How Learning Analytics Will Forever Change Education zite.to/1r5eEW3
Predicting Dropout Student: An Application of Data Mining Methods in an Online Education Program zite.to/1d4t0yb
Machine Learning for Automatic Grading zite.to/1fy3MKv
Semi-automatic method for grading a million homework assignments zite.to/GHISdf
How Learning Analytics Are Being Used In Education zite.to/W5ri3R
SPSS software used to develop performanace models in grades K-12
As students head back to their pencils and books, school districts in several states are turning to predictive analytic tools to meet the data aggregation and analysis requirements of the No Child Left Behind Act and to focus teacher efforts on boosting academic performance.
Research firms Analytic Focus LLC and Reveal Technologies LLC have partnered with Chicago-based SPSS Inc. to help develop models that can predict student performance in grades K-12 based on current instructional methods used in a school. School districts in New York, Colorado, Minnesota, Alabama and Iowa will put in place the SPSS predictive analytics tools and participate in this new program, according to an SPSS announcement.
In addition, the Naperville, Ill., school district, located in the suburbs of Chicago, this summer has been training principals in its 21 schools how to use SPSS’s predictive analytics software, said Alan Leis, superintendent of the Naperville Community Unit School District 203.
“No Child Left Behind [legislation] forces us to focus on individual student data… and large groups by schools,” Leis said. “[SPSS] will allow us to see which students are on a normal growth path and which students are below it… and to predict which students are most at risk for not meeting achievement standards.”
The district began working with SPSS last year to build a master data warehouse that could pull together data from disparate databases containing test scores, demographic data and other information needed for predictive analysis, Leis added. This school year, the district will begin using the software to analyze data and build growth plans for schools and the district’s 19,000 students. The software will replace the time-consuming process of manually analyzing data from test score spreadsheets, Leis added.
“Now we can give [users] a CD with all this data on it so they can do the what-if analysis,” he said. “It allows you to not spend all this time figuring out the data but… figuring out what you did right and what you need to do better.”
Leis said he hopes to eventually expand the use of the software to the district’s 1,200 teachers.
Phil Ashworth, coordinator of testing data for the Hamilton County school district in Chattanooga, Tenn., said he has been using SPSS predictive analytics software for several years to analyze testing data. A year ago, he added SPSS’s Clementine data mining tool to the mix to provide a graphical representation of test scores from the district’s 40,000 students.
The tool allows him to set up the parameters for analysis and to run a report and apply those parameters to any of the 80 schools in the district without having to rewrite any of the instructions, he said. In addition, while testing is commonly done in the spring, it is at the beginning of the school year that teachers need to know their students’ strengths and weaknesses, Ashworth said. The SPSS tools allow him to provide testing data to each teacher at the beginning of the year, he said.