David Williamson Shaffer

David Williamson Shaffer

@DWShaffer

Followers504
Following247

Data Philosopher, Quantitative Ethnographer, Learning Scientist Also, dad, author, reader, sailor, musician....

UW Madison
Joined on May 01, 2013
Statistics

We looked inside some of the tweets by @DWShaffer and here's what we found interesting.

Inside 100 Tweets

Time between tweets:
a month
Average replies
1
Average retweets
1
Average likes
4
Tweets with photos
14 / 100
Tweets with videos
0 / 100
Tweets with links
0 / 100

Replying to @PeterT @DWShaffer

Hi Peter! No, this isn’t merely about data types. DWS seeks to harmonise those methodological worldviews

1
1
2
Here's my review https://doi.org/10.18608/jla.2019.61.7 … of Quantitative Ethnography http://www.quantitativeethnography.org  by @DWShaffer — and why I'll be at the first International Conference on QE in Oct http://icqe19.org 

Here's my review https://doi.org/10.18608/jla.2019.61.7 … of Quantitative Ethnography http://www.quantitativeethnography.org  by @DWShaffer — and why I'll be at the first International Conference on QE in Oct http://icqe19.org 

Quoted @sbuckshum

1 month to submit to the inaugural International Conference on Quantitative Ethnography — join the network coming together around deriving culturally contextualised meaning from large human activity datasets http://icqe19.org  #learninganalytics #learningsciences

Not quite a month now, but still time. It's a great chance to people to learn more about Quantitative Ethnography and/or get feedback on your work. Workshops included with conference registration, and there is a doctoral consortium (funded!) as well as an early career workshop.

0
4
6

Seriously, though: this is a really thoughtful review of Quantitative Ethnography. Thanks to @sbuckshum for doing such a great job with it!

0
0
3

All done with #AERA19

0
0
1

Just watched Desperately Seeking Susan with my daughter for the first time. Her reax? “In a world of Garys and Jims, find yourself a Dez.” #successfulParenting #happyDad

0
0
1

End of marathon! Thanks for tuning in.... Now back to our regularly scheduled tweets. /24x PS: Mega-props to @cody_marquart for going from back of a napkin to prototype in 24 hours *while on vacation at a water park with his kids!*

0
0
3

Even more to the point, people working with other models (for whatever reason) can visualize them without creating an arbitrary layout of nodes with link weights indicated by numbers *and* without introducing arbitrary visual artifacts that confound interpretation of results./23

1
0
1

More to the point, since clearly there *will* be some cases where an ordered model is better and some where a co-temporal model is better, now ENA can do both. /22

1
0
1

I think it is an open question, still, as to whether order in this sense matters more than co-temporality. But now we have a good tool to assess that in a direct comparison, rather than comparing two different techniques, with the obvious confounds that introduces. /21

1
0
1

So state transition diagrams, darling of NLP folks, can easily be represented and visualized in a way that lets us quickly model and compare larger number of diagrams. Instead of looking at just the average diagram for a group, we can compare groups, means, individuals, etc./20

1
0
1

So if you're interested in being a beta tester, please let us know! Once you can visualize a directed network, you can use ING to model any set of directed networks (that have similar nodes). /19

1
0
1

What ING does is model *and* visualize an a directed graph correctly. I won't spoil the fun telling you how, but @cody_marquart has a working version of the visualization and the @EpistemicUW crew has code for the ordered model. /18

1
0
2

That is easy to interpret, but the interpretation is actually wrong! A network with all connections A -> B and no B -> A would be the same mathematically in the model as one with all B -> A and no A -> B. So that doesn't really work either. /17

1
0
1

Second, just position A and B so that changes in A -> B and B -> A have the same impact on the position of the network in space -- that is, *show* the difference between A -> B and B -> A but don't have it impact the model. /16

1
0
1

So you can see the problem: If we have separate links A -> B and B -> A, what do we do with the co-registration? We've tried two things in the past. First, just treat A -> and -> A as separate nodes. That works mathematically, but is hard to interpret. /15

1
0
1

The co-registration works because ENA uses an optimization algorithm (developed by my brilliant friend @JeffLinderoth) that positions the nodes of the network graph so that the centroid of the network graph approximates the position of the point in the network space. /14

1
0
2

In this way, we can interpret the space such that networks with (for example) larger x-coordinates have more connections between particular kinds of nodes (the ones further to the positive X side of the space). /13

1
0
1

The key to making this work is that network graphs are co-registered with the space, meaning that changes in the graph (more connections between nodes in the "upper" part of the graph) correlate with changes in position the network space (larger values for the y-coordinate)/12

1
0
1
Next Page