Skip to main content

[Critique] Positional error in automated geocoding of residential addresses

Article by Michael R Cayo and Thomas O Talbot
Geographic Research and Analysis Section, Bureau of Environmental and Occupational
Epidemiology, New York State Department of Health, 547 River Street, Room 200,
Troy, NY 12180-2216, USA

The authors have sufficiently summarized the study by including the purpose, methods, results, and conclusions or summary. The study has identified positional error correction as their purpose, with the objective of determining if the errors in geographic encoding could impact results. The study focused on the discussion of the ability of GIS-based applications to locate residential addresses as closely as possible with the proposed alternative method of geographic encoding using residential property parcel data, instead of the automated geographic encoding.

The study emphasized on the application of GIS-based systems in public health, and how important precision in geographic encoding is to ensuring that public health studies are not compromised. There is extensive literature review on geographic encoding. The authors have noted that the underlying concepts and theories of this study have been documented in other studies, in that they were able to present the problem and justify the need for research. The study also indicates that there was limited public information on positional error correction, further emphasizing the need for research.

The problem statement has clearly identified what needs investigation, which emphasized on finding out if positional errors could impact public health data analysis results. As a solution, the authors have proposed their method of geographic encoding, which they expect to address these concerns.

The authors were able to specifically indicate their methods for collection of geographic data, beginning with residential addresses. This data is used as a starting point for geographic encoding and analysis. The authors indicated some exceptions, which according to them, will greatly increase positional errors in any geographic encoding method because of some missing details in the geographic data. The geographic encoding process was presented, together with the validation methods that were used in showing how much positional error resulted from the geographic encoding methods used. The authors were likewise specific in the analysis methods that they used for the study.

Although quite brief, the authors were able to present the results of their study well, using tables to compare positional errors between the geographic encoding methods used. The researchers were also very thorough in their discussion of their findings, pointing out the obvious improvements in positional error correction with their proposed method of geographic encoding. They also made suggestions as to how parcel points can be standardized, because the encoding and storage of this data is done by the local government and may not exist in similar formats.

The authors made recommendations for improvements, stating that their study is localized and may not be reflective of the level of positional error that the current geographic encoding method has, using topographically integrated geographic encoding and reference files.

Notes:
URL: http://www.ij-healthgeographics.com/content/2/1/10
Received 10 September 2003
Accepted 19 December 2003
Published 19 December 2003
© 2003 Cayo and Talbot; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.

Comments

Popular posts from this blog

Family Visits Singapura

My parents and in-laws traveled to Singapore last September and it was truly memorable! I'm trying out some code to display a Google Photos album and it's quite cool that I'm able to embed the album here. Credits to Pavex for the great tool. He details out how to use it here . Check out the album below!
March was a good month. Not just because it's my birthday month, but I've been able to do more physical exercises as far as I'm allowed, and I've also been able to get through my first month back in the job. For the first couple of weeks, I thought I was going to do 4 hours a day at most, and turn up the heat bit by bit until I'm able to do 100% again. But I suppose having to rest over a month does have its toll. It took significant effort to try not go all in with guns blazing on the first day back at work. LOL Anyway, I'm trying something new this year. We've migrated our applications to a better platform, and we'd like to push that to the cloud this year. It'll obviously need a lot of liaising and cooperative work with other teams and domains so I'm going to do more engagement and solutioning than the usual engineering this time. Last week, we did a platform-related hackathon and a critical refresh across clusters. There's going to be a lo

Get through your email faster with Gmail Priority Inbox

This is great for people who do not use a GTD system yet. Try it and see how fast it learns. Priority Inbox Beta   Welcome to Priority Inbox! By automatically separating out your most important messages, Priority Inbox makes it easy for you to read and respond to the messages that matter.   Get through your email faster Try reading and replying to the messages in the "Important and Unread" section first. Mark anything that requires follow-up with a star, then go through the "Everything Else" section. If you leave Priority Inbox, you can return to it by clicking the link next to Inbox on the side navigation of Gmail. How it works Gmail's servers look at several types of information to identify the email that's important to you, including who you email and chat with most, how often you email with these people, and which keywords appear frequently in the emails you read.