Twitter and Obesity
Background
Data
What is obesity?
Obesity is a medical condition in which abnormal or excessive fat accumulation presents a risk to health.
Obesity is preventable
By exercising regularly and eating healthy on a consistent basis, you can prevent unhealthy weight gains and related health problems.
A crude way of measuring obesity within a population is the body mass index (BMI): a person's weight in kilograms divided by their height in metres squared.
Obesity increases the risk factors for various chronic disease including:
- Cancer
- Osteoarthritis
- Diabetes
- Heart disease
- Obstructive sleep apnea
It can also lead to chronic illnesses such as:
- Eating disorders
- Poor self-esteem
- Depression
The data for our project was acquired from our client Michael Martin, who is currently a PhD student working with Dr. Nadine Schuurman at Simon Fraser University. Michael’s Twitter database is comprised of an algorithm that has been scraping a random one percent of all tweets in North America since 2013.
His database contains over 700 million randomized tweets, including each tweets username, time stamp, geo-tagged location, and absolute location. From these tweets, we only worked with data that originated from the City of Vancouver.
This database represents a considerable amount of private and sensitive information, if revealed to the public it would compromise not only the integrity of our study but
also the private lives of the Twitter users involved.
Steps were taken to ensure that no one Twitter user was singled out in the duration of the study and that our outputs and results did not reveal any sensitive information that could potentially be used against an individual or reveal their private lives.
Unhealthy Tweets
Healthy Tweets
Health Survey
The other source of data used in this study was a health survey conducted by Dr. Scott Lear, a professor in the Faculty of Health Sciences at Simon Fraser University. This survey was conducted in three cities including: Vancouver, BC, Hamilton, ON, and Montreal, QC. This survey contained information of each city’s population and information related to their health, including height, weight, BMI, muscle mass index, etc. Ultimately, the majority of this survey was not used in this project because of the inability to discern the meaning of the data fields. The only statistic we included in our project was the BMI measurement of survey respondents.
