Laura Lanwermeyer lists items on the board — baby carrots, string, cornstarch, sugar, salt, water, plastic bags, baking soda, ice.
"Now use any of these items to design an experiment to show osmosis," she tells her class of 9th-grade biology students at Selinsgrove Area High School in Pennslyvania.
The students have been introduced to the concept of osmosis — water moves from solutions of low concentration to solutions of high concentration — but Lanwermeyer hasn't told them how to show osmosis itself. As the students set to work planning their experiments, it's clear they have no idea what they are doing. A few discuss the possible uses of string. Others debate which "chemicals" cause water transfer. Lanwermeyer doesn't intervene, other than to ask a question here and there.
Lanwermeyer's teaching style, called discovery learning by experts, is one of the hottest trends in education theory. Discovery learning teachers allow students to discover scientific concepts on their own through trial and error, debate and reflection. "The point isn't to get the right answer," she said, "but to reflect on the process of learning." This is in contrast to direct instruction, wherein the teacher explains in detail all topics and activities for students.
Discovery learning is part of a growing trend stretching across the country. Its principles inform the objectives of the new Common Core State Standards Initiative, a new standards-based curriculum for students from elementary school through high school. The Common Core, which emphasizes experimentation and problem-solving skills, will be implemented in 45 states across the country in August. While there is no doubt about the rising popularity of discovery learning, there is considerable debate about its effectiveness.
Expert opinions run the gamut. Initial assessments research shows direct, explicit instruction is more effective and efficient than discovery learning. On the other hand it appears that in the long-term, discovery learners have better understanding, retention and ability to transfer knowledge, said Deanna Kuhn, professor of education and psychology at Columbia University's Teachers College.
"How we instruct students is important for achieving educational goals," said Chipper Dean, assistant professor of psychology at Bucknell University.
If educators and parents want students who can quickly master concepts, pass tests and move on to the next thing, then direct instruction works well, said Dean. On the other hand, if the goal is to equip students with the ability to problem solve, to think critically and to work collaboratively, then discovery learning really works. Parents, teachers and policymakers need to think about which skills students need to be sucessful in today's information-based global economy.
Direct vs discovery
To better understand direct and discovery learning, consider the differences in the presentation of a math lesson on adding decimals.
A teacher using direct instruction begins the lesson with a problem: 12.1 + 3.3 = ? Then she will break the problem into smaller steps. First, write the problem so the decimals line up. Second, bring the decimal point straight down into the answer. Third, add the numbers beginning on the right and moving to the left. Teachers model several problems like this, and then allow students to practice until they've mastered the task.
A teacher trying to encourage discovery learning also begins the lesson with the problem, 12.1 + 3.3 = ? But in this case, instead of immediately showing the students how to solve it, she asks them to brainstorm possible solutions in small groups. She'll provide few, if any, prompts to guide thinking. Then the class reconvenes to debate the solutions. It may be some time before the teacher indicates which solution is correct. The crucial idea is that through the process of trying to solve 12.1 + 3.3 and discussing possible solutions, the students discover the relevant mathematics.
Where direct instruction is to the point and efficient, discovery learning can be oblique and time-consuming. But which produces better academic outcomes?
One of the most commonly cited studies by supporters of direct instruction is a 2004 report out of Carnegie Mellon University. Researchers considered which teaching style, direct instruction or discovery learning, produces better outcomes among 3rd and 4th grade students.
They found dramatic differences in the performance of the two groups of students. About 75 percent of direct instruction learners earned mastery level performance, whereas only 23 percent of discovery learners earned mastery level.
"These results challenge predictions derived from the presumed superiority of discovery approaches in teaching young children basic procedures for early scientific investigations," wrote David Klahr, professor of psychology at Carnegie Mellon University and principal researcher of the report.
Though Klahr considered his work definitive proof of the superiority of direct instruction, others were skeptical. Dean, then a graduate student at Teachers College, wondered if Klahr's findings held over time. Using a similar research design and test population, Dean set out to examine how recipients of discovery learning and direct instruction performed on tests over a period of 17 weeks.
Like Klahr, Dean found that on the initial assessment, students who received direct instruction outperformed their discovery learning peers by considerable margins. Over time, however, he noticed a significant shift. Testing students 11 weeks after the initial assessment, Dean found that no recipients of direct instruction achieved mastery level, while just under 10 percent of discovery learners did.
He tested the students on three subsequent occasions, the last test 17 weeks after the initial assessment. By that point almost 50 percent of the students who received discovery-based instruction achieved mastery level. By contrast, just more than 10 percent of the recipients of direct instruction earned the same score.
For Dean, the implications of his findings were clear: In the long-term, discovery learning pays. "Although in the short term discovery learning appears to be inefficient, it is very efficient if we take a long view," Dean said. "We retain information we work with," he said. "Discovery learning provides students with an opportunity to work with information."
Lanwermeyer, the Pennsylvania high school biology teacher, agrees. Discovery learning works because it provides students with an opportunity to critically engage in problems, she said.
Room for debate
Even assuming that discovery learning is the more effective of the two teaching styles, there are significant barriers to its implementation in the traditional school system. Most prominent are concerns about covering the required course content in the allotted school time.
"We can take a week on a single experiment," Lanwermeyer said of her 9th-grade biology class.
"With current school structures and curricula, many times it is impossible to allow the time needed for discovery learning," said Roger Schank, an education advocate and CEO of Socratic Arts, a company that designs educational lesson plans. This is especially problematic when teachers are held accountable for the curriculum they teach based on student test scores.
Still, advocates of discovery learning say the method works.
"Even if we don't cover all the curriculum, the kids know the things we've covered better," said Britnie Powell, a Salt Lake City-based 6th-grade teacher who uses a discovery learning approach.
"It means they have a better chance of getting questions that deal with that material right," she said. According to Powell, the proof is in the pudding. Despite not always covering all the material in the curriculum, her students considently score well above average on their state exams.