We may operate from a place with quite a nice climate, plenty of sunny beaches and all that, but from time to time we take a break and do a bit of actual work.
Here are some of the things we've been up to most recently.
•Released a working prototype of Lertap 5 for Macintosh computers equipped with Excel 2016 (Mac).
This has been a long time coming. It seems that the Mac version of Excel 2016 is finally very (very) nearly on par with the Windows version. Interested? Got a Mac with Excel 2016 installed? Write to email@example.com for instructions on how to get a copy of the prototype.
•Improved performance, now up to version 5.10.9
Re-programmed the computer code so that Lertap 5 has a lot more vim and vigor. Running times have been slashed by 50% to 70% depending on the size of the test/survey being processed. For example, a 60-item cognitive test with responses from 1,800 students now takes about 15 seconds to process with Excel 2016, down from 52 seconds in the previous version (5.10.8).
Lertap 5 is now also smoother when it runs, free of the former screen flashing at times witnessed when running with Excel 2016. Refer to this document for more information.
Users with Excel 2010 (our favorite version of Excel) will experience better performance too -- under 10 seconds to process the 60-item cognitive test (down from 22 seconds).
•Added a paper on using R and RStudio
These two systems, R and RStudio, can receive data and appropriate programming code from Lertap and do wondrous things. This new paper has the scoop.
•More support for IRT users
Rasch analyses? Building on work done by the TAM team, when the Omega1 macro runs it now readily leads to R and Rmd programs designed for use with TAM's Rasch capabilities. See this link to work currently on going.
•Enhanced the Omega1 macro
This macro is designed for use with an omega reliability routine built into the free Psych Package from CRAN. Omega reliability estimates are generally superior to other variance-based reliability estimates, such as Cronbach's alpha.
Now the macro is easier to use, and its graphics output, a plot of factor loadings, is automatically captured as a png file. Read about it here, and enjoy.
•Added a "cleaned quantiles" ability
Have a look at these packed plots when you have a chance. They're new. And, then again, they're really not new: all they are are conventional quantile plots with the trace lines for the distractors removed. The idea is to make it easier to spot trends: does the response trace for an item start low on the left, say below 0.20, and steadily climb, getting over 0.80 or better on the right without taking a dip anywhere? If so we've got a discriminating item and, were we to try and fit some sort of IRT model, we might have cause to be optimistic for an okay fit.
•Data from FIMS now up and ready
Results from the First International Mathematics Study are now available in a Lertap 5-ready workbook. They include student responses from two countries, Japan and Australia. Gender information is also included. Best of all, the actual test items are available. This will be useful in measurement classes. Travel to FIMland with a click here.
•Added two riveting new documents
Results from an international study of numeracy among junior high-school students are discussed in this document with an eye to demonstrating how to process mixed cognitive item types with Lertap 5. Constructed-response items were mixed with multiple-choice items in this test.
A review of selected free (for students) item analysis programs is presented in this document. The programs include jMetrik, Lertap 5, and SAS University. Mention is also made of Iteman, Xcalibre, Bilog-MG, and a promising new offering in R, "TAM".
•Made improvements to the Pearson VUE Exam Developer macro
The number of Lertap 5 users who also make use of Pearson VUE's Exam Developer platform is not quite approaching infinity yet, but the in-built macro continues to make it truly easy to "port" Lertap's item and test statistics over to Exam Developer, especially after a couple of recent modifications and minor enhancements. Read about it.
•Added a new sample dataset with items having multiple responses
Multiple-response items are multiple-choice questions where it's necessary for students to pick out all the correct responses to an item, and there will be more than one. For example, to get item "I1" correct, students must select both options B and D. This is a type of the "multiple-mark" item format discussed in Haladyna & Rodriguez (2013, p.121).
Click here to be transported to the "Zmed" sample dataset.
•IRT with EIRT (<--click here)
"EIRT" originated in 2006 as a counterpart to "RIRT", a package of item response theory routines for use in the R environment, put together by French-speaking colleagues in France and Canada. The E in EIRT stands for Excel -- this is a package designed for use with Excel, and it goes very well with Lertap 5, making it a reasonably straightforward process to have IRT parameter and theta estimates written as another worksheet within a user's Lertap 5 workbook. Very handy; give it a burl. It works tres bien.
•IRT with SAS (<--click here)
The SAS University Edition is a powerful data analysis system, and it's free. A few years ago, 2014, SAS' IRT routines moved from beta testing to production, and they're potent indeed. Now there's support for SAS IRT in Lertap 5 -- take it for a spin.
A link to the complete "updates summary" page is here.