Archive for ‘Influence’

November 11, 2013

You are what you tweet: Researchers predict users’ gender with 92% accuracy

How often do you think about what you are telling the world about yourself when you post an update to your social media profiles? Well, it turns out that you are being studied, whether you know it or not. Earlier this fall, PLOSone published a study that aimed to link the vocabulary netizens use with their age, gender, and select personality traits. The unique twist on this study was the methodology. Instead of using known word correlations to base their analysis on, they adopted an open vocabulary approach in an attempt to “find connections that are not captured with traditional closed-vocabulary word-category analyses”.

The use of the open vocabulary approach yielded some interesting results:

1. Men are much more likely to use profanity and talk about gaming while women seem to be much more positive and upbeat. *The size of the word in the word clouds below indicates the strength of the correlation; color indicates relative frequency of usage. Underscores (_) connect words of multi-word phrases.

MenvsWomen

 

2. Your age can be determined based on whether you talk about school, work or family.

Age

 

3. Extroverts like to party, introverts like the internet, neurotics use angry and depressed language, and the emotional stable like….basketball?

PersonalityTraits

 

4. Finally, the people in your social media networks who’s updates are negative, profanity-filled, and often tiresome, may rank low on the agreeableness, openness, and conscientiousness scale.

Agreeable

Beyond simply being “interesting”, these correlations will further help communicators and marketers get their message in front of the right audience – You need to know where your audience “lives” before you can influence them. Be sure to check out the full study “Personality, Gender, and Age in the Language of Social Media: The Open-Vocabulary Approach” for the complete methodology and findings.

October 17, 2012

Not sure I endorse LinkedIn’s new “endorsements” feature

Over the past few weeks, you’ve probably noticed a few emails about people “endorsing” you on LinkedIn.  I got a few of them too, so I thought I’d check it out. I logged into my account and viewed who has endorsed me, and started endorsing people.

LinkedIn’s blog announcement says “We are introducing Endorsements, a new feature that makes it easier to recognize them for their skills and expertise. With just one click, you can now endorse your connections for a skill they’ve listed on their profile or recommend one they haven’t added yet. Think your connection is great at programming AND project management? Let them know!”

And with this, LinkedIn is trying to join the game of online influence. Essentially, they’ve created their own version of Klout, or PeerIndex, or Kred.  Just take a quick look at the profiles on these services below…

…LinkedIn “endorsements”…

 

…kind of looks like…Klout Topics…

…which kind of looks like…the Kred Dashboard…

….which kind of looks like…PeerIndex…

In a world saturated with “topic influence” services, I fear that this isn’t the right move for LinkedIn. Besides competing with an already crowded marketplace, the “endorsements” feature has the potential to harm the LinkedIn experience itself by detracting from the “recommendations” feature. In short, I think people will opt to endorse somebody rather then recommend them, and here’s why that matters: “recommendations” are written explanations about why an employee is valued. They usually refer to specific  experiences, work projects, or professional attributes, and the relationship of the author to the person they are recommending is disclosed. An “endorsement”, on the other hand, is a quick “+1″ on the user’s self-proclaimed skills. One can simply scroll down a user’s profile and click on each of their skills. No explanation, no disclosure of relationship, no details.  This quick “+1″ ability may also leave the door open to “I’ll endorse you if you endorse me back”  behaviour.

Trying to measure online influence isn’t a bad thing. In the past, I must admit, I’ve tried to boost my Klout score. I made sure my content was focused, interacted with others regularly, pushed out a lot of updates, etc. But there have also been times when I just didn’t care about it and I completely ignored my social media accounts. Neither of these actions seemed to have influenced my score much, so I can’t see myself continuing to be active in the “endorsement” game.

The Project Management LinkedIn group has a decent discussion this week about “endorsements”. A few good points were made:

What do you think? Are “endorsements” a welcome way to give kudos to colleagues, or a wasteful exercise in meaningless rewards?

May 1, 2012

Want to know if you’re an online jerk? There’s an app for that! [Interview]

Tom Scott – Creator of Klouchebag

By now, I’m sure you’ve heard of Klout.com, a service that attempts to measure how influential your social media profiles are. People are scored out of 100 by an algorithm and are assigned influence ratings on topics. They can also compete for titles and rewards. It’s the gamification of influence online. Some people love it, some hate it. Last week one Londoner decided to poke fun at it by creating his own service designed to measure your online “asshattery“.

This new parody site, named Klouchebag.com, was created in a few hours by Tom Scott  (@TomScott) – a British geek comedian, programmer, and social media contributor.

Klouchebag ranks Twitter accounts by evaluating online activity using the ARSE system:

  • Anger: use of profanity and rage.
  • Retweets: “please RT”s, no or constant retweeting, and old-style.
  • Social Apps: sharing every useless check-in on Foursquare or its horrible brethren; and
  • English Usage: if you use EXCLAMATION MARKS OMG!!!, or no capitals at all, this’ll be quite high.

Getting a good chuckle at this site, and learning that I’m “Quite Noisy“, I shot Tom a quick email to get some more background on his latest creation. Here’s what he said about Klouchebag….

What spawned this idea and how fast did it come together?  
I had the idea on April 26th, 2012, after reading this article in Wired. I’d been annoyed with the idea of Klout for a while, and that crystallised it. On April 27th, 2012, I registered the domain name (sadly, “klunt.com” was already taken) and built it in a couple of hours of spare time.

What type of feedback have you been getting so far?
Almost all positive – fortunately no-one seems to be taking it seriously!

How much traffic did you get on the first day?
No idea. The stats won’t be in for a while. It’s certainly the fastest-launching project I’ve ever had.

Is there anything that you left out that you would have liked to have included?
Given another few hours, I might have added some awards or badges – but I’d worry about people competing for them!

Who, besides yourself, has the highest score?
Someone did discover one natural 100, which I didn’t think would be possible – a US morning radio show! For their sake, I’ll keep quiet as to who it is.

After a little Googling, I found this…

Do you have an axe to grind with Klout? or could this have been any “influence rating” service? 
Klout annoys me for the same reason that search engine optimisation annoys me: it’s an enormous amount of effort designed to game an arbitrary and often-changing system. Imagine if all that time went into actually making interesting things, or caring about the people around you. To quote the WOPR computer from WarGames: “the only way to win is not to play”!

Social media in a business context is all about measurement. If influence isn’t a viable measure for ROI, how would you suggest evaluating the effectiveness of social media campaigns? 
I’m not sure I can actually answer that question without throwing up a bit in my mouth.

Now what do you think? Do  you care about your Klout score? Do you try and improve it?

Follow

Get every new post delivered to your Inbox.

Join 81 other followers

%d bloggers like this: