based on the prior assumption of required sleep duration = average sleep duration, I am now able to calculate my sleep deprivation.

Daily sleep deprivation

According to Wikipedia, sleep deprivation is the condition of not having enough sleep. I can calculate my daily sleep deprivation by subtracting my daily sleep duration from my required sleep. This sleep deprivation is visualised in the chart below.

It's important to point out that a positive value in this chart is actually a good thing. The chart shows a positive value if I slept longer than required, and a negative value when I'm sleep deprived.

I've added the cumulative sleep deprivation and charted it on the right axis. This shows you exactly what my sleep habits are. I tend to not sleep enough during weekdays, from which I need to recover on the weekdays.

This matches my suspicion: I highly value my sleep on the weekends. Waking up early gets harder as the week progresses, and I'm usually pretty tired on Fridays. My sleep habits would certainly not win any awards for Best Value or Most Durable. No way.

You know now that my sleeping habits are not optimal, and I'm very much aware of it. By shifting my sleeping times like this, I am constantly living on a jet lag. This is called a social jet lag. This is certainly something I should be trying to optimise.

One more thing I want to highlight before showing you my full set of data is that the cumulative sleep deprivation ends exactly on zero. This is a result of my big assumption, that my required sleep duration equals my average sleep duration.

The full set of data

Let's have a look at the total set of data. This includes all the days in which I've tracked my sleep. This started on the 17th of March, 2015. The chart below contains a range of approximately 1.000 days, so you might want to scroll to the right to see the entire thing 🙂

Except for a couple of periods, I have been living with a social jetlag for the entire duration of this analysis. The pattern is mostly the same: sleep deprivation during the weekdays, and recovery during the weekends.

There are also gaps in this data! *gasps for air*

How can an article about tracking sleep - posted on a site about tracking happiness - have gaps in the data?!!

There are a couple of reasons for that, of which one is that I simply forgot to start this sleep tracking application before going to sleep on some days. No excuses there! This results in the small, single day gaps you see in the data. What caused the bigger gaps in this data set were my holidays. During some of these holidays, I was sleeping in a tent without the possibility to simultaneously charge my smartphone and track my sleep. I personally think that's a good enough reason, so I would appreciate it if you could forgive me for these errors.

These gaps are discounted in this analysis, meaning that they do not influence the outcome of this exercise.

The average sleep duration on which I have survived and functioned just fine so far is 7,16 hours per day.

Let's see how this translates into my sleep deprivation calculation!

As you can see, the cumulative sleep deprivation varies quite a lot. The periods with the steepest increases and decreases in cumulative sleep deprivation deserve some additional context.

For example, have a look at the Christmas period of 2015, starting on the 20th of December. At the time, I had a 10-day streak of great nights of sleep, lasting until the 31st of December. This was the result of the holiday period, during which I rapidly increased my sleep buffer!

Another example is a streak of sleep-deprived days, starting on the 3rd of July, 2017. This was actually the start of a very busy period at work, from which I only fully recovered two months later during my holiday to Norway.

For more details on the different things that were going on in my life during this entire period, you can check out my 'Personal Happiness' section. It covers every single month of my life.

Sleep duration per day

You might be interested to see a quick visualization of my average sleep duration per day.

It's safe to say that there's some room for improvement here. As of right now, I rely on every single weekend in order to catch up on lost sleep. It'd be much better if I could manage to evenly distribute my sleep, without depending on a particular day of the week.

Some disturbing notes about this data

I must confess something. This data is nowhere near 100% accurate, and it would be naive to think otherwise. Allow me to explain.

For example, the 21st of May, 2015 looks to have been a terrible night for me. If you have a look at the chart, you will see that I had a sleep deprivation of 5,73 hours that night! Only 1,43 hours of sleep? What the hell happened there? Well, I was actually traveling to Costa Rica on that day. Therefore, I not only faced a huge jetlag and difference in time zones, I also didn't activate my sleep tracking app while dozing off in my seat on my long flight.

Coincidentally, the 7th of April, 2016 has the exact same issue. On that day, I was flying back to the Netherlands, from a second visit to the same project in Costa Rica.

I also need to point out that my data is inaccurate due to yet another reason. That reason is: I don't instantly fall asleep the moment I press start on my sleep tracking app. If only that was a possibility, right?!

I fall asleep quite easily. It usually doesn't take me more than 30 minutes. I can confidently say that, because I always sleep with music on, and I set my MP3 player to shut down after 30 minutes of inactivity. 99% of the times, I don't notice it when the music stops, meaning I'm already flying with dragons, exploring beautiful forests and fighting off villains in my imaginary dreamworld!

A number of sleep sequences, highlighting the durations at the start of my sleep laying "Idle"

On rare occasions, however, I find it extremely hard to fall asleep. It has happened on multiple occasions that I hit the sheets at 22:30, after which I have a staring contest with the ceiling until the clock passes 03:00. Even though it doesn't happen often, it totally sucks when it does. I have since learned that this usually happens after I go out to an all-you-can-eat dinner. I'm not kidding. Eating too much is causing me sleep insomnia...

These "idle" times - a.k.a. the moments when my app is measuring my sleep but I'm actually still awake - are somewhat distorting this data analysis. I can only hope this doesn't ruin my data beyond any use. We'll have to see about that!

Happiness and sleep

In addition to tracking my sleep data, I've also been tracking my happiness. If I want to determine whether or not my happiness is influenced by my sleep, I'm going to have to combine these two sets of data.

My happiness tracking data consists of two important variables: my happiness ratings and my happiness factors. By now, I'm assuming you know what these things are. If not, then that's okay! Just head on over to my method page to learn more about these crucial variables. 🙂

My happiness ratings

The chart below shows you the same set of data as before but now includes the happiness ratings as well. Please notice that these ratings are charted on the right axis.

So this chart shows you 3 things: my daily sleep deprivation, my cumulative sleep deprivation and my happiness ratings. I have tried to include some comments here and there. It's my attempt to provide extra information to this chart, since it's rather hard to read as it is.

Can you determine whether or not I'm happier during the days where I have slept like a baby?

I didn't think so.

You should be able to see big dips in my happiness ratings. These were never caused by a lack of sleep though. Similarly, my happiest days were not caused by an abundance of sleep. It's impossible to determine any correlation at all based on this graph. I know my happiness is influenced by a lot of factors, but so far I cannot tell if sleep is one of them.

Again, you can head over to 'My Personal Happiness' section for more details on my general life. I try to accurately describe what moves my happiness in these posts, so you might be interested in them. 🙂