Construction Stimulus Funds: The Importance of Data Reporting
In 2012, stimulus funds from the American Recovery and Reinvestment Act (ARRA) are coming to a close. Since the construction industry was promised roughly $131 billion of those stimulus funds, I thought it’d be a good idea to check up on the effectiveness of the funding. I had a simple question: did construction stimulus funds help create (or retain) jobs?
Since I’m a Texas resident, and because Texas has a relatively strong construction economy, I decided to focus on jobs created or retained in Texas, specifically. Luckily, a website called Recovery.gov–the website devoted to helping the public track the flow of stimulus funds–aggregates a host of information for the public to access and analyze.
Unfortunately, the way the data is collected and distributed makes it next to impossible for the average person to answer my simple question. I’d like to share a few of the issues that I came across while trying to determine whether construction stimulus funds were a boon to Texas construction employment.
State-level Data is Not Detailed Enough
The first issue that I ran across was that state-level data was not detailed enough to come up with a good answer. For instance, while the total number of jobs created is reported by the fund recipient (either a state, city, agency or company), the type of job is not. So while we may know the number of jobs that were created, we have no idea if they were educations jobs, construction jobs or some other type of jobs.
As an example, the Texas Department of Transportation reported that they received roughly $2 billion in funds and created or retained 33,800 jobs, and surely some of them involved construction projects. However, since there is no coding for the type of job, there’s no way to tell how many of the 33,800 jobs were construction jobs.
Since state-level data didn’t have the needed information, I decided to access national statistics. Unfortunately, that information proved cumbersome to work with as well.
National Level Data is Too Difficult to Manipulate
After I did a bit more research, I noticed that national data has a North American Industry Classification System (NAICS) code–a code used to classify statistical data by industry–attached to each project. Since the construction industry is classified as 23, I could theoretically filter out Texas projects with code 23.
I tried this, but the spreadsheet of nationwide data is 320 megabytes and contains 587,150 rows of data. Because of this, every attempt to filter out this information crashed my computer. I was left with a set of state-level data that was incomplete and a set of nationwide data that was more complete, but one that I could never filter to produce a reasonable answer to my question.
A Lesson in Data Reporting
The difficulty that I had in finding an answer to this question actually teaches a good lesson: data in and of itself isn’t enough to answer your questions. In order to answer the important questions, it’s necessary to plan ahead for how you’ll classify and report data. Because, after all, data is meaningless if it’s not reported in the right way.
If Recovery.gov had better planned out what questions they’d like to answer, they likely would have required data to be reported in a more efficient and accessible manner. As a business, it’s important to think ahead and determine how your data will be used, what questions you’ll want answers, and how the data will have to be classified to derive those answers.
What are your thoughts? What are your best practices for data reporting? Stop by Software Advice to leave me your thoughts at: Did Construction Stimulus Funds Create Jobs? Good Question.
