Foodborne Chicago Affected by Twitter Geocoding Issue

Foodborne Chicago is an application that provides an easy online interface to report incidents of food poisoning to the Chicago Department of Public Health through the city’s open311 interface. This allows residents to fill out a simple online form to report an incident instead of calling 311 directly.

Regardless of the mode of reporting, the reports end up at the Health Department, where they are reviewed. If a given case meets the Health Department’s criteria, a CDPH inspector is sent to the restaurant to do an inspection. In many cases, nothing is found, but in some cases, serious violations are found and fixed.

The second part of the application uses the Twitter API to proactively find folks complaining of food poisoning in Chicago and send them a link to the form to fill out a report. The Twitter API takes a bounding box (the Chicago area) and a search term (food poisoning) and returns those tweets. Since the end of November 2014, the number of tweets that the API returns have gone down drastically. Here’s a graph of the number of daily tweets since 2014:

total-number-of-tweets-as-of-april-2015

Foodborne Chicago was not the only application to be affected by this issue.

There was an issue logged on the twittercommunity.com message board. An engineer from Twitter replied that there are two issues affecting the number of tweets returned. One is low coverage of tweets with explicit geocodes attached and the second issue is the lack of fallback to the user profile for location approximation. At first the engineers gave an estimate of days to get the issue resolved. The final update has a start date for work on the fix no earlier than April 1, 2015. Hopefully, a team has been assembled and is working on it now.

The Foodborne Chicago application continues to see an increase in the number of submission via the web form, but this Twitter issue has affected our ability to be proactive and respond to those directly affected with food poisoning. This is a graph of the number of submissions per week through the Foodborne Chicago web form since January 2014:

submissions-as-of-april-2015

This Twitter API issue hasn’t prevented us from being effective in Chicago, but it’s unclear how many more submissions and cases of food poisoning would have been caught with a proactive approach that is enabled using the Twitter API. Twitter has been a great tool for us, and we hope they can fix this issue soon.

Chicago Health Corps Turns 20! By Alicia Phillippou

DSC00219Editor’s note: The following post is from Alicia Phillippou – a member of the Chicago Health Corps

What do Habitat for Humanity, City Year, and the Smart Chicago Collaborative all have in common? They are all programs that benefit from the help of AmeriCorps members! This year marks the 20th anniversary of AmeriCorps, the National Service Organization which provides year-long service opportunities to people of all ages interested in making a difference. The Chicago Health Corps (CHC), an AmeriCorps program that has been part of the organization since its inception, is also celebrating its 20th Anniversary.

In honor of this, the CHC and its members have been spreading the word about all the work that members participate in throughout the Chicagoland area. Chicago Health Corps members serve at Erie Family Health Centers educating patients about diseases and disease prevention. There are also members serving with Chicago Public Schools working to educate students of all ages about health issues ranging from dating and sexual health to proper dental and eye care. There are even members serving as two of the Smart Chicago Collaborative’s many Health Navigators at the General Medicine Clinic at Fantus Health Center. Feel free to take a look at what CHC members have gained so far from their year of service!

Given the CHC and Smart Chicago’s shared dedication to the betterment of human lives within Chicago, it was only natural for a partnership to develop. As the two CHC members serving at Fantus Smart Health Site, we engage patients in activities that promote access to care and increasing health and technological literacy skills. While these are lofty goals on a daily basis our work is much more straightforward. We inform and assist our patients with gaining access to the patient portal, a secure website that gives patients’ access to their Electronic Health Records (EHR). This allows patients to see upcoming appointments, review summaries of past visits, and even communicate with their medical team from within the portal. From conversations with our patients, we have found that having access to this information promotes well-being, comfort, and trust in the healthcare setting. It helps people feel more connected to their doctors when, given the enormity of the clinic, it can be so easy to feel isolated.

Although establishing access to their medical records is very helpful to patients once they have left the clinic, we also focus on helping them navigate the system when they arrive for their appointments. We hold orientation classes for new patients and refresher classes for seasoned patients who are trying to get used to the new system of healthcare delivery that the clinic is implementing. We have also produced a detailed video that describes clinic resources that are available to patients. This video, now in English and Spanish, plays in the waiting rooms and serves to educate patients about who the different medical professionals in the clinic are, how to get their medications, make new appointments, and contact their doctors. We know that doctors and nurses are incredibly busy and do not always have time to explain these systems to patients every time they come in; our resources serve to save that time by helping patients avoid confusions that can arise when medications run out, when appointments are scheduled, and when navigating our health system.

While we have had many successes this year, there have also been a number of difficulties. The General Medicine Clinic is the largest clinic in the Midwest with an incredibly diverse patient population. Many of our patients travel for hours by car, bus, and train to get to the clinic. We also have a substantial number of patients who do not speak English. Many of our patients do not own computers, or do not understand how to use a computer. We have even met a handful of patients who admitted that they are unable to read or do not know what an email address is. All of these things make it difficult for us to promote the patient portal and overall health literacy.

These barriers are daunting and are not always surmountable, but we have found that ingenuity, patience, and understanding have really helped us. We create email addresses for those patients who do not have them so that we can sign them up for the patient portal. We give basic computer tutorials to those patients who do not know how to turn on a computer or who don’t know how to open a web browser. The video describing the clinic’s resources that we have created relies very heavily on images so people of all literacy levels can understand our message. Furthermore, listening to our patients and empathizing with them has been incredibly useful. All in all, thus far, this year has been filled with challenges, successes, and an incredibly large number of learning opportunities, and we have enjoyed every single one of them!

Primer for Experimental Modes Meeting

Here at Smart Chicago, we’ve always had three areas of focus:

  • Access to technology and the internet
  • Digital skills for all
  • Meaningful products from data

This focus keeps us on the right path— one that requires us to lay practical groundwork before delivering cool apps— to put people first. We’ve done this since day one.

On a personal note, I’ve been a worker in community technology for a long time. I love it. In the early 2000s, I started a side business to help people get an internet life. I learned that nearly no one goes beyond default configurations, or even knows they can.

Internet Life Services

In 2004, I conducted bilingual computer training at my church to teach people how to post to our blog. I learned that everyone has a thirst to express themselves.

Bilingual computer training

In 2006, I taught a 6-week course in websites for small businesses. I learned that people love certificates.

Websites for small businesses

I’ve also been a part of a parallel path, which started taking off right about this time: the open data and civic tech movements. As a co-founder of EveryBlock, one of the earliest examples of a site that sought to use civic data in communities, I helped shape and build things like the 8 Principles of Open Government Data and  Open Gov Chicago, a gathering of technologists in the field started in 2009.

EveryBlock Launch Screenshot

We like to think these worlds— those of community technology, grounded in the needs of the people, and civic technology, driven by the most technical people, are aligned. When we’re at our best, they are. Very often, however, they are worlds apart.

This is why Smart Chicago exists. Our mission, grounded in our areas of focus, situated directly in the community (as we work here at the region’s community foundation), based in community technology research (lead by the MacArthur Foundation), and fully engaged with the governments and institutions that serve the people (including the City of Chicago, one of our founders)—this is us. This is why we’re here today.

We designed this project to fit under a larger area of work that Smart Chicago: the Knight Community Information Challenge grant awarded under their Engaged Communities strategy to the Chicago Community Trust “as it builds on its successful Smart Chicago Project, which is taking open government resources directly into neighborhoods through a variety of civic-minded apps”.

Materials for today:

  • Here’s all of the research and synthesis that led us to this meeting in one handy PDF.
  • Here’s live meeting notes— follow along starting at 9AM Saturday, April 4, Central time.
  • Here’s the form we’ll be using in our Case Study Sprint: https://smartchicago2012.wufoo.com/forms/diy-case-study-civic-engagement-in-civic-tech. Fill it out!

There is no other organization in the country that is more qualified to lead this thinking. I am proud of where we are, and steeled for the work ahead.

CUTGroup #13 – mRelief

Receiving a text message from mRelief on a SamsungFor our thirteenth Civic User Testing Group session, we tested mRelief, a website that helps residents check their eligibility for social services benefits in Chicago and Illinois. This in-person test took place at one of the Connect Chicago locations – Chicago Public Library King branch at 3436 S. King Drive in the Douglas community area.

The mRelief team was interested in testing these things:

  • Language: We heard that language is very important in understanding if someone is eligible for social services. We wanted to see if testers understood terms such as: gross income (what to and not to include), value of assets, household size, etc. The mRelief team also uses pop-overs and wanted to see if testers use them and if they were helpful.
  • SMS: We wanted to test the text-messaging platform to see if users found it easy to use and how long each session took to receive and send messages
  • Website usability: In addition to testing SMS, we wanted to see how testers liked the mRelief website and how easy it was to use.

Segmenting

On January 21, we sent out an email to all of our 834 CUTGroup participants. We wanted to know if they would be available for an in-person test on January 29 for about 30-45 minutes. We asked some screening questions to gather information. We wanted to focus on testers who were currently enrolled in or qualified for social services programs. We also wanted to focus on residents of the Southside of Chicago

We ended up having 11 testers participate in this test. In addition, 1 individual from the library wanted to be involved in the CUTGroup, and therefore, we tested with him, but did not include him in the results because he would not have qualified for the test.

Screening Questions

We heard form 71 CUTGroup participants through our callout for testers. We received a lot of good information just from the screening questions. Here is a look at what we learned:

  • 90% of CUTGroup respondents said they have a smartphone
  • 33% of respondents are currently enrolled in social services programs

Test Format

Due to the personal nature of discussing social services benefits with testers, we decided early on that we wanted to do one-on-one proctoring. We had a total of 7 people from Smart Chicago and mRelief help with welcoming testers and proctoring.

This was the first test that we were not only testing a website, but also wanted to test the text messaging platform. Testers were asked to go through both processes, and we wanted to see how they compared and if testers found one platform easier to use than the other.

Results

8 out of the 11 testers were currently enrolled in social services benefits, and all of these testers were enrolled in Food Stamp benefits. The 3 testers who were not currently enrolled seemed to have qualified for social services benefits based on their household size and income.

 Language

We learned from this CUTGroup test that testers found the form(s) when checking their eligibility to be a simple process. 7 out of 11 testers (64%) mentioned that the questions were easy to understand or straightforward. Although most of the language was very clear, however, only 1 tester clicked on a pop-over to find out more information. Therefore, any question that requires additional help text should be added after the question.

In addition, some testers did not notice that some questions might ask for monthly income vs others that ask for annual income. These are key pieces of information that should be made clearer in the question format.

SMS

Out of 11 testers, 10 testers (91%) do not pay per text message.

Only 5 out of 11 testers (45%) thought that the questions were “easy” or “straight-forward” when going through the text messages. The biggest thing we heard was that testers were “frustrated” or “annoyed” that they were unable to fix a mistake and had to start from the beginning.

Lastly, some testers received messages out of order, and some terms were split up between messages (ex: “Medicaid”).

1 of our testers, Simple One (#9), experienced a delay between text messages that lasted anywhere from 1 to 2 minutes. She thought it was “annoying” that she did know when the end of the questions would be. 3 testers thought the process was pretty fast!

Website usability

When reviewing the homepage, 6 out of 11 testers (55%) mentioned the graphics. 3 testers felt that the icons were very intuitive and understood that they connected to the social services benefits. The other 3 testers did not think the icons were as intuitive. 4 testers thought the graphics were icons that they could click on.

Most of the testers had a clear direction after reviewing the homepage. The majority (9 testers) clicked “Get Started” while the other 2 testers clicked on “Programs.” It was great to see that testers had a clear path in using the website that got them to the most important part — checking their eligibility for social services.

When checking eligibility, we noticed that a lot of testers liked the idea of the calculator to calculate income, but most testers did not use if because they had already answered the questions before coming across the calculator located at the bottom of the screen.

Documents & More Information

We wanted testers to tell us how important it would be for mRelief to provide additional information about documents that they might need to bring, or money that they would receive from the benefit. We hoped this would be a help for mRelief to decide what the next steps for the website might be. Here is what we learned:

How important is it for you to know what documents you need to prepare for an interview with a caseworker to receive public assistance?

5 Very Important      100% (11)
4 Important                    0%
3 Neutral                         0%
2 Slightly Important     0%
1 Not Important            0%

Is it important for you to know why you do or don’t qualify for public assistance?

5 Very Important      82% (9)
4 Important                     18% (2)
3 Neutral                          0%
2 Slightly Important      0%
1 Not Important              0%

How important is it for you to know how much money you will receive from public assistance before going to file with a caseworker?

5 Very Important      46% (5)
4 Important                     9% (1)
3 Neutral                          9% (1)
2 Slightly Important      18% (2)
1 Not Important             18% (2)

How important is it for you to know how to file an application for public assistance?

5 Very Important      91% (10)
4 Important                     0%
3 Neutral                          0%
2 Slightly Important      0%
1 Not Important              9% (1)

Final Report

Here is a final report of the results with the analysis of the questions we asked, followed by each tester’s responses, and copies of other questions we asked:


The raw test data can be found below with the written answers from every tester.

Tomorrow: Convening on Experimental Modes

Tomorrow Smart Chicago is hosting a convening at The Chicago Community Trust as part of our Experimental Modes of Civic Engagement in Civic Tech Project.

The gathering, led by consultant Laurenellen McCann, brings together people who provide community leadership in the creation of technology. Each participant works in different contexts, with radically different types of technology, but what unites us is the degree to which our work is done with, not for, the communities we serve.

Here’s all of the research and synthesis that led us to this meeting in one handy PDF:

Here’s live meeting notes— follow along starting at 9AM Saturday, April 4, Central time.

Here’s the form we’ll be using in our Case Study Sprint: https://smartchicago2012.wufoo.com/forms/diy-case-study-civic-engagement-in-civic-tech. Fill it out!

All are welcome to participate in the meeting notes, to complete your own case studies, and to hang out of Twitter— @SmartChicago and @elle_mccann are good places to start.

The Chicago Community Trust

Two New Health Navigators

Yesterday we on-boarded two new Health Navigators in our Smart Health Centers program: Anthony Green, who will be working at Family Focus and A.C.T.S. OF F.A.I.T.H., and Daniel Broome, who will work with people at Ann & Robert H. Lurie Children’s Hospital.

Anthony Green and Daniel Broome

Kyla Williams runs this and all of our health projects. Yesterday Anthony and Daniel were trained by Patrice Coleman and Diana Beasly. Welcome!

IMG_3182