YaleGlobal Ends Publication

Thank you, Nayan and Susan, for highlighting interconnectedness of humanity over all these years!

Sorry to see it close, especially at a time when it is most needed.

I will really miss your publications!

Thank you Nayan. YaleGlobal has been a great service and will be sorely missed.

Hello Susan, I hope you are well in these crazy Covid times. You and I corresponded back in January, when I wrote to you with some strong concerns about the inaccuracies and poor methodology featured in an article regarding homelessness that was posted on Yale Global Online website at https://yaleglobal.yale.edu/content/cities-grow-so-do-numbers-homeless. In your January response, you admitted that there are severe factual and methodological issues, and you said that you were going to deal with it by putting the OECD’s warning on the article about the methodological issues. Which I see you have done, but unfortunately that has not fixed the problem.

I am still seeing that article being cited by others – indeed, on your home page it shows that this article is your “most popular”-- using it as the basis for claiming that the homelessness problem in the US (or in San Francisco where I live) is not so bad because we are lower than a number of other countries in Europe (as is Mexico, according to the data in your article). Yet, for reasons that I previously outlined, that conclusion is absolutely insupportable by the evidence that is buried several layers down in the citations for this article. At this point, this shoddy research and journalism is providing dis-information for misuse by opponents of better homelessness policy.

The problem is the chart titled “Homeless as Percent of Total Population” that you still are featuring on that article, which is completely misleading and based on the wrong data. Yet that chart is the takeaway that readers and researchers are quoting from the article, despite the OECD and other warnings you have posted. But as I previously detailed, the headline on that chart does not match the actual OECD chart that is the source of the information. As I previously wrote:

“The bar graph in the article called “Homeless as a Percent of Total Population, OECD countries, 2015” has a link to an OECD citation at the bottom of the graph which takes you to the following webpage: http://www.oecd.org/els/family/HC3-2-Homeless-strategies.pdf That webpage actually has NO DATA at all about “homeless percent of total population”, instead it is a document showing strategies to counter homelessness in a range of OECD countries.

However, the text of the OECD article itself contains a second link to this document: http://www.oecd.org/els/family/HC3-1-Homeless-population.pdf. (See the text that reads “Trends in homelessness among OECD countries with available data are mixed. In recent years rates of homelessness are reported to have increased…”), and that document does contain data like the kind that is in Mr. Chamie’s article. However, the data in this second OECD document does not match the data in Mr. Chamie’s bar graph. For example, look at Germany in both the bar graph and in the OECD document. The bar graph shows German at about 0.19%, but the OECD document shows .41%. Austria in the bar graph is about 0.18%, but in the OECD document it is 0.25%. There are a number of inconsistencies between the bar graph and the OECD document, even though it appears that the data for the bar graph is drawn from that OECD document.”

In addition, as I previously wrote: In the chart on your webpage, “the nation with one of the lowest percentage of "Homeless as % of total population" is Mexico, at 0.04% compared to US at 0.17% (over four times higher), or Germany .41% (TEN times higher than Mexico?), or Austria .25% (six times higher than Mexico? And 50% higher than in the US?). And Greece is listed at .19%, less than half of Germany. For work, I have traveled to all of these places, and lived in Berlin for most of 2016-18, and honestly those numbers simply do not match the on-the-ground reality.”

Susan, I am very dismayed that this article is still being cited to give out a wildly inaccurate picture of comparative rates of homelessness in various OECD countries. To make it really simple – the chart that you have displayed in this article on your webpage is COMPLETELY WRONG. It’s citing the WRONG data from the WRONG OECD chart. It’s not even clear where the data in the chart that you are featuring comes from, because none of the sources cited for it actually show that data.

The warnings and caveats you have put in the article are completely inadequate, and have done nothing to prevent others from continuing to cite the false information in this “fake OECD chart.” The fact that you, as editor, and that Joseph Chamie, as the author, continue to publish this fake chart and fake data is extremely alarming.

If you don’t have time to actually make the data right, it seems to me you have two choices, in order to remain the intellectual integrity of this website. First, just completely remove that wildly inaccurate OECD chart, as well as other quantitative information, from this article. The article itself has a decent discussion about the difficulties of counting homelessness, so if you just remove all the quantitative information from it then perhaps this article could still be useful. But that would take some work on somebody’s part, and it appears that your website is not operational anymore (according to information on your home page), even as other people on the web continue to find this misinformation on your website. Your homepage says that it reaches 500,000 people annually so I’m sure many of them are continuing to come back to your page. I am guessing that you, personally, would like to see better policy that reduces homelessness in the US, and around the world. Given that assumption, honestly I don’t know why you have left such a grossly inaccurate article there to live for eternity, spewing out misinformation about homelessness that will be cited by opponents of good homeless policy.

So the second option, which might be easiest for you if there is no one left minding the store there, is to simply delete this article entirely. The article as it stands is based on factual misinformation, is sloppy in the extreme, in terms of matching data to sources, and is helping opponents of better homeless policy. If no one has time to make it right, then for heaven sakes show some journalistic integrity and kill the damn thing. That way no one will continue to cite it to support their backward political agenda. That’s what is happening right now, and at this point you are aiding and abetting such efforts. Please make it stop.

As a Yale graduate, I find it appalling that this kind of sloppy academic misinformation is being allowed to live next to the name “Yale.” As the author of seven books, and hundreds of op-ed’s, articles and media interviews for media outlets like the New York Times, Washington Post, CNN, CNBC, the Atlantic, and more, I understand that sometimes we make mistakes as journalists. But this is the SECOND time now that I have written to you at length, providing great detail regarding why this article, as it is presently, violates a number of journalistic norms, and must be removed from the public domain until it can be fixed.

Please, either make this article right or make it go away. And please let me know when you have rectified the situation. Thank you. (I have also sent this to you by email, including our previous correspondence back in January 2020)

All the best,

Steven Hill

www.Steven-Hill.com
www.ExpandSocialSecurity.net
www.RawDealBook.com
@StevenHill1776

Hi Susan, thank you for your response. At least now the data in the article on your website is actually sourced correctly, i.e. to the OECD chart that actually has comparative data about homelessness from country to country.

But as previously mentioned, the other problem I have with your presentation is that the OECD data itself, as presented in the chart in your article at https://yaleglobal.yale.edu/content/cities-grow-so-do-numbers-homeless, simply makes no sense, and deserves greater scrutiny in order to live up to the most basic norms of journalistic integrity. Despite all of the caveats and warnings about methodological differences and challenges, as a journalist if I were trying to cover the issue of homelessness, I would not simply be satisfied with a chart from the OECD that doesn’t make sense. I would contact the OECD, try to find the author(s) of that chart, and ask them some hard questions.

For example, the OECD chart at http://www.oecd.org/els/family/HC3-1-Homeless-population.pdf lists homelessness in Germany at 0.41% based on data from 2018. But here is a recent article from Deutsche Welle (February 7, 2020), reporting on how the government of Berlin just executed its FIRST HOMELESS COUNT EVER in January 2020. From the article: “Dubbed the "Night of Solidarity," around 2,600 volunteers were divided into teams of three and five, who scoured more than 600 different areas in Berlin to conduct the census, which took place on the night of January 29-30.” Berlin is the largest city in Germany. If Berlin had never before done a homeless count before 2020, how could the OECD be quoting data for all of Germany from 2018? Link to article https://www.dw.com/en/germany-nearly-2000-homeless-in-berlin-says-census...

Going deeper into that data, this recent homeless count in Berlin found a total of 1,976 homeless people. Berlin is a city of about 3.8 million people. Therefore, the rate of homelessness in Berlin, according to this official count, is .05%. As I’m sure you know, homelessness is largely an urban phenomenon, so if Germany’s most populous city has a homelessness rate of .05%, how could all of Germany have a homelessness rate of .41%, which is eight times greater than its largest city? In fact, given that Berlin is actually poorer than other large German cities, such as Munich, Hamburg and Stuttgart, those cities probably have an even lower homelessness rate than Berlin. And the whole country would have an even lower homelessness rate than Berlin.

So the OECD data simply makes no sense for Germany, no matter how many caveats and warnings you put around it. It also doesn’t make sense for other countries, as I previously wrote. For example, the OECD data that you cite lists the lowest homelessness rates in some of the poorest countries in the OECD – Mexico (.04%), Croatia (.01%), Hungary (.01%), and others. So Germany’s homelessness rate is FORTY TIMES that of Hungary and Croatia? Or 10 times higher than Mexico? Tidy, well-kept Austria’s homelessness rate is listed as .25% – 25 times Hungary and Croatia, and six times greater than Mexico? Greece is listed at .19%, less than Austria and less than half of Germany? The US homeless rate is listed at .17% – four times that of Mexico, less than Austria and less than half of Germany?

As a journalist, I have traveled to all of these places, and lived in Berlin for most of 2016-18, and honestly those numbers simply do not match the on-the-ground reality.

As researchers, journalists and social scientists, I believe we cannot simply regurgitate any data set that fits our frame or the story that we are trying to convey. We should not cherry pick data. And sometimes we have to try harder to get at objective data based on science and facts, and when that data seemingly does not match reality, we have to dig deeper. The author of this article, Joseph Chamie, should have tried harder to contact the OECD and figure out some of these gross inconsistencies. And whoever edited this article for Yale Global – whether it was you or someone else – should have insisted that he do it.

As it is, I don’t believe this article passes a basic test of journalistic legitimacy or integrity. Again, as I said before, I think you should try harder to make this article right, in the ways that I have proposed, or take it down. At the very least, update the chart, as well as the narrative, to give the latest data from Berlin (and by extension Germany). As it is, this article is being found on the Internet by people who are using this skewed data for partisan purposes (it doesn’t matter to them that you have listed numerous caveats about the possible unreliability of the data or methodology used). And it will live forever on the Internet, forever putting forward this terribly distorted picture about comparative homelessness rates from country to country. It is listed as the most viewed article on your website, and when I search on Google for “Estimated number of homeless people, by country,” this article is ranked number six.

I deeply hope that you care enough about this issue, as well as the reputation of your website, to either rectify this situation or “unpublish” this article. There are a number of other skewed articles out there, such as on Wikipedia, that also use questionable data. As a Yale alumni, I think any article associated with Yale should strive to do better than Wikipedia, and should strive for journalist excellence that makes a REAL contribution to this important issue.

Thank you, all the best

Steven Hill

www.Steven-Hill.com
www.ExpandSocialSecurity.net
www.RawDealBook.com
@StevenHill1776

Why is YaleGlobal ending publication? This is not explained to us.

As a Sinophile and an internationalist, I and all my brethren have regrets. Yale Global provided depth and width to international activity and behavior.
The balance it provided with other thoughtful sources can only, in my mind, be judged as irreplaceable. I look forward to seeing renewal, with optimism.
Dan Alexander

Thank you for the years of insightful, thoughtful dialog about globalization. YG has source of much needed analysis about this critical aspect of the modern world. You will be missed...