The annual meeting of the Psychonomic Society will go under way in about 99 hours, give or take 12 hours depending on your time zone. The program has been available for some time, and here I only briefly review the digital aspects of the meeting.
Quite a few of the digital team will be in attendance (myself included) and we will be live tweeting the conference using the hashtag #Psynom15. This is known as “microblogging”, and I have written about this before.
One might think that constant tweeting is distracting, but in my experience it is not: When multiple audience members capture a talk “live” in 140-character chunks, it provides a written interpretative record of what the speaker is saying—opening up the option to check back in time if you have missed something.
The Society has scheduled a social media session for Saturday, 12:00 noon-1:00 p.m., in Joliet, which I will run with the Society’s Executive Director, Lou Shomette. Anyone interested in our digital activities, past, present, and future, is most welcome to attend. My intention is to review and foreshadow our coverage, and seek suggestions and ideas from the audience.
Hope to see you on Saturday at noon!
Digital event the following week
One of the most successful recent digital initiatives was the “digital event” involving the Interface Theory of Perception (you can re-read the week-long sequence of posts from this entry point.) Based on this success, we have scheduled another event for the week after the Psychonomics meeting.
This next digital event will deal with the issue of confidence intervals: Do they really mean what most people think they mean? Can confidence intervals give us confidence about a parameter estimate? This is an intriguing question that has recently gained prominence in the literature.
I don’t want to give away the story, but one intriguing issue surrounding confidence intervals is “coverage”: that is, do they cover the true value of the parameter? If my 95% confidence interval is, say, 50% ± 10%, how often will that interval cover the true value of the parameter?
This is not always an easy question to answer.
There are exceptions, however: My prediction that #Psynom15 will commence in around 99 hours (using one’s local time as the reference time, so considering 4pm wherever you are to be 4pm Thursday in Chicago) can only be off by at most 12 hours either way as there are only 24 time zones. So this is one of those cases where I can be confident of my 100% confidence interval provided I pop this post up within the next 13 minutes.
Stay tuned for more on confidence intervals during the week after next.