Located southwest of New Orleans along the banks of Bayou Lafourche, Lafourche Parish Public Schools serves nearly 15,000 students in 30 schools and has approximately 2,400 employees. This school district is a part of Acadiana, or French Louisiana, which is known for its Cajun ancestry and culture.
Reflecting a trend that is occurring in other school systems around the country, the administrators at Lafourche Parish Public Schools have been facing a rise in the number of student disciplinary referrals. At the end of the first semester in 2006, Lafourche administrators had already recorded 17,647 disciplinary cases. By March 2006, that total increased to 26,000. This growing concern prompted school officials to take a much closer look at how they addressed student disciplinary issues.
Using SPSS Inc.'s predictive analytics software, administrators conducted research to see if the programs currently in place effected the desired changes in students' behavior. Officials also wanted to see if they could uncover patterns from previous disciplinary incidents to ensure that similar future occurrences would be dealt with in a consistent and appropriate manner.
Our predictive analytics software enables Lafourche Parish Public Schools to address both of those needs.
"With SPSS Inc.'s software, I'm able to dig deep into the data I have and categorize student behaviors by various actions associated with them,” explained Chris Bowman, supervisor of federal programs at Lafourche Parish Public Schools. “By uncovering trends and patterns of students who are referred for discipline—for example, the student who dressed a particular way or who used profanity on a regular basis or was more prone to getting into trouble in a particular manner—I can provide critical information that helps us better execute the program and, more importantly, better address the needs of the affected student. I'm not saying we can change the behavior of every student, but we can certainly do a better job of correcting it."
Interested in Lafourche Parish Public Schools? Download the PDF
Lafourche administrators relied on our text mining and data mining software to pinpoint patterns of student disciplinary problems and categorize the results. For the first time, Lafourche Parish Public Schools was able to extract key concepts and code responses with accuracy and consistency, as well as without prejudice or forethought.
What I like most about using the computer to do this work is the fact that it can handle mindless, boring, and repetitive tasks better than I can. . . . After PASW Text Analysis for Surveys performs its automated tasks, I have an opportunity to do what I do best. I move into the inquisitive, intuitive, and innovative realm of interpretation.
Chris Bowman
Supervisor of federal programs
Lafourche Parish Public Schools
"We had never attempted to quantify free form text fields before because it was beyond what a human being could be expected to do," said Bowman. "With SPSS Inc. text analysis software, rule associations, cluster analysis, and other data mining models are able to reveal patterns, some of which confirm previous assumptions based on anecdotal evidence, while others disprove our conventional wisdom. Our initial findings have already shown us that many of our past decisions were made in spite of data, not based on it."
For example, Bowman discovered that there is no one day of the week that is more likely to have an increase in behavior-related problems—although, he said, it has often been suggested otherwise.
After Lafourche Parish Public Schools saw the advantages that our predictive analytics solutions provided, they decided to use them for other analyses. During the summer of 2005, administrators had used our text mining software, PASW Text Analysis for Surveys, to investigate employee attitudes. That project took Bowman no more than five hours to code the responses by himself—without any training. Compare that to their experience with a previous survey: without our text analysis solution, it took 20 people six hours to code the 400 responses.
"What I found interesting was that in using PASW Text Analysis for Surveys, I came up with different conclusions about various issues like teacher pay, paperwork, discipline, etc. Another thing that I noticed was that I could now compare categories to categories, which gave me an opportunity to see the relationships among them," Bowman said.
In many cases, Bowman added, "manual coding" is completely out of the question. "By the time you have finished reading two, 10 or 20 responses, it all seems the same. What I like most about using the computer to do this work is the fact that it can handle mindless, boring, and repetitive tasks better than I can. With PASW Text Analysisfor Surveys, I can extract concepts and code responses reliably and consistently, and without prejudice or forethought. After PASW Text Analysis for Surveys performs its automated tasks, I have an opportunity to do what I do best. I move into the inquisitive, intuitive, and innovative realm of interpretation. That can’t be done without the program."
In the future, Bowman plans on conducting a closer examination of in-school suspensions. He wants to determine, through predictive analytics, if they really work for Lafourche Parish Public Schools. If not, he believes there has to be a better alternative; and he will recommend finding it.
Bowman would also like to use predictive analytics to examine the effectiveness of the Positive Action School Site (PASS), which Lafourche Parish Public Schools facilitates for suspended or expelled/excluded students. PASS is designed to continue the educational process in the general curriculum at an alternative school site. Again, Bowman wants to see if it's truly working. If not, he wants to find out the reasons why and help effect the needed changes.
Worldwide Offices | Communications | Site Map | Contact Us | Terms and Conditions | Privacy Policy
© 2010 SPSS Inc., an IBM Company All rights reserved.
SPSS Inc., an IBM Company Headquarters, 233 S. Wacker Drive, 11th floor
