Comments for The Magic Bean Laboratory

Comment on Forecasting Google search volume using R by Josh Dulberger

Josh Dulberger — Thu, 25 Apr 2013 18:06:41 +0000

Nice post, and definitely good to do this in R instead of excel. That said, array formulae do work in mac excel – use command+shift+enter. For an easy example, try the frequency function: create some data, some bins, then highlight the column/rows next to the bins, go to the forumla bar and enter frequency(array, bins), then command+shift+enter. It works on the skeezy workbook too, and I think John is on a mac.

Comment on Can we calculate party affiliation using Twitter networks? (US Congress Edition) by Respondent-driven Sampling | KnowledgeWatch

Respondent-driven Sampling | KnowledgeWatch — Fri, 19 Apr 2013 14:38:40 +0000

[...] there are still some connections between the two clusters from the people who work on both sides. http://blog.magicbeanlab.com/networkanalysis/calculating-party-affiliation-us-congress/ NOTE: that this might not be the best example because a lot of Congressman did not tweet each [...]

Comment on How should Page Admins deal with Flame Wars? by Estimated Follower Accession Charts for Twitter | OUseful.Info, the blog...

Estimated Follower Accession Charts for Twitter | OUseful.Info, the blog... — Fri, 05 Apr 2013 10:32:16 +0000

[...] or so ago, Mat Morrison/@mediaczar introduced me to a visualisation he’d been working on (How should Page Admins deal with Flame Wars?) that I started to refer to as an accession chart (Visualising Activity Around a Twitter Hashtag or [...]

Comment on The Engagement Thing by Digital Marketing Blogs | top10digitalmarketingblogs

Digital Marketing Blogs | top10digitalmarketingblogs — Mon, 01 Apr 2013 20:28:49 +0000

[...] http://blog.magicbeanlab.com/data-viz/the-engagement-thing/ - Appearing independent and described as Data-driven social media insights and opinion. Digital Marketing guru, stats and analytical, research driven findings. [...]

Comment on Forecasting Google search volume using R by Alex Zverev

Alex Zverev — Fri, 08 Mar 2013 19:41:12 +0000

ok, will try

Comment on Forecasting Google search volume using R by Mat Morrison

Mat Morrison — Wed, 06 Mar 2013 11:39:53 +0000

I'm afraid not — I was using the search data as sample data to test the process, rather than for any specific business need. But it should be possible to use data from Google Insights to test the accuracy and fit.

Comment on Forecasting Google search volume using R by Mat Morrison

Mat Morrison — Wed, 06 Mar 2013 11:37:45 +0000

I’m afraid not. But it should be possible to use data from Google Insights to test the accuracy and fit.

Comment on Forecasting Google search volume using R by Alex Zverev

Alex Zverev — Fri, 01 Mar 2013 07:33:38 +0000

do you test this method in real case?

Comment on A perl script to create Twitter friend/follow matrices by Mat Morrison

Mat Morrison — Mon, 25 Feb 2013 12:22:48 +0000

I didn’t really; but in subsequent projects, I did work out that it was cheaper (in API terms) to use the GET followers/ids method. After that, it’s relatively easy to process the data. Here’s the script I came up with. I make no guarantees that it still works…

Comment on Changing to Facebook Timeline has no impact on Engagement by After Timeline, Top 50 Brands See Facebook Comments Dip 11%, Likes Rise 11% | Ignite Social Media

Mon, 25 Feb 2013 05:00:59 +0000

[...] There have been several other studies looking at this already. Some say it has had no impact and that engagement was down for everyone. Others say it lead to a 46% increase in engagement. Others point out that since most engagement takes place in the news feed, engagement should not materially change. [...]