Discussion about stat consulting, stat packages, applied statistics and data analysis by statistical consultants of Academic Technology Services at UCLA. Please leave your comments or send us email at stattalk at ats.ucla.edu

Tuesday, April 10, 2007

We Feel Like A Million

We have just run the stats on our web server and are pleased to announce that we achieved over one million hits last month. In March, we had 1,003,103 total hits. These million hits are the total numbers of pages hit at our website after we subtract out all of the hits generated in our lab and by our personal office machines.

This was not our first million hit month, we also had one in March of 2006. Since then we have been hovering around 900,000 hits per month.

As of the 1st of April, the total cumulative hits since mid-1999 is just over 37 million. If you look at the results by stat package, the most popular pages are on SAS followed by Stata and then SPSS with the remaining specialty packages trailing behind.

In the last three months the SAS pages had 939,953 hits, Stata 793,129 hits and SPSS with 518,087 hits. For the specialty packages Mplus lead the pack with 66,685 hits in the last three months. The next closest was HLM with 18,473.

One of the more surprising findings was that our limited Splus and R pages managed 39,963 hits since January 1st.

All-in-all not bad results for a website that consists entirely of geeky stat stuff.

pbe

Wednesday, April 4, 2007

Roadtrip

This past Saturday the all the ATS Stat Consultants piled into the Statmobile and went on a roadtrip to the 26th Annual Workshop in Applied Statistics put on by the Southern California Chapter of the American Statistical Association. Fortunately, it was a short drive since the meeting was held on the UCLA campus in the Bradley International Center.

This year's speaker was Bengt Muthén talking on recent developments in statistical analysis with latent variables. The presentation went into how the idea of latent variables captures a wide variety of statistical concepts, including random effects, sources of variation in hierarchical data, frailties, missing data, finite mixtures, latent classes and clusters.

The presentation began a little after 9 am with cross sectional models and finished around 4:30 pm somewhere in longitudinal models. The presentation moved along nicely thanks in part to Professors Muthén's subtle Swedish sense of humor. Although there was no hands-on component the crowd got into the swing of things during the lively question and answer periods. Even with a whole day discussing these topics the material covered was only a fraction of what Professor Muthén usually covers in his five-day workshop.

I'm sure the conference would have run much later but many wanted to get home to see the UCLA-Florida basketball game. Too bad the game didn't go as well as the conference did.

pbe

Contributors