Psychology professor's research looks into solving word problems

Assistant Professor of Psychology Allison Butler explores modifying word problems to improve performance of low income students.

Assistant Psychology Professor Allison Butler's recent research underscores the need to identify new and innovative ways to improve the math performance of children from lower socioeconomic backgrounds.

This was part of her findings in "The Problem with Word Problems: The Impact of Problem Format on Accuracy, Self-Efficacy, and Algorithmic Strategy Use."

"It is critical that we avoid falling into the trap of a 'one-sized-fits-all' curriculum when today's learners are so diverse in terms of their backgrounds and experiences," says Butler, who spent three years earlier in her career teaching fifth grade.

"I encourage school teachers to be teacher-scholars, consider their unique student population, and be well-versed in math education research when determining the best ways to facilitate learning for their students," she adds.

Butler's research also shows that using algorithms -- well-defined sets of instructions to solve a problem -- are critical in helping students get the right solution to math questions.

"While it may be important conceptually for students to have experience with alternative problem-solving strategies -- such as drawing pictures or diagrams -- the step-by-step algorithms seem to be the most reliable in terms of leading to students 'correct responses,' she says.

'Strengths-based view'

As an elementary school teacher, Butler says her favorite subject to teach was math. She remembers how much her students dreaded word problems, and she was always looking for new techniques that might help. This curiosity eventually led to her pursuing a Ph.D. in applied development and educational psychology, which she earned before coming to Bryant in 2008.

"I wanted to learn more about how children learn to solve math problems, and I wanted to contribute to the research base that informs current and effective teaching practices," she says.

As part of her research, which she says rejects the "deficit model" of low-income children's cognition and instead promotes a "strengths-based view," Butler created a 16-problem math test to identify the types of word problems that students perform better on. She included word problems in four categories:

Everyday activity: You buy 16 cupcakes at your school bake sale for a quarter each. How much did you spend?
Social-cognitive: John thinks he will get better grades in math if he is the teacher's pet. If he spends 15 minutes each day helping the teacher, after four days how many minutes will he have helped the teacher?
Traditional textbook: There were 15 bundles of sticks and each bundle has 7 sticks. How many sticks are there?
Symbolic: 14x7 =?

In the test, which she administered to 68 low-income and 98 middle-income fifth graders, Butler uncovered that low socioeconomic students perform better on social-cognitive problems as opposed to traditional textbook questions. That is the case, she reasons, because social-cognitive type word problems capitalize on the strength that low-income students have in social reasoning.

"Finding practical and creative ways to capitalize on students' particular strengths in the classroom may prove to be valuable in improving the academic achievement of children from lower socioeconomic backgrounds who continue to lag behind their more socioeconomically advantaged peers in math problem-solving," she says.

Faculty student collaboration

During the fall of 2009, Butler began working with one of her students, Brynn Woodland '10 (Orleans, MA), to code data about students' problem solving strategies. They used statistical software to enter the data and then analyzed it together. Last summer, Woodland joined Butler to present their findings at the American Psychological Association conference in Toronto. They also took part in Bryant's first College of Arts and Sciences colloquium during the fall semester. Butler also presented their research at the 4th Faculty Research Day in January.

"I consider Brynn to be more of a co-investigator than research assistant," says Butler. "I have been continually impressed by her engagement in all aspects of the research process. She has a tremendous work ethic, is a sharp thinker, and has always demonstrated enthusiasm about the project."

Woodland, an applied psychology major, plans to pursue a master's degree in social work after graduating in May. She has not yet decided which school she will attend, but she has already been accepted to Boston College, Boston University, and Simmons College.

She says the chance to work so closely with a professor is a hallmark of the Bryant experience and has given her a leg up as she continues her education and research at the graduate school level. "Being an undergraduate student, I was nervous about presenting alongside veteran professionals," says Woodland, of her experience at the APA conference. "Professor Butler answered all of my questions, and she was a great mentor for my first psychological conference."

Currently Butler and Woodland are working on a journal article that they hope will be published in a peer-reviewed education journal that is read by math teachers. "All of the work I have done with Professor Butler," says Woodland, "will help me succeed as I continue my studies and begin my career."