Buffalo Most Anti-Gay U.s. City?

[drpaladin]

A recent unofficial study using the origin of tweets on Twitter has given Buffalo, NY the dubious distinction of being the most anti-gay city in the U.S.. It was followed closely by Arlington, TX and a number of cities in California.

 

 

 

This study also studied anti-black slurs. Some cities made both lists. There are a few qualimetrics that come to mind viewing these results. Does this reflect the views in general in thes cities or is this for some reason just limited to Twitter users? Twitter seems to be a straight mind to world medium that is prone to more free writing than anything other than speech itself.

 

[MJ85]

And who exactly commissioned this, ah, study?

 

For being an "unofficial" study, whoever's directed it sure wants to make it seem like something with meaning.

[drpaladin]

And who exactly commissioned this, ah, study?

 

For being an "unofficial" study, whoever's directed it sure wants to make it seem like something with meaning.

 

You seem eager to dismiss the validity of the study. However the raw data comes straight from Twitter, so the figures themselves are undeniable. To answer the source of the study, it was Adobo, which is an apartment listing and locating service. I'm sure they intended it to be an aid for those with thoughts of possibly relocating. You realize  that similar demographics are psted for other indices such as crime, best schools, shopping, and economic aspects. As we have seen in the past, Twitter is a good indication on what is trending in certain areas.

[Ron]

I can see the numbers but that doesn't mean they have any real significance. Just as with any "unofficial" study, the parameters (if any) aren't stated. What sort of comment is considered anti-gay or -black, by the way? A question that doesn't seem to be given consideration. Also, there could be a certain number of people who consistently (and vehemently) make derisive comments about gays. The same can be said for some bigots who are vociferous about their hate. Are these comments counted as one for each person, or is each tweet counted individually? Drpaladin writes that the study was based on origin, which indicates location only and which is a very good measure for the purpose of this study. Because, I say, overeaching was the study aim, and for that reason, it is spot on in its purpose. I see no good reason here to pass judgment on entire cities, no good reason, at all.

Edited March 7, 2016 by Ron

[MJ85]

Indeed, it strikes me as something that can be intended as a click-bait weapon while at the same time having a foundation about equivalent to a house of cards.

[drpaladin]

I can see the numbers but that doesn't mean they have any real significance. Just as with any "unofficial" study, the parameters (if any) aren't stated. What sort of comment is considered anti-gay or -black, by the way? A question that doesn't seem to be given consideration. Also, there could be a certain number of people who consistently (and vehemently) make derisive comments about gays. The same can be said for some bigots who are vociferous about their hate. Are these comments counted as one for each person, or is each tweet counted individually? Drpaladin writes that the study was based on origin, which indicates location only and which is a very good measure for the purpose of this study. Because, I say, overeaching was the study aim, and for that reason, it is spot on in its purpose. I see no good reason here to pass judgment on entire cities, no good reason, at all.

 

It is certainly fr from an exacting scientific study and I did qualify it as such. Please note the question mark in mytitle. We don't know what percentage of the population in these cities are tweeting. The terms they used to determine hate speech are easy enough to imagine. I know what I don't like to hear and what constitutes 'fighting words.' You will note that Baltimore, MD is at the top of the list in anti-black slurs and there we can also see that there has been significant racial trouble in that city. Is the frequency of tweets fueled by those incidents or is it a sign of the way people think? You get into the chicken and egg argument there.  You can't really ask openly if someone is prejudiced or not and get any accurate answers. Any survey would need to be extremely deft to come up with valid results. However this does indicate what is appearing frequently in these cities and I wouldn't tend to dismiss it out of hand without seriously considering it.

 

If you want to look at a breakdown of other results from this study, you can see them here:

 

http://www.thedailybeast.com/articles/2016/03/02/the-most-prejudiced-places-in-america.html?source=TDB&via=FB_Page

[drpaladin]

Indeed, it strikes me as something that can be intended as a click-bait weapon while at the same time having a foundation about equivalent to a house of cards.

 

If you look at the frequency of hate crimes committed in Buffalo, it isn't as you say 'a house of cards.' Those are way, way above average. That would tend to correlate with the study.

 

In addition, four of the cities in the list are in states that still have sodomy laws on the books despite the fact that the Supreme Court ruled them unconstitutional over a decade ago. Those laws are rarely enforced on heterosexual couples and can be viewed as the prevailing thought in those areas.

[spike382]

I don't really have time to do extensive research to check the validity of the findings about actual concrete facts like hate crimes committed against gay people. But I will say that this is entirely based on tweets, which isn't actually a good indication of anything, because it's online, and I'v since lost count of the racial and homophobic slurs that are posted online. If I were to judge that I'd start to think everyone in the English speaking world was a hardcore racist homophobe.

[Valkyrie]

I live in Buffalo, NY and am appalled to be on the top of this list.  It is definitely NOT my experience as someone born and raised here. Are there homophobes and racists here?  Yes, but I don't think with any more prevalence than the rest of the country. 

[Carlos Hazday]

I live in Buffalo, NY and am appalled to be on the top of this list.  It is definitely NOT my experience as someone born and raised here. Are there homophobes and racists here?  Yes, but I don't think with any more prevalence than the rest of the country. 

By any chance do you know an attorney named Michael Serotte?

[Graeme]

I live in Buffalo, NY and am appalled to be on the top of this list.  It is definitely NOT my experience as someone born and raised here. Are there homophobes and racists here?  Yes, but I don't think with any more prevalence than the rest of the country. 

Look at the numbers: 168 anti-gay tweets out of 100,000. That's 0.168%. There's a lot of uncertainty as to how they're counted if one person sends multiple tweets, but I think you could probably safely say from this that less than 1% of the population is openly anti-gay, based on their tweeting habits. Would that agree with your experiences?

 

To me, the numbers are too small to be useful. The margin for error is too high.

[drpaladin]

I live in Buffalo, NY and am appalled to be on the top of this list.  It is definitely NOT my experience as someone born and raised here. Are there homophobes and racists here?  Yes, but I don't think with any more prevalence than the rest of the country. 

-

Consider the possibility that you just aren't aware of it. Many of the people who live where I live think that it's a nice quiet place, except that it's not. Very little of what goes on gets into the newspaper or on TV. What does get there is seldom the truth.

 

Here are some hard statistics for you straight from crime reports. These are hate crimes from the NYS Division of Criminal Justice Services for Buffalo PD.

 

2009 - 32

2010 - 32

2011 - 28

2012 - 37

2013 - 17

 

Now consider that during the same time frame that NYPD listed the following for the same time frames: Kings County NYCPD 92 / 135 / 106 / 157 / 123.

 

NYC has millions of people and Buffalo around 258,000. To equal the rate in Buffallo, the numbers in NYC would need to be in the thousands. Please note that the statistics don't specify what kind of hate was involved.

[Bill W]

I live in Buffalo, NY and am appalled to be on the top of this list.  It is definitely NOT my experience as someone born and raised here. Are there homophobes and racists here?  Yes, but I don't think with any more prevalence than the rest of the country. 

 

Val, do you think some of the tweets (possibly the majority) may be coming out of the University of Buffalo or Canisius? 

[drpaladin]

Look at the numbers: 168 anti-gay tweets out of 100,000. That's 0.168%. There's a lot of uncertainty as to how they're counted if one person sends multiple tweets, but I think you could probably safely say from this that less than 1% of the population is openly anti-gay, based on their tweeting habits. Would that agree with your experiences?

 

To me, the numbers are too small to be useful. The margin for error is too high.

 

Pollsters use numbers smaller than that all the time. Gallop and Pew must have it all wrong then.    I'm not saying that this information is 100% accurate. Let's say it's trending.

[spike382]

But there's all kinds of room for uncertainty based on tweets. The hate crime statistics are a much better indication of anything. I could have more than one Twitter account and say whatever I want and see what gets moderated and what doesn't. Two or three people could have done all that.

[drpaladin]

But there's all kinds of room for uncertainty based on tweets. The hate crime statistics are a much better indication of anything. I could have more than one Twitter account and say whatever I want and see what gets moderated and what doesn't. Two or three people could have done all that.

 

So two or three people are sending out all those tweets over months to make Buffalo look bad in a study that no one knew was going to happen? I wouldn't think so.

[spike382]

So two or three people are sending out all those tweets over months to make Buffalo look bad in a study that no one knew was going to happen? I wouldn't think so.

To make Buffalo look bad as the sole motivation? No. And it's probably more than two or three, but there's a lot of factors of uncertainty. I mean trolls are always a factor to keep in mind. People who say whatever wherever just to stir up a ruckus. I just don't think tweets are really enough to say an area is more racist or homophobic. I mean this assumes most of the population uses Twitter. I don't, I know plenty of person don't. Granted all I have to back that up is anecdotal evidence, but tweets are suspect in and of themselves. Like I said, the actual crime statistics are a much better indication.

[Valkyrie]

By any chance do you know an attorney named Michael Serotte?

Nope, never heard of him. 

 

As for the rest of the comments, I am too tired to think of a cogent reply right now. 

[drpaladin]

To make Buffalo look bad as the sole motivation? No. And it's probably more than two or three, but there's a lot of factors of uncertainty. I mean trolls are always a factor to keep in mind. People who say whatever wherever just to stir up a ruckus. I just don't think tweets are really enough to say an area is more racist or homophobic. I mean this assumes most of the population uses Twitter. I don't, I know plenty of person don't. Granted all I have to back that up is anecdotal evidence, but tweets are suspect in and of themselves. Like I said, the actual crime statistics are a much better indication.

 

And the statistics tend to back up the evidence from Twitter. You might have overlooked it, but Buffalo made the ani-black slur list also. Most of a population doesn't have to participate to get a selection of attitudes. Pollsters don't contact every person in an area to get their opinion either, but more times than not polls prove to be fairly accurate. Twitter is as anonymous as the person using it chooses to make it and that lends itself to unfiltered opinions. You can't punch a tweet in the nose or picket a twitter account. As ISIS has discovered, it is a perfect way to rapidly spread hate and shouldn't be easily dismissed or ignored. This is why Anonymous targets ISIS and the KKK on social media accounts. You notice Baltimore is at the top of the anti-black list and given the problems there it makes perfect sense.

[Tiger]

The only things that surprise me is that St. Louis is not in the top 10 list for black slurs and that there are a no lists for anti-white, anti-Hispanic, and anti-Muslim slurs. 

[drpaladin]

The only things that surprise me is that St. Louis is not in the top 10 list for black slurs and that there are a no lists for anti-white, anti-Hispanic, and anti-Muslim slurs. 

 

The link I posted in post #6 contains some other graphs for Hispanics, women, overweight people. Perhaps anti-white and anti-Muslim escaped them. You have to remember that this came from a journalism article. For a great many journalists whites don't experience prejudice, they simply deserve it for past sins, just like that myth that it is impossible for a black person to be a racist. Unfortunately racists come in all colors.

[Graeme]

Pollsters use numbers smaller than that all the time. Gallop and Pew must have it all wrong then.    I'm not saying that this information is 100% accurate. Let's say it's trending.

Pollsters also include an error in their figures which is usually around 2-3% mark. When the % being reported is 0.168%, a 2-3% error rate (taking a polling standard) means that the numbers being reported are statistically indistinguishable from zero.

 

Okay, I'm being a bit harsh because what pollster do and what this study is doing are two completely different things, but to me that 0.168% figure is so low that without some rigorous methodology and long-term history behind it (which applies to medical statistics that can be of a similar order of magnitude), the numbers aren't reliable.

 

I've just gone to the article that was linked and it appears that this was a study of tweets from June 2014 to December 2015 - either 18 or 19 months, depending on whether the endpoints were included and whether they were whole months. That's actually not a bad baseline, but I'd like to see the level of variability over that time. I'd also like to see this extended out over 5 or 10 years, possibly with trends over that time. For example, that period covers the Supreme Court decision in Obergefell vs Hodges in June 2015. I would suspect that there was a large spike of anti-gay tweets in response to that decision, but how much did that spike affect the overall averages? Sometimes an average can hide the volatility in the underlying data. One or two large spikes in response to one-off events can distort the underlying average.

 

In short, without some indication of variance (they should be able to calculate standard deviations, given the raw data), it's not possible to judge if the numbers supplied are meaningful.

[drpaladin]

Pollsters also include an error in their figures which is usually around 2-3% mark. When the % being reported is 0.168%, a 2-3% error rate (taking a polling standard) means that the numbers being reported are statistically indistinguishable from zero.

 

Okay, I'm being a bit harsh because what pollster do and what this study is doing are two completely different things, but to me that 0.168% figure is so low that without some rigorous methodology and long-term history behind it (which applies to medical statistics that can be of a similar order of magnitude), the numbers aren't reliable.

 

I've just gone to the article that was linked and it appears that this was a study of tweets from June 2014 to December 2015 - either 18 or 19 months, depending on whether the endpoints were included and whether they were whole months. That's actually not a bad baseline, but I'd like to see the level of variability over that time. I'd also like to see this extended out over 5 or 10 years, possibly with trends over that time. For example, that period covers the Supreme Court decision in Obergefell vs Hodges in June 2015. I would suspect that there was a large spike of anti-gay tweets in response to that decision, but how much did that spike affect the overall averages? Sometimes an average can hide the volatility in the underlying data. One or two large spikes in response to one-off events can distort the underlying average.

 

In short, without some indication of variance (they should be able to calculate standard deviations, given the raw data), it's not possible to judge if the numbers supplied are meaningful.

 

The 2-3% would be applied to the final results themselves not to the percentage of the representative sampling.

 

An additional graph with a timeline indicating spikes would have been useful. With that you could establish a baseline of sorts.

Edited March 8, 2016 by drpaladin

[LeonMichaels]

It is interesting to use twitter data in this way, but I'd have to agree with many posters here that I'm not sure how meaningful these results are. There are just too many unknowns in their method for me not to be very skeptical.  

 

Anyway, after some clicking I did find the list of 'derogatory' keywords in their search. Here's what they used (although I have no idea what they did with 'neutral language' hits):

 

Gay and lesbian people:

• Negative language: fag, faggot, homo, dyke, sodomite, lesbo
• Neutral language: gay, lesbian

 

The full list is here