Saturday, February 14, 2015

Evaluating Error Correction Procedures for Oral Reading

Joseph R. Jenkins and Kathy Larson authored Evaluating Error Correction Procedures for Oral Reading for the University of Illinois at Urbana-Champaign as a technical report in June of 1978. Although this research is three and a half decades old, its results are interesting.

The authors note that when proficient readers have miscues they either disrupt meaning and the reader engages in fix-up strategies or they do not disrupt the meaning and the reader ignores the error. This self-correction is possible because proficient readers have good word recognition skills. Poor readers, on the other hand, may not even notice that meaning is disrupted, in part because they do not expect reading to make sense. They have fewer tools for self-correction and less word identification skills. When teachers are working with oral reading, the most common correction tool they use is word supply. When a student makes an error, the word is corrected and the reading continues. The authors sought to understand how effective this interruption is in increasing word identification skills and if other strategies would be more useful.

The authors had five junior high school students aged 13 or 14 who were reading at least four years below grade level and had been identified as having a learning disability. They received resource room support every day for 50 minutes. The intervention used approximately 35 minutes of that time and was delivered individually. Each treatment method was used for at least seven days for each student.

Material from instructional basal readers, presumably at the individual's grade level was used. In all treatment approaches lists of misread words were kept. Half the day's passage was read under the control condition and half under the experimental condition. The following day the student was asked to read the previous day's list- both as individual words and as words within the context of the original sentence- and data was recorded on the success. Below is a summary of the treatment strategies used and their results.


Results- successful reading rate on the next day of instruction for the list of misread words.
Word Study (WS)
Teacher supplies the correct word after student error. Student repeats the correct response.
Isolation- 33.4-41.1%
Context- 46.9-54.2%
No Supply (NS)
After an attempt, no correction was provided.
Isolation- 36%
Context- 34.9%
Sentence repeat (SR)
Teacher supplies the correct word after student error. Student repeats the correct response. Student is asked to reread the entire sentence again.
Isolation- 49.1%
End of Page review (EPR)
Teacher supplies the correct word after student error. Student repeats the correct response.  Error words printed on a list. At the end of each page, student read the word list.
Isolation- 49.0%
Context- 61.5%
Word Meaning (WM)
Teacher supplies the correct word after student error. Student repeats the correct response. Then teacher asked, “What does this word mean?” If student could not provide an adequate definition, one was provided and the student was asked to repeat it. Then the student took off from where he left off. If the word was again misread, the procedure was repeated. At the end of the page the list of words was presented to be read and definitions of the unknown words were requested. If student could not provide the definition, it was again provided, and the student was asked to repeat it.
Isolation- 57.5%
Context- 67.1%
Drill (D)
Teacher supplies the correct word after student error. Student repeats the correct response. All misread words were recorded on individual index cards. Deck was read, correct words were removed, incorrect words were corrected, student was asked to repeat the word, and the card was put back in the deck. The deck was then shuffled and the process was repeated until the student read through the deck correctly on two consecutive occasions.
Isolation- 77.4%
Context- 84%

For all conditions except no supply the students read better in context than in isolation. Although SR, EPR, WM and D all had significantly better results than the traditional WS or NS conditions, the D condition had vastly superior results to everything else. For teaching word recognition, drill and practice is clearly the best methodology.

In the years since this study, drill and practice has taken a great deal of abuse. It is boring and does not increase engagement. While drill should not be used exclusively during reading instruction, it has a place. We know that students need to have a large sight word vocabulary in order to effectively and efficiently read material. In order to get that vocabulary, students need to a) read widely and extensively and b) have many exposures to individual words. Drill on these sight words is far more effective than strategies that teachers often use, especially at the secondary level, where there appears to be an assumption that sight words have been mastered, even if all our data reveals that only about one-third of eighth grade students are proficient readers.

No comments:

Post a Comment