CIT 115: Data Project

Website created by Adam Gutonski

Comparing the Number of Driver-added Stickers Automobiles in Two Separate Parking Lots

	For our study of basic data gathering and manipulation, we decided to observe samples of automobiles in a CCAC parking lot 
and a public grocery store parking lot in Pittsburgh Pennsylvania.  Specifically, we counted the number of driver-added stickers on each automobile.  
In total we made 30 vehicle observations per parking lot.  We counted every fourth car, starting from the first parking space at the entrance of each lot.
	Our definition of “driver-added sticker” counted any sticker placed on the car.  Position did not matter, meaning that the sticker could be placed 
anywhere on the body of the vehicle.  We excluded any factory, emissions, registration, any legally required stickers/tags, as well as any dealership branding.  
We wanted to capture the number of times that a car owner placed a sticker on their vehicle that included any imagery, business promotion, school name or 
parking pass, text, political and religious identifiers, etc.
	Our data categories are the “car observation number”, “make”, “model”, “category”, and “number of stickers”.  To simplify our data, we refined the 
make/model classes to four specific body type categories.  The make/model data allowed us to identify body styles via Google searches.  These included a “Truck” 
category, which represents any vehicle with an open flatbed (covered or otherwise) and tailgate door.  The “SUV” category represents any sport utility vehicle.  
The “Sedan” category represents any other four door car or hatchback, and the “Coupe” category represents any two-door vehicles.  The first two data tables represent an 
individual parking lot and show the total number of driver-added stickers per car, the average stickers per car, least/most amount of stickers on a given category, 
the total stickers per category, and the standard deviation.  The third data table contains a scatter plot of the data and shows the data correlation value of ~0.080.  
	The first question posed was ‘Will a college campus have more bumper stickers per car than a public shopping space?’ We expected that a college campus would 
have more total stickers, but the correlation value tells us that there is no relation between the amount of driver-added stickers at a college parking lot and a 
grocery store.  Next we wondered, ‘Of the four body type categories, will a particular body style have more or less stickers?’Our hypothesis was that the 
“Truck” and “SUV” class will have more driver added stickers and the “Coupe” class will have the least.  The =max function shows us that in the CCAC lot, the vehicle
with the most driver-added stickers was a Ford 4x4 in the “Truck” category with a total of 21.  The =max also shows a “Truck” style Dodge Ram in the Pittsburgh lot, 
with the most driver-added stickers at a total of 8.  
	Reflecting on our study, we see that there was opportunity to collect more insightful and revealing data.  
This would also require a greater sample size than what we used.  With a greater sample size, we believe that there would be a substantial 
negative correlation between the CCAC parking lot sticker number and the Pittsburgh lot.  Meaning, as the number increases for driver-added stickers 
at the Pittsburgh lot, the number of stickers will decrease in the CCAC lot.  We would want to create a field for tracking types/styles of driver-added 
stickers to see if there was any regional, political, religious, or other type of driver-added sticker that could explain the negative correlation 
(if it in fact was negative).

Thanks for visiting!