Social Media User Analysis

1.1 Overview

Nowadays, Social Media Users is everywhere.In 2018, roughly 2,234 Millions people are using Facebook, 980 Millions People are using Wechat, 813 Millions people are using Instagram, 330 Millions people are using Twitter and don’t forget, there are tons of Social Media Tools avaliable. As social media users, we are interested in investigating into different social media tools we are using now and what social media users worldwide look like. According to eMarketer, one in three people—2.48 billion—worldwide used a social network. Relativly Saturated social network use in Western Europe and North America are observed but significant rising social network use in emerging markets in Asia-Pacific, Latin America and the Middle East and Africa.

1.2 Comparison of Different Social Network User Numbers

Through graphing the total worldwide users of Facebook, Twitter, and Instagram, we can see that Facebook is dominating the world of social network: the worldwide user number of Facebook is much greater than that of Twitter and Instagram. In addition, by just looking at the graph, we can see that the growth rates of Facebook is also greater than that of the other two social medias.


1.3 Facebook Market Share

Facebook is dominating the world social media market. As you can see in the pie chart below, facebook is dominating the social media market with 62.2% market share as of 2017. All the other social medias possess the rest of 37.8% market share.


1.4 Penatration by Country

Here, let’s have a look of this fancy waffle graph. Different colors here represent different continents. Around the circle, there are labels of different countries. This graph is showing the ratios of social media user numbers over total population. As you can see in the graph, the larger the shape, or, the wider the position, the greater the ratio. Some Asia-Pacific countries have very big ratios while other Asia-Pacific countries have small ratios. Similarly, some European countrie have very big ratio while other European countries have small ratios. For other continents such as Latin America and North America, the difference of ratio also exists.

1.5 Leading Countries with the Most Numbers of Facebook, Twitter, Instagram, and LinkedIn Users

The four pie charts here show the leading 5 countries with the most number of users of Facebook, Twitter, Instagram, and LinkedIn. The upper left pie chart shows that the top 5 countries with most Facebook users are India, USA, Brazil, Indonesia, and Mexico. The upper right pie chart shows that the top 5 countries with most Twitter users are USA, India, Indonesia, Japan, and China.The bottom left pie chart shows that the top 5 countries with the most Instagram users are USA, Brazil, Indonesia, India, and Turkey. Finally, the bottom roght pie chart shows that the top 5 countries with most LinkedIn Users are USA, India, Brazil, UK, and China. We want to explore the reasons behind the difference of number of users in different countries. The following section will tell us what factors lead to the differences.

This map shows the five leading countries with the most Facebook Users. The more purple in colour indicate the greater users, the more pink in colour indicate the fewer users. The countries painted in yellow are out of the top ranking.

2.1 Internet Penetration

First, I choose the social media users (% of population) intex and plot it on the worl map. we could conclude that the more darker color a country has, the larger proportion of social network users of this country. We could see that the distribution of social medai users focus on America and Europen.

## OGR data source with driver: ESRI Shapefile 
## Source: "/Users/HeRen/RH/My Documents/Master/QMSS/Data Visualization/DV_CU_course_material/Lectures/Week07/data/world_map", layer: "TM_WORLD_BORDERS_SIMPL-0.3"
## with 246 features
## It has 11 fields
## Integer64 fields read as strings:  POP2005

### 2.2 Freedom of Press Second, as the first plot shows, there is some consistency between proportion of social network users and counties’ distribution. I used color scale to plot the freedom of press Index. As we could see, forexample, Canada have a high level of freedom of press as well as high social network users percentage.

## OGR data source with driver: ESRI Shapefile 
## Source: "/Users/HeRen/RH/My Documents/Master/QMSS/Data Visualization/DV_CU_course_material/Lectures/Week07/data/world_map", layer: "TM_WORLD_BORDERS_SIMPL-0.3"
## with 246 features
## It has 11 fields
## Integer64 fields read as strings:  POP2005

### 2.3 Third, we would like to visualize the relations between Internet accessibility and Social Media Users percentage. So we plot these two information into one interactive map. And we also cold conclude some consistency between percentage of social media users and Individuals using the Internet Index.

## OGR data source with driver: ESRI Shapefile 
## Source: "/Users/HeRen/RH/My Documents/Master/QMSS/Data Visualization/DV_CU_course_material/Lectures/Week07/data/world_map", layer: "TM_WORLD_BORDERS_SIMPL-0.3"
## with 246 features
## It has 11 fields
## Integer64 fields read as strings:  POP2005

3.1 Age

There are some interesting findings about the relationship between age of Facebook users and the number of Facebook users.

## # A tibble: 24 x 5
## # Groups:   Age, Country [24]
##        Age Users Country perc_population percentage
##  *  <fctr> <dbl>  <fctr>           <dbl>     <fctr>
##  1    0-11   0.4  Canada            0.09      9.30%
##  2   12-17   2.0  Canada            0.85     85.00%
##  3  18-24    3.1  Canada            0.94     94.00%
##  4   25-34   4.4  Canada            0.87     87.40%
##  5   35-44   3.8  Canada            0.77     77.40%
##  6   45-54   3.2  Canada            0.63     63.00%
##  7   55-64   2.6  Canada            0.52     51.70%
##  8     65+   1.7  Canada            0.27     27.10%
##  9    0-11   1.7   China            0.72     71.60%
## 10   12-17  65.0   China            0.89     89.10%
## # ... with 14 more rows


4.1 MeetUp.com Findings

By analyzing MeetUp.com data, we have found the most popular topics by counting the number of users in each MeetUp category created in MeetUp.com. ‘New Technology’, ‘Art’, ‘Watching Movies’, ‘Wine’, ‘Photography’ are the Top 5 most popular topics people in MeetUp.com like to have activities on over time.

Interesting ! Do these topics vary across city ? As New Yorkers, we would like to know what are the top topics in New York. Therefore, we selected data from three different cities: New York, Chicago and San Francisco. We found that these three countries are sharing most of the top topics such as ‘Tech’,‘Career/Business’,‘Socializing’, but it seems that New Yorkers are lacking of passion on ‘Sports/Recreation’ and Chicago people are lacking of passion on ‘Spirituality’.


4.2 LinkedIn Findings

As Social Media Become more and more popular, people started using social medias for job huntings. LinkedIn is one of the most popular social network platform for job seekers. The effect of linkedIn widely spreads to many industries. People with different majors and skills present their educational and working experiences as well as skills on LinkedIn and utilize LinkedIn to find jobs that are related to their skills.

Here we did some data analyses and visualizations on the skills that appear on LinkedIn profile most frequently and that are the most popular skills which companies are looking for.

As shown in the above point graph, the top 1 skills in countries listed here are either Statistical Analyses and Data Mining or Cloud and Distributed Computing. In the US, the top 1 skill is Cloud and Distributed Computing. The skill trend shows us that Statistical Analyses and Data Mining, and Cloud and Distributed Computing are the 2 most demanding skills nowadays.

The USA skills graph above lists out the top five most in-demand skills that appeared on linkedIn profiles. As we can see, Cloud and Distributed Computing, Statistical Analyses and Data Mining, Mobile Development, Storage System and Management, User Interface Design are the top 1 to top 5 skills that appeared on LinkedIn profiles most frequently in the USA, 2016.

4.3 Twitter Findings

In Twitter self-descriptions, there are some words that female users use a lot and there are some words that male users use a lot. A Twitter Gender Identifier was designed for identifying genders through looking at users’ self-descriptions. The mechanism behind this Twitter Gender Identifier is shown in the wordcloud: when encountering these high-frequency words, the identifier is able to identify whether the user is a female or a male. By looking at the wordcloud, we can see that the most frequently used words by female users are beauty, mother, mum, happy, and queen, while the most frequently used words by male users are technology, business, director, producer, and football.