Finding a New Home
Learning about neighborhoods far and near.
Finding a New Home eases home buyers into a new city, visualizing areas of interest and filtering on parameters such as crime rates and housing prices. The site is designed to help people explore neighborhoods beyond obvious choices and find hidden gems.
This application draws upon both academic papers and commercial endeavors that came before it. Exploration began in urban studies, with the goal of defining neighborhoods and understanding how people conceptualize the space around them. Jacobs’ work in understanding what makes a neighborhood showed that housing groups can vary in scale and strength of affiliation. By asking people how they visualize a city, Lynch determined that cities can be viewed as a combination of paths, edges, districts, nodes, and landmarks. Both of these perspectives helped in visualizing different aspects of neighborhoods
To determine which features were most important, I posted an initial survey on three home-buying websites to collect information about the experiences of recent home buyers. I asked the users whether they used the internet to find information about neighborhoods, what they wish they had known before they bought a home, and for basic demographic information. There were seven respondents in this initial survey. They ranged in age from under 24 to 45-65 years.
After compiling the feedback, four features were chosen for implementation: commute time, price, location, and crime. Although crime was not a deciding factor for survey respondents, it is a critical issue in urban areas. Crime is typically a non-negotiable characteristic, even if the rest of a user’s criteria are met. These features were chosen because they were mentioned multiple times by previous home buyers, and because they give a broad overview of the neighborhood.
- Easy to understand and use by novices.
- Powerful filtering to meet any user need.
- Always up-to-date records are relevant to the user.
When the user first loads the Finding a New Home application, they are presented with a map of the Atlanta area with different neighborhoods displayed in a neutral gray color. By modifying any of the variables listed on the right, the map will begin to change and the number of neighborhoods visible will decrease as the system gets a better idea about appropriate areas for the user.
Price can be entered as a minimum, a maximum, or both values. The input supports variation such as extra commas or decimal places. As the price changes, the map updates instantly to reflect the new range. The user can explore how their budget will affect their purchase decision, and determine whether they will be likely to find a property in a target neighborhood.
Crime information is only displayed when it is a problem in a neighborhood. The user can select violent crime, nonviolent crime, or both when filtering information. Warning symbols alert the user about which neighborhoods have the highest crime rates for the selected types of crime.
A functioning prototype was built with access to local neighborhood ammenities via the Yelp database and crime information from the City of Atlanta Public Records.
Observation of Performance
Users completed tasks about online home searches while speaking about their thoughts and actions out loud. These reactions were recorded for comparative analysis. Clarification was offered when needed. After the users completed the first task using existing websites, they were shown the new website and asked to complete those same tasks. Having the users complete the tasks using both systems allowed for a comparative study in both the interview and usability evaluation.
Users were asked to reflect on their experience with both the existing and new systems. Questions were meant to spark a discussion about the merits of each method.
The usability evaluation utilized NEC’s System Usability Scale (SUS). This document included ten Likert scale questions regarding aspects of the usability such as ease of use, integration of functions, consistency, familiarity, and integration of components. Participants were asked rate each factor from a scale of 1 to 5, with 1 meaning the user strongly disagreed with the statement and 5 meaning the user strongly agreed with the statement.