Gardner and Murnane Debate at Spencer Foundation Event

Harvard Graduate School of Education Professors Howard Gardner and Richard Murnane took center stage on Saturday, May 1, at the Spencer Foundation's annual convening at the AERA conference. Held in Denver's Marriott City Center Hotel, the event brought together several hundred scholars and Spencer affiliates to watch Gardner and Murnane debate the uses and limits of data in improving education. The debate, which followed the presentation of Spencer's Exemplary Dissertation Awards, was moderated by New York Times columnist David Leonhardt.

After introducing Gardner and Murnane, Leonhardt invited each scholar to deliver his prepared remarks. Gardner spoke first about the limits of using data in education. He criticized the current climate of testing in America for confusing figure and ground. In focusing our attention on testing and data, Gardner argued, we've lost sight of the society in which we want to live and for which we want to prepare our children to become contributing members. He said he worries about the messages we send children when we place such weight on tests and rankings and neglect hard to measure areas such as respect, civility, ethics, and the arts.

In contrast to Gardner, who focused on the society in which we want to live, Murnane spoke about the society in which we currently live. He pointed out that our current education system does not serve all students equally well. As a result, the socio-economic mobility in America remains extremely low relative to other developed countries. Data are needed, he argued, to shine a light on these inequities and to track our progress as we try to overcome them.

Leonhardt drew on these opening remarks as he moderated the ensuing discussion, in which the following two questions emerged as central points of focus: When are data helpful and when are they not? If we don't use data, how are we to know if students are learning?

With respect to the first question, Murnane pointed out that the usefulness of data depends entirely on the uses to which they are put. For instance, data may show that the students who pass through one teacher's class consistently score lower on state achievement tests than the students in another teacher's class. Murnane said he would oppose using such data to make decisions about teacher pay or termination. Instead, he argued, the root cause of the disparity in student achievement should be probed. Perhaps the teacher with low-scoring students is not receiving support from her colleagues, or perhaps struggling students are disproportionately placed in her class. Once the source has been identified, steps can be taken to provide the necessary supports to improve teacher practice and student learning.

Perhaps surprising some members of the audience, Gardner avowed that he has nothing against assessments. Assessments of progress and achievement might actually be quite useful, he said, provided they are valid and do not rely too heavily on only one format. To improve validity, he suggested that states employ different assessment formats of important skills each year in such a way that it becomes impossible for teachers and students to "game the system."

Gardner agreed with Murnane that the uses to which data are put matter a great deal. He saluted Murnane for the work he does to help teachers and school administrators make sense of and use data to inform their practice. Gardner said his primary objection to using data in education lies in the league mentality to which "test mania" gives rise. He worries that the sorting and ranking of students and schools sends a disturbing message to students about what is and is not valued by society.

Leonhardt said he understands Gardner's aversion to the league mentality. Yet, he finds it difficult to see how standards can be established and maintained without some form of measurement. Without a set of agreed upon standards, wouldn't everyone simply claim to be the best?

Gardner addressed this concern by calling for a broadening of how we define and where we look for evidence of learning. In the community of Reggio Emilia in Italy, for example, schools document student learning in a variety of ways, none of which involves administering multiple choice tests. As well, the air of respect and high levels of student engagement are palpable in these schools and should be counted as valid measures of success. Most important, Gardner commented, "The community works."

Murnane said he supports the idea of broadening our measures of success. In fact, he would like to see a waiver process associated with the No Child Left Behind (NCLB) law's testing mandate. Under this amendment to NCLB, schools that develop rigorous alternative forms of student evaluation could appeal to an outside agency for approval to use their homegrown assessments in place of state assessments. Yet, until all schools develop the capacity to develop such alternatives to state tests, Murnane maintained that we must not discard the imperfect measures we have now; they are needed to keep the light shining on the inequities that pervade our current education system.

In the end, the audience was the clear winner of the evening's debate. We were treated to a skillfully moderated discussion that engaged three fine minds in a conversation about a timely and critically important topic in education.

Katie Davis, Ed.M.'02, Ed.M.'09, is a HGSE doctoral student in Human Development and Education.