My sleeping patterns

Surprisingly, I started to prepare for this week's assignment a little bit earlier. While working on the group's project (where we are using datasets retrieved from quantifiedself.com) I decided to try self-tracking tools on myself in order to come up with ideas on what can be interesting to study. Hence, I downloaded a bunch of apps that can track my personal data. One of them – SleepBot – interested me the most since my relationships with Morpheus have always been very intriguing. So, on 9th of October this application started tracking my sleep patterns.

It took quite some time to extract and clean up the data from my phone. Not going deep into the details, it should be noted that finally I had a dataset consisting of information about my sleep patterns from 9th till 21st of October 2014.

On the first table you can see the duration of my sleep in particular day. We can see that generally I was sleeping more than eight hours a day.





Next table depicts the details, for example, the time when I fell asleep and woke up:




Application gives to its users an option to rate each sleep (from 1 to 5). I was interested in looking closer on how did I define the quality of my sleeps those days, so next bar chart shows both the duration of the sleep and the grade I gave to it. We can see that the lowest grades were given when the duration of the sleep was less than eight hours.

This is another way of visualizing the same data:

Moving to the last graph, we can see that the highest grade - 5 - was given to the sleep when its duration was less than 10 hours, while naps longer than 10 hours (wow) got 4, and, finally, naps continuing less than 8 hours got the lowest grades.





In order to sum up, I should note that this exercise was very useful for me. Actually, I have never really been focusing on the fact that I’m oversleeping that much.

P.S. SleepBot records users during the night, which is really creepy, isn’t it?

My forever-politicised newsfeed

Reading posts of my groupmates who were failing to find any political discussions in their news feeds, I already knew that my Facebook would show radically different picture. Considering the fact that I’m from Ukraine, country that recently has gone through revolution and now going through war, it’s not a surprise that my news feed is full of politics.

For this assignment I decided to count all the posts from my news feed on one particular day and find how many of them are political. Here are the results:

As we can see from the pie chart, 66 out of 157 posts were political (42%), while 91 (58%) were connected to other topics. To the group of ‘political posts’ I included: news disseminated by mass media, posts from different pages (e.g. Facebook page of Ukrainian Ministry of Defence) and status updates of my friends/people that I’m following  - all of which were touching political issues as well as starting political discussions.

Frankly speaking, I was sure that the majority of posts in my newsfeed are connected to the political situation in Ukraine, so the results that showed relatively low amount of such information (42%) surprised me. Probably this gap between my expectations and reality happened because I generally do not focus my attention on non-political news (except the updates from my close friends)? Or maybe the character of my newsfeed has shifted just recently - after I moved to the Netherlands and added a lot of new people, who are not interested in Ukrainian affairs, to my friendlist?

Turning to the content of political posts, I found out that they hardly confront my political views. It is not surprising: as long as we can choose which mass media, Facebook group or person we want to follow, we will get expected result in our newsfeed. The only exception here – the content provided by our friends. Fortunately or not, but I cannot show such examples as political posts from my friends were not contradictory to my own opinions.

Even though my newsfeed is not so politicized as I expected, I believe that it still contains much more political messages than newsfeeds of many of my classmates. In this context I should say that this informational shift on my Facebook happened suddenly. My newsflow in social networks also used to be full of ‘funny’ pictures, lifestyle-articles and so on. Of course there were a few voices of political activists that I followed in that flow, but generally Facebook was not a platform from which I gained political information, as it is now. Obviously, my newsfeeds in social networks became so politicized after the beginning of the revolution in Ukraine and for so far it is not likely that they will ever be light and entertaining again.

P.S. This is the first political post on my page in Instagram. On the foto you can see a small picture that policemen gave to me during the revolution in Kyiv. Date: January 2014

The island of freedom: What emotions do Sziget’s guests spread through Twitter?

Sziget, which is held every year on one of the Budapest’s islands, is one of the largest music festivals in Europe. Since 1993 people from different parts of Europe visit this event for good music and unforgettable atmosphere of freedom. Is this festival still giving positive emotions, or it has become just a well-known spot?

To reveal this I decided to analyze recent tweets with the hashtag ‘sziget’. Which emotions do people feel while writing about the festival?

To begin with, we should stress out that positive emotions in tweets prevail negative:

The next bar chart depicts which emotions (positive/negative) do citizens of different countries share with their friends more/less frequently. Generally, we can see that people from the UK and Hungary write the biggest amount of negative-emotion tweets. However, citizens of the UK also also show the highest figure of positive emotions in their tweets. This probably means that users from this country were presented better in the Twitter on the day when data had been retrieved. To back up this statement, we can look at my previous research, which clearly shows that the biggest amount of tweets was from the UK.

If we want to look closer at links between pos/neg emotions in tweets and users’ countries of origin, we can turn to the next two bar charts that show more detailed picture: 

Positive emotions - Countries

Negative emotions - Countries

I  was interested about the content of ‘negative’ tweets. As the result, the next chart shows text of tweets with negative emotions. We can see that verb ‘miss’ is the most widespread within them, which means that even the so-called ‘negative’ tweets concerning sziget are not negative about the festival.

Turning to the next chart, we can see that tweets with positive emotions were retweeted more frequently.

The last table shows the link between pos/neg tweets and their source. We can see that the highest number of positive posts is from Tumblr, while the highest number of negative – from both Twitter for iPhone and Twitter Web Client.

To conclude, we can say that generally people spread positive emotions with the hashtag ‘sziget’. Furthermore, these ‘positive’ tweets are retweeted more frequently. However, if we turn to the content of ‘negative’ tweets, we can see that they are not negative about the festival at all. Backing up on the countries of users’ origin, we can claim that the highest number of positive emotions are spread by the UK’s citizens, while the highest number of negative were spread by people from both UK and Hungary.