Determinants of health, dimensions of health, positive health and spectrum of...
COMMON PROBLEMS IN ANALYZING DATA OBTAINED FROM FACEBOOK
1. Common problems in analyzing data
obtained from Facebook
EDRD 6000
By Grace Kargobai
University of Guelph, 2017
2. Overview of the presentation
● Brief background of Facebook
● What is Data Analysis ?
● Tools used in analysing
facebook data
● Highlights of Common problems
● Example : U.S 2016 election :
Real news vs Fake News on
facebook
● References
3. Background of facebook
● An American for-profit
corporation and online social
media and social networking
service based in Menlo Park,
California.
● Facebook is owned by Mark
Zuckerberg and was launched on
February 4, 2004.
● Facebook has 1.86 billion monthly
active users which is 17 percent
increase year over(Facebook MAUs)
4. What is data analysis + Facebook data
● Data Analysis
This is the process of systematically applying
statistical and/or logical techniques to
describe and illustrate, condense and recap,
and evaluate data.
● Facebook Data
The categories of data being processed by
Facebook and the personal data that Facebook
receives from Facebook members.
5. Tools used in analysing facebook data
● Facebook Insights
It provides detailed metrics about your posts and
the engagement they earn. Audience analysis,
including demographic and location breakdown, can
help you better understand your fans.
● LikeAlyzer
A simple, free-to-use tool that allows you enter any
Facebook page without authorization to measure and
analyse its performance.
● Agoropulse
The main suite is a management and engagement tool
for several social media accounts, including
Facebook
6. ...tools used in analysing data
● Simply Measured
Simply Measured has four different free Facebook reports
which offer some real insights.First is the Insights report, which
repurposes Facebook Insights data into some great looking
graphs. Information includes engagement, reach, stories, post
type, impressions, page likes, traffic sources, fan statistics,
activity, and demographics.
● Sociograph.io
It shows the total number of posts, authors, commenters and
likers. It also displays the average number of likes, shares and
comments per post, the different post types, and the top posts
in the chosen timeframe.
7. Highlights of Common problems
● Redundancy
Always difficult to limit the amount of times
people answer questions or surveys on facebook
computer IPs are written down
● Selection
Sampling and selection error by non probability
method
● Using the wrong parameters for statistical
significance
8. ...common problems
● Privacy Settings
Most people do not share their
information publicly, thus making it more
challenging for researchers to analyze
data and get a holistic picture
● Timeliness
If the data is old or stale then the
information derived from the analysis
could be of no use.
9. Example : abstracts from a write up the ceo of facebook released about real news vs fake
news on facebook
● All the content on facebook, more
than 99% of what people see is
authentic.
● We have already launched work
enabling our community to flag
hoaxes and fake news, and there is
more we can do here.
● Identifying the “truth” is
complicated.
10. REferences
Top 20 Facebook Statistics - Updated January 2017
https://zephoria.com/top-15-valuable-facebook-statistics/
Facebook's Mark Zuckerberg wrote a 6,000-word letter addressing fake news and saving the world
Kara Wagner - http://www.recode.net/2017/2/16/14632726/mark-zuckerberg-facebook-manifesto-fake-news-terrorism
Lin KY. Lu HP. Intention to continue using Facebook fan pages from the perspective of social capital theory. Cyberpsychology and
Behavior Social Network. 2011;14:565–570. [PubMed]
Zimmer M. “But the data is already public”: on the ethics of research in Facebook. Ethics and Information Technology.
2010;12:313–325.
8 Free Facebook Analytics Tools (and 7 more!)
https://www.brandwatch.com/blog/8-free-facebook-analytics-tools/
Papacharissi Z. The virtual geographies of social networks: a comparative analysis of Facebook, LinkedIn and ASmallWorld. New
Media Society. 2009;11:199–220.