“Lots” of hype
The big to-do today online is the story of “a lot” of racist tweets after Nina Davuluri was crowned Miss America last night. Ms. Davuluri is the first American of Indian descent to be crowned in the contest.
A few online publications like Buzzfeed posted headlines that gave the impression that the racist backlash against the decision to crown Ms. Davuluri was rampant and overwhelming. Words and phrases like “A Lot” and “Many” and “Several” were written into headlines to describe the amorphous amount of racist tweets coming out last night. Yet Buzzfeed could find a mere 23 to post (and publicly shame their authors out of context. What if one was being sarcastic?).
No numbers? No deal!
Be wary of any headline that uses non-numerical quantifiers or adjectives like many, several, or lots. Usually these words are used when real numbers can’t be pinned down, or the real numbers aren’t impressive.
Compare the information from this example headline:
“Many tweeters upset about Miss America being of Indian descent.”
to the information in this example headline:
“[n]% of US Tweets in the 2 hours following the crowning of Ms. Davuluri were negative and racist in nature.”
The first headline leads us to believe that the majority of Americans were not happy with the Miss America decision to crown an Indian-American. When we see the word “many” we generalize it to mean “majority” or “most.” That mental mashing of “many” to mean “most” is what makes the headline catchy. If the editor wrote “We found 23 tweets that were racist and we think that’s many tweets,” people would stop reading that publication. Here’s the thing: that’s exactly what happened. They used the term “many” to quantify a few tweets. It’s irresponsible and misleading, and it’s meant to get clicks to sell ads. It’s crap.
Let’s do the math
The overwhelming majority of tweets in the last 24 hours were not about Miss America at all. How do I know this, when I have no way to parse out the data? Let’s look at some real numbers and some common sense experience that we all have with Twitter.
I once heard a state representative say that it’s common local-politics lore to assume that 10,000 of their constituents agree with each letter they receive. So 1 letter=10,001 supporting opinions in their district. This apparently is a tried-and-true rule of measure within local politics. Let’s adopt this idea for a second.
Even if each of those 23 tweets Buzzfeed put up represent, say, a million people’s opinions – each – those 23 million tweets can’t responsibly be qualified as “several” “a lot” or “many” of the overall number of tweets about the Miss America pageant or of the total number of tweets in any 24 hour period.
The site averages 5,700 tweets per second.
That’s 342,000 tweets per minute.
20,520,000 tweets per hour.
492,480,000 per day.
Let’s say every tweet in the 2 hours after Miss America 2013 was crowned were racist crap. That would be 41,040,000 tweets. Let’s say that represents all of the racist tweets for that day. That would put racist tweets about Miss America 2013 at around 8% of all tweets for that day.
We can drill down, sum up, move around and get jiggy with these numbers, but the fact is, neither Buzzfeed or PurpleCar has the exact data for the number of racist tweets about Nina Davuluri. We can assume, though, based on our experience, that all tweets for 2 hours after the Miss America 2013 pageant ended were not about or related Nina Davuluri or Miss America. So the racism tweets have to be spread out, because other people are clogging up the stream with pictures of their food or complaints about insomnia.
When I woke up this morning, #MissAmerica was trending. When I checked the tweets, I saw not one racist tweet using that hashtag. Instead, it was post after post after post with a link to the Buzzfeed or similar articles about the racist tweets found the night before. Not one of the articles had any kind of statistics or numbers. They all used non-numerical, amorphous quantifiers. I went searching for the tweets myself. I searched with hashtags, without hashtags. I used all sorts of racist terms. I included references to Miss America and excluded them. Go try it yourself. You may lose your appetite for dinner, but you won’t find millions of racist tweets in general, let alone racist tweets about Miss America.
Real meaning
Now that we’ve concluded that these types of headlines are misleading, let’s discuss the good or the damage or they do.
Benefits:
1. Buzzfeed & other offenders make money on ad clicks/exposure. (Didn’t qualify who receives the benefits, did I? One could say ads power the Internet…)
2. Racism is brought to the forefront, perhaps making some oppressed individuals feel vindicated (Had a conversation about this on Twitter today. I’ve seen this argument about supporting the highlighting of public displays of racism. The logic is that if it is never reported then no-one will believe it exists).
3. Parents can be informed that online social media sites contain dangerous garbage (It’s possible there may be a parent out there who does not know this yet).
Faults:
1. The mental exchange of “many” to “most” makes racism seem more rampant than it is (Say what you want, but societal etiquette has changed drastically in the last 50 years. Perhaps people are more racist now but they certainly display it less).
2. The hopeless feeling the bad “news” engenders can deplete people’s hope and well-being (Several studies show that news like this affects us poorly).
3. The United States and online social networks like Twitter get a needlessly tarnished reputation (We have enough damage control, we don’t need more to do).
We can think up more pros and cons. How you think about this issue will all depend on your own views on racism, social networking and media outlets. Your own experiences inform your opinion. Perhaps you feel this type of racism isn’t brought to the forefront enough, and you are happy to see that someone picked out a small number of these tweets as proof. Perhaps you’ve been online so long that headlines like this seem sensationalistic and unnecessary.
Personally, I think the reports do more harm than good because they do not include actual numbers or real data. By giving the impression that the majority of Americans (and not just American tweeters because people generalize) are racist will only create the unintended consequence of more public racism. If a person is on the edge about expressing a racist opinion and then sees a trending topic that “many” people agree with her, she will be encouraged to share that thought. This is how social rules change. Slowly but surely, our culture has changed to the state where expressing racist views in a public forum is met with disdain. We should continue to reinforce that disdain by not sensationalizing racist tweets, and we should definitely think of the real numbers each time we see “many” headlines like these.
Comments on this entry are closed.
While I agree that “no numbers = no deal” … 41,040,000 tweets is a boat load of tweets about one subject, even if the time frame was spread across a full year. Most brands & events dont see a fraction of that volume.
Yes, it’s an impossible boatload of tweets, yet it still doesn’t come close to being thought of as a majority of tweets. My point was to be overly generous to the proposed amount of racist tweets and even then showing that it is irresponsible to qualify them as “many.”
The proper way to calculate would be to look at all Miss America tweets, and pull out how many are ignorant / racist using specific keywords. It’s doable. Try Topsy.com.
Topsy’s fun. Thanks for sharing that link. One can lose hours there. I speak from experience.
There really would be no one proper way to measure racial tweets. The variation on wording, phrases, etc., would make it impossible to catch it all. I’m sure at least a few people tweeted without using the hashtag or any identifiable marker. Also, we’d have so many other questions to ask. What tweets? “Most” of what time period and geographic location and language? It’s just a messy study. This is all the more reason to sincerely doubt headlines with “most” when it comes to tweets.