Email Status

Volume 17, Number 5
September/October 2001

Solving Problems with ‘Action Research’

A conversation with Pedro Noguera


While teaching at the University of California, Berkeley, Pedro Noguera led The Diversity Project at nearby Berkeley High School, an initiative designed to address the disparity in achievement between white students and students of color and to investigate the causes of racial separation in the school. Using an action research approach, he collaborated with administrators, teachers, students, parents, and other community members to produce findings that Berkeley school officials now use to address inequities. This approach brings research design and implementation directly into schools to tackle what on-site practitioners see as important. HEL assistant editor Michael Sadowski recently spoke with Noguera, now Professor of Communities and Schools at the Harvard Graduate School of Education, about how action research can help schools.

Can you start by giving us a working definition of action research?

I think of action research as research that makes itself directly relevant to practice and policy. That is its goal, to influence either or both of those. Therefore, it needs to be intelligible. It needs to be useful. It needs to be collaborative, whenever possible. And it needs to be driven by the concerns of those who are doing the work, as opposed to by the concerns of the researcher.

Let’s say a group of professionals in a school building or a district—a superintendent, a principal, some teachers—identify a serious systemic problem. They think an action research model might help solve it. How do they begin?

The best place to start is with the data you already have. Schools amass a lot of data related to attendance, grades, test scores, disciplinary issues, [and] data on course enrollment, if it’s a high school. All of that can say something about what’s going on in the school, if it’s [broken down and] analyzed by different categories that are relevant, such as race, geography, or socioeconomic status. You can also collect qualitative data. Focus groups with kids and teachers, surveys, even discussions with parents give you a sense of how people connected to a school perceive the issues in that school.

The next step is to ask the question, “What are the patterns?” You’re going to look for patterns which might tell you something about how well different kids are being served at the school. The data isn’t magic by itself; it doesn’t speak for itself. It needs to be interpreted.

Once you’ve collected, analyzed, and made sense of the data, the next question is, how do you present it and use it as the basis for discussion with that school community? [The reason it’s] important, and why action research is helpful, is that it can provide a way for people to challenge their assumptions about what’s going on in a school.

In particular, I think it’s very important for people to problematize failure, rather than seeing failure as normal. Data can help in doing that. Data also provides a certain amount of detachment, a way for people to engage in a conversation about some really complicated and controversial issues without getting defensive and without personalizing blame.

What about moving beyond the data one already has? What are the next steps educators need to take to design a comprehensive research plan on a particular issue?

The first step is to build a team to do the work. It helps if you have a university partner who’s done research before, who can help with both the collection and the analysis. [It] helps if you can pay for people’s time. Teachers can’t do this on top of their existing schedules. In our project, we bought teacher time, one or two periods, so they would have time to work on the research. We also compensated parents and kids who worked with us. That’s the way you get consistent participation. The collaboration of all those constituencies was important. Our sense was that it’s the process of inquiry, as well as the product of it, that’s transformative. Posing the question, coming up with the answers, and then discussing them—that whole cycle is what leads to new ways of thinking about familiar issues.

How do people in a school district connect with a university researcher?

If you have a college or university near you, [find out] who on the faculty, based upon their background and interests, might be willing to work with the district in a collaborative way on an endeavor like this. The thing to keep in mind, though, is that universities often don’t provide a lot of support for faculty to get involved in this kind of work because it’s not the traditional approach to research. It’s easier if you go after someone who’s a little more established, rather than someone who’s brand new. And it’s easier if it’s someone who understands schools and the issues that schools are going through, rather than someone who primarily views schools as sites for research but is not open to collaborating with practitioners, kids, and parents. It’s a different way of thinking about research.

What are some of the pitfalls of action research?

One issue is denial within the school about what’s going on—or at least a lot of rationalization. Especially in a school where there have been consistent patterns of failure for certain kinds of kids, it’s often the case that people locate the source of that failure in the kids themselves, or in their culture, their community, or their parents. All of this means the school is unwilling to take responsibility for what it can do to address the needs of those kids. Getting people to the point where they’re willing to take some responsibility is an important step.

That’s where the research can play a role in challenging people’s assumptions and getting them to see how they can think differently about why kids succeed or don’t succeed. Some teachers are very willing to accept credit for success—the kids who go to good colleges—but they’re not so willing to take responsibility for the kids who don’t succeed.

How can you make sure things don’t fall apart at the implementation level?

That’s the hardest part. That was the issue that we encountered at Berkeley High. We did the work, we generated good findings, we shared it with the school board. [But] the school itself did not have the capacity to implement the ideas. It had gone through three principals in four years. It was in disarray, from an organizational standpoint. In that kind of environment, it’s very hard to get people to think about things like student achievement and equity because they’re worried about whether or not the bathrooms are going to work, and whether or not they can get copies made. So the learning goals take second place to the survival goals.

Without good leadership to follow through, not a whole lot can happen. This constant turnover of principals at Berkeley meant it took each new principal a year just to figure out the job, much less what should be done. Implementation really depends on institutional capacity and leadership. Do you have leaders who know how to go about the implementation? Do they have the buy-in of their staff, and do they have the capacity, the resources, to pull it off?

Are there any issues that would be tough to address through action research?

Issues around teacher effectiveness need a different form. Those issues are very personal for teachers. That work is very important, but it needs to happen in a setting where people don’t feel as though they’re going to be scrutinized and their weaknesses are going to be used against them. We thought about how to provide teachers with support but do it in a way that’s safe for them. We ended up with an action research project that was all teacher run, where we had teachers actually collecting data on their own work and sharing it with each other. That seemed to work well. So I think there are ways in which you might have to make modifications in the plan [and] format of the research, to take into account certain sensitivities and controversial issues.

Any other advice for educators who are thinking of starting an action research project?

The most difficult part is the public discussion of the research. You want to do that in a way that’s constructive, that doesn’t result in incrimination. It’s very important to think that through ahead of time because the data can sometimes seem to indict the school. How are you going to make sure this is constructive? What are the goals? What are the next steps?