Bill Erickson Has a Spreadsheet with 3.2 Million Rows

Bill Erickson is a senior research specialist at the Yang-Tan Institute on Employment and Disability in the Cornell ILR School. He is also the lead researcher for the DisabilityStatistics.org website. To mark the site’s Feb. 26 relaunch, I interviewed Bill to learn about his role at this site and why people consider him a wizard when it comes to statistical measures of disability. This interview has been edited and condensed.

What is DisabilityStatistics.org?

The DisabilityStatistics.org website and its related Annual Disability Status Report provide easily digestible information to people who don’t necessarily know anything about disability or about statistics. It is digestible in ways that are useful for individuals, whether they’re writing proposals to justify funding needs or developing a business plan.

The goal is to make it easy for people to find the right information. And to provide charts and text that describe how to interpret the numbers. You can see the numbers, but not necessarily understand what they mean or how to use them – or know the implications of using this data point versus that data point. We can provide more clarity. —Bill Erickson

One person who contacted me wanted to offer hairdressing services to older people and people with disabilities. She had realized there were a lot of people who couldn’t easily access a hairdresser, so she was focusing her business plan around serving people with disabilities. She wanted to know how many people with disabilities were in her area and what kinds of disabilities they had.

[DisabilityStatistics.org is published by the Northeast ADA Center, a grant-funded project housed in the Yang-Tan Institute.]

How do you turn raw data from the Census Bureau into user-friendly output?

I work with Public Use Microdata from the Census Bureau – this is American Community Survey data. I connect my PC remotely to Cornell Center for Social Sciences services to analyze the three million observations we are working with. I do all my analysis and run my programs there.

Each year, I check if the survey questions and the variables are in alignment with the past year. Once I confirm this, I will make any needed changes to the programs. For example, St used to be the code for States, and it’s been that way for 20-plus years. But this year they decided to change it to the word state. Unfortunately, I was using state in some of my programs for my own variable, so I had to change my variable.

Once that’s done, the programs parse the data and split it out and develop all these estimates and funnel them into one massive Excel spreadsheet that’s almost 3.2 million rows of data.

Does Excel really do 3.2 million rows?

It doesn’t handle it very happily. But with enough memory, it can.

How does it get from Excel to DisabilityStatistics.org?

I hand it off to the web team [at the Yang-Tan Institute]. They do their magic to put all these numbers on the web site and into the annual reports. They get the report translated into Spanish and provide a Spanish version on the site.

Besides your obvious passion for statistics, what is your main personal interest?

My main hobby is cycling. When I go to my office at Cornell, I ride my bicycle. I have rain cycling gear and winter cycling gear. The ride is six miles long, and I enjoy the exercise and fresh air. I like not having to park a car. I’ve belonged to the Finger Lakes Cycling Club for many years. I help with their races, and I’ve organized some of their rides.

You occasionally write academic papers. Is there one you’d like to highlight?

One paper that I was the lead writer on was Disability-Inclusive Employer Practices and Hiring of Individuals with Disabilities. It looked at a study that we did about employer practices that increase the likelihood of hiring a person with a disability. [This peer-reviewed paper won the 2015 best research paper award from the American Rehabilitation Counseling Association, an organization that is currently known as NARRTC.]

Part of your job is technical assistance. What is that?

This is generally answering emails from people looking for help with statistics about disability.

Right now, I’ve got one looking for assistance for the number of people in a specific area of Kentucky in ages 18–24, but that’s not an age group that the Census Bureau provides. But I can do the calculations focusing on the specific population they’re interested in.

I also try to identify data sources or papers that people can go to if they have a disability statistics question that’s not looked at by the ACS [American Community Survey].

That’s my favorite part of my job. It’s something I can respond to and take care of. It’s fun to give people an answer right away, and people are genuinely extremely appreciative. I can provide background information that is important for them to understand. Not only just “here’s this number,” but this is what it means – and the caveats around it. Like how the ACS is not including people with mental illness.

Is there a misconception that you run into a lot?

When people look at estimates that combine disability prevalence and unemployment, they don’t realize that unemployment means something specific. The unemployment rate leaves out people who are not working, unless they are actively looking for work. For example, it leaves out discouraged workers – people who have given up looking for work. Many people think the unemployment rate is 100 minus the employment rate. And that’s just not the way it’s calculated. It doesn’t consider anyone who’s not actively looking for work.

But when you consider the employment rate, it’s important to know what ages you are talking about. If you are talking about anyone older than age 18, you are including a lot of college students and retired people. Or you could look at people ages 21–64 to include what’s generally considered the working age population.

What is a typical weekend day like for you?

Take the dog for a walk with my wife, and usually get out for a hike or a ride or a ski with friends. I’ve got this cohort of people that I’ve been riding with for 30-plus years.

What is your family like?

My wife and I have three kids. They are grown up. They are ages 22, 28 and 30. One of them is living at home right now. We have a 3-year-old rescue dog, Charlie.

We’ve been living in the Ithaca area for a long time. We raised our kids here. We enjoy the natural beauty of the area.

If you had 15 minutes with a policymaker, what would you most want to explain to them?

If they hadn’t already thought about it, the complications of how to define disability. Sure, there is the ADA definition. But there’s no data that matches that. We can ask survey questions about disability, but there isn’t one definition. There are also temporary disabilities – like people with broken legs. Another angle is that Veterans Affairs defines a service-connected disability in a way that doesn’t match the American Community Survey’s way of looking at disability.

Disability is complicated. And measuring disability is complicated.