What the National Reading Panel Report Did NOT Say

Whether we agree with some of the summary judgments of the National Reading Panel (NRP) report or not, there is no denying the political and possible curricular impact this report is having on reading education in America. There have been numerous articles and media reports written in support of the NRP, however, one must investigate what was not said and what should not be defended based on the findings of the separate sub-committees. There seem to have been some inconsistencies between the NRP’s findings, the summary document and commercial reading companies contentions.
Looking at the big picture, there were a substantial number of areas of inquiry that were not addressed by the NRP report. Some of these include; effectiveness of decodable and predictable texts, second language issues, motivation and attitudes towards reading, family literacy, reading-writing connections and the role of school and classroom libraries (availability of reading materials). In addition, the reduction of the number of studies actually used in the final report should cause any reader to question their findings. Going by the trendy moniker “scientific-based research” the panel reduced the number of studies reviewed in the report from over 100,000 to approximately 300. That means that less than 3 one-thousandths of the studies available met the standards for inclusion. I have to believe that some good studies were neglected in the name of “science”.

A major concern of the NRP report is that fact that the panel did not define reading as a process, instead it was operationally defined as a set of isolated skills and was thus investigated as such. In other words, the assessments used in many of the studies included defined and measured reading as decoding words in isolation (on word lists), as pronouncing nonsense words from lists, segmenting words into phonemes /c/a/t/, and pronouncing individual letters (/b/ say buh). This is not reading as far as I am concerned. The NRP panel defined phonics as a “means to an end”. In a quote from the report, “reading programs that focus too much on the teaching of letter-sound relations and not enough on putting them to use are unlikely to be very effective.” (p. 10). I wish the panel would have looked at studies that matched their own definition, phonics as a tool for meaning to be used in the context of reading. The studies reported did not measure what they reported to be measuring. The report warns, “there is a need to be cautious to giving blanket endorsement to all kinds of phonics instruction.” I couldn’t agree more.

The report focused on the following categories or aspects of reading: decoding, reading non-sense words, word list reading, spelling and reading short passages and answering multiple choice questions. No single study included in the analysis measured reading based on more than one or two of these aspects or isolated skills. Whether these categories together can be synthesized into “General Reading” as the panel did, is debatable. Although, it may be inferred from the report that children need to develop an understanding of letter-sound correspondence in order to become better readers, how children acquire this understanding and which programs or instructional practices are most effective in developing this understanding could not be determined from the studies reviewed by the panel. The report states that phonics should not be allowed to become the dominant method of instruction based on time or emphasis. Again, I couldn’t agree more. Maybe the report should be renamed the National Decoding and Non-Sense Word List Panel Report.

Another major concern with the final report issued by the panel involves their determinations about comprehension. Cunningham, in his article in Reading Research Quarterly, reveals a weakness in the report of the sub-committee reviewing comprehension studies. The committee did not have enough studies to conduct a meta-analysis, the kind the NRP used in every other category studied. The small number of studies and the variation of the dependent variables included in the research design made it impossible to conduct a meta-analysis. However, this did not stop the panel from ignoring its own guidelines and making suggestions based on the small number of studies available. I find this problematic.

Obviously, I have many challenges with the report itself and how it is being used to guide reading instruction. Decoding is not reading and I am reluctant to agree with many of the inferences that are suggested in the summary document. Do children need to develop letter-sound correspondence? Absolutely. However, decoding should always be used in the context of actual reading to make sense of the texts being read. An overemphasis on decoding as an end in itself can paralyze young readers. I am also concerned about commercial programs (ie. Open Court, SFA) that contend they are supported by the findings of the NRP report. I am concerned about the reduction of available research to what fits a narrow, pre-determined definition of science and its effect on future reading research. The report concludes “By emphasizing ALL of the processes that contribute to growth in reading, teachers will have the best chance of making every child a reader.” I couldn’t agree more.