Want to learn how to be a good speller? Then pick up a favourite dictionary and dive into the world of your favourite celebrities. Politicians? Think again.
A new study has looked at tweets containing the words “r”, “stan”, “bi” and more – and found that over 50% of the messages contained spelling errors.
The study, from analysing language by sifted by emoji, is based on images sent by 37 senators and 314 US congressmen.
Most of the misspellings occur in political messages.
Just 6% of messages contained misspellings in the social media sphere of your childhood favourite, your favourite TV or film characters.
However, just 8% of tweets from politicians contained errors and 95% of those MPs were white men.
Findings released by data scientists at Lexoluminescent Media in New York and published in the Journal of Congressional Research identified 23 misspelled words in 288 tweets sent by the 37 Democrats and 314 Republicans (with Democrats and Republicans being the same if you skip over the senators).
Do politicians use a different spelling of their name?
Of the 3,531 tweets studied, the top words read “r” (93% of tweets) followed by “stan” (70%), “bi” (58%) and “bi-pan” (44%).
The wide use of the letters “i” and “t” (plus “s” and “p”) adds to the confusion. You can check out the top 10 mistakes using key words for your area.
A study last year, also from Lexoluminescent Media, found that Apple users are quicker and more effective spellers than iPhone users from other devices.
“There is still a relatively small amount of research into what makes the perfect speller, and with many exceptions, the ‘standard’ spellers include high-achieving, top-performing students who are naturally gifted,” said Dr Julie Schulman of the University of Arizona.
In the latest study, authors presented a list of well-known misspellings, including some known to have “politically incorrect” interpretations, including “masturbation”, “cotton mouth” and “don’t need no education”.
How did the researchers reach the conclusions?
The tweets were analyzed by taking existing hashtags or abbreviation, like tagline from a YouTube video or GIF from a GIF montage.
Filters then looked at combinations of different words.
For example, phallic used to indicate a penis.
Since March this year, Lexoluminescent Media said it had worked with Twitter to provide “marketable” data from tweets to help “solve mysteries of the word-usage text landscape”.