What is Balanced


Contact Us

Developing Research-BasedResources for the Balanced Reading Teacher


Return tothe
BalancedReading.comHome Page



Myth #5 -- Reading involves usingsyntax and semantics cues to "guess" words, and good readers make many"mistakes" as they read authentic text


Both of these claims are quite wrong,but both are surprisingly pervasive in reading instruction (they are especiallyinfluential at the pre-service level).  The idea that good readersuse context cues to guess words in running text comes from a method ofassessment developed by Ken Goodman that he called "miscue analysis" (whichhas given rise to the popular "running records" assessments).  Forhis dissertation, Goodman examined the types of mistakes that young readersmake, and drew inferences about the strategies they employ as they read. He noticed that the children in his studies very often made errors as theyread, but many of these errors did not change the meaning of the text (e.g.misreading "rabbit" as "bunny").  He surmised that the reason mustbe that good readers depend on context to predict upcoming words in passagesof text.  He further suggested that for good readers, these contextcues are so important that the reader only needs to occasionally "sample"from the text (i.e. look at the words on the page) to confirm the predictions. Children who struggle to sound out words, Goodman says, are over-dependingon the letter / word cues, and need to learn to pay more attention to thesemantic and syntactic cues.

Goodman's model, that eventuallygave rise to the "Three Cueing Systems" model of word recognition, is veryinfluential in reading instruction, but unfortunately, it has never beensupported by research evidence.

In fact, repeated studies have shownthat only poor readers depend upon context to try to "guess" words in text-- good readers depend heavily upon the visual information contained inthe words themselves (i.e. the letter / word cues) to quickly and automaticallyidentify the word.  Keith Stanovich has been especially critical ofthe three cueing systems model because the predictions made by the modelare exactly the opposite of what has been observed in research studies.

Philip Gough and I addressed thesecond claim and showed that, in fact, good readers almost never make anymistakes at all when they read, which means the notion of conducting a"miscue analysis" is somewhat suspect -- how can you perform a miscue analysiswhen there are typically no miscues?  We had over 400 students reada passage of text from Ken Goodman's book "Phonics Phacts," and showedthat the modal number of mistakes made by these students was zero -- almostall of the students read the passage flawlessly.  To suggest thatgood readers are correctly guessing the words in the passage with one-hundredpercent accuracy stretched the boundaries of credulity.

However, to be sure, we examinedhow accurate people would be if they were forced to use semantics and contextas their only cues.  We concealed the passage of text and asked studentsto guess each of the words in a passage; after each guess, the correctword was revealed, and students were asked to guess the next word. This process was repeated for every word in the passage, so the studentsalways new the words leading up to the unknown word.  We found that,given unlimited time to ponder, students were able to correctly guess oneout of ten content words in the passage.  That's a ninety percentfailure rate, as opposed to the zero percent failure rate seen in skilledreaders who were not forced to make guesses based on context.

It is clear that good readers dependvery heavily upon the visual information contained in the word for wordidentification (what is commonly called the graphemic information or orthographicinformation).  The semantic and syntactic information are criticalfor comprehension of passages of text, but they do not play an importantrole in decoding or identifying words.  Good readers make virtuallyno mistakes as they read because they have developed extremely effectiveand efficient word identification skills that do not depend upon semantics/contextor syntax.  For good readers, word identification is fast, fluent,and automatic -- it needs to be so that the their attention can be fullyfocused on using semantics and syntax to comprehend the text.


Returnto 10 Myths of Reading Instruction

BalancedReading.com    P. O. Box 300471   Austin, TX  78703

Doyou have comments or questions about this site?

Wouldyou like to contribute material or information to this site?


Last Updated 8-7-03