Tag Archives: Dr Tim Rasinski

Assessing Reading Fluency–Word Recognition Automaticity

By Dr. Timothy Rasinski – guest blogger

In my previous blog I presented a simple approach for assessing reading fluencystudents’ word recognition accuracy.      Certainly accuracy in word recognition is important – you can’t read if you can’t decode the words in print.   However, word recognition accuracy is not enough.    Proficient readers are also automatic in their word recognition.    Automaticity refers to the reader’s ability to recognize words in text instantly or effortlessly.     When word recognition becomes automatic readers are able to move their cognitive resources from the word recognition task in reading to the more important comprehension task.

One outcome of automatic word recognition is an increase in reading speed.  As readers become more effortless and instant in their recognition of words, their reading rate will naturally increase as well.   So, reading rate or reading speed has become a simple way to measure word recognition automaticity.    Here’s how you do it:

Use the same reading that the student did for word recognition accuracy.   In addition to marking uncorrected errors, simply mark where the student is at the end of 60 seconds or reading.  Then, determine the number of words that were read correctly in the 60 second period; this is the student’s word recognition automaticity score and is usually stated in terms of words read correctly per minute (WCPM).

Now, because students’ reading fluency rate increases with age, you’ll need to compare their WCPM scores against grade level norms.  Fortunately, these norms are easily available.   Jan Hasbrouck and Gerald Tindal have developed a valid set of automaticity norms.  Below I provide the end of year 50th percentile norms for grades 1 through 8 based on Hasbrouck and Tindal’s norms:

EOY 50%ile Norm                       Threshold for Concern

Grade 1:               53 wcpm                                              40 wcpm

Grade 2:               89 wcpm                                              67 wcpm

Grade 3:               107 wcpm                                           80 wcpm

Grade 4:               123 wcpm                                           92 wcpm

Grade 5:               139 wcpm                                           194 wcpm

Grades 6-8:        150 wcpm                                           112 wcpm

These norms represent normal end of the year performance at each grade level.   So, for example, if you are a third grade teacher, your end of year goal is for your students to be reading at least 107 wcpm on third grade texts.     Scores that are say 25% below these norms should be cause for concern – such students are not developing sufficiently in word recognition automaticity to comprehend texts well.    A complete listing of Hasbrouck and Tindal’s norms can be found at http://www.jhasbrouck.com/ORF2005_dataBRT.pdf

Determining reading rate is a simple and effective way to measure word recognition automaticity.  However, I need to point out that directing students to read faster is NOT the way to teach automaticity.   When students are directed to read faster and faster they lose sight of what reading is all about – making meaning.   We want students to improve their automaticity (and speed) the way that you who are reading this developed your own automaticity and speed – through lots of authentic and engaging reading,   The more we read, the more automatic we become at recognizing words, and our comprehension and reading speed will naturally increase.

Word recognition accuracy and automaticity are critical for reading success.    And being critical, we need to monitor students’ development of these key competencies.  In my next blog, I will discuss assessing one last component of reading fluency – prosody.

reading fluencyDr. Rasinski is a professor of literacy education at Kent State University and director of its award winning reading clinic. He has written over 200 articles and has authored , co-authored or edited over 50 books or curriculum programs on reading education. He is the author of several bestselling books on reading fluency. His research on reading has been cited by the National reading Panel and has been published in many high profile journals.

Assessing Reading Fluency – Word Recognition Accuracy

By Dr. Timothy Rasinski – guest bloggerReading fluency

Reading fluency has been identified by the Common Core Standards as a foundational reading competency.   This means that readers must have a solid foundation in reading fluency before they are able to move on to deeper and more sophisticated levels of reading (e.g. deep reading).   The Common Core indicate that reading fluency is made up essentially of three sub-components – word recognition accuracy, word recognition automaticity, and expressive (prosodic) reading that reflects the meaning of the text being read.   In order to determine how well students are doing in their development of fluency, we need ways to assess the various aspects of their fluency.  In this blog entry, I describe a simple way to assess word recognition accuracy, the first component of fluency.

In order to read one needs to be able to accurately recognize (decode or sound out) the words in the text.   Clearly, if a reader is unable sounding out 1 out of every 10 words in a text comprehension will suffer.     How do we assess this ability in our students?  We’ve all heard about the five finger rule – is on a given page of text a reader has trouble with five or more words, then that text is probably too difficult for the reader.    While intuitively appealing, this rule is rather imprecise.   A page of text can have anywhere from 20 to 100 or more words.  Five problematic words can yield a word recognition accuracy level of 75% to 95%.  That’s a wide range that really does not provide good guidance on the appropriateness of a passage.

There is however a more precise way to assess word recognition accuracy.    Here’s how it goes.   Find a passage of text of about 100 words that has a readability (Lexile) level that matches a student’s assigned grade level.   Ask the student to read this passage orally to you.    As the student reads you follow along and mark any word recognition errors that a student makes that goes uncorrected.  A word recognition error is simply when a student gives a pronunciation for word that does not match exactly the word in print. For example, word recognition errors may include a student reading big for bag, dog for dogs, saw for was.     A word recognition error may also include words omitted by the student or words in which you the teacher have to pronounce the word for the student after they have examined or attempted the word unsuccessfully for a count of three.

Once the student has completed the reading you simply determine the percentage of words that were read accurately.    This can be done by dividing the number or words read accurately (total words in the passage minus word recognition errors) by the total number of words in the passage; and then moving the decimal two places to the right.   For example, suppose a third grade student is asked to read a third grade passage that contains 110 words.  She makes 6 word recognition errors that she does not correct.   To determine the percentage of words read accurately, we divide 104 (110 total words minus 6 errors) by 110 total words.   This can be stated as a fraction: 104/110, a decimal: 94.5, or percentage: 94.5 %.

Now you need to interpret this percentage.      Here’s a good guide to use:

91% or below:   Frustration level.   The reader’s word recognition abilities are not sufficient to successfully read passages at the grade level of the passage read.

92-98%:  Instructional level.  The reader’s word recognition abilities are adequate to successfully read passes at the grade level of the passage read with some instructional assistance.

99% and above:  Independent level.  The reader’s word recognition abilities are at a level that will allow the reader to successfully read passages at the grade level of the passage read without assistance.

Let’s return to the student for whom we earlier calculated her word recognition accuracy percentage.  Her score of 94.5% suggests that she is at her instructional level for grade three material.  As a third grade student then, we can assume she is just where she should be — she is able to read third grade material successfully, but may need some assistance and further instruction in word recognition.

A word recognition accuracy score of 99% for a third grader reading third grade material suggests this student has a relatively strong ability to decode words at her assigned grade level.  On the other hand, a third grade student who scores at say 89% word recognition accuracy is likely to have difficulty in word recognition of third grade material (and higher), and this difficulty is likely to cause difficulty in comprehension of third grade material.   Students who score at 91% or below when reading grade level material are likely to be in need of additional instruction and support in learning to recognize or decode words.

This simple protocol for assessing word recognition can easily be modified to gain more precise information.   For example if a student score poorly on a third grade passage, it is reasonable to check the student out on second or even first grade material to determine where the student’s instructional level is at.   Similarly, for the student who decodes well on the third grade passage, you may wish to have the student read even more challenging material to determine her instructional level for word recognition.

Word recognition accuracy is critical for successful reading.   With this simple method for assessing word recognition accuracy you can easily determine which students are doing well in this competency and which students may need additional intervention in this key skill.   In my future blogs I will share simple approaches to assessing those other two foundational competencies – automaticity and prosody.

Dr. Rasinski is a professor of literacy education at Kent State University and director of its award winning reading clinic. He has written over 200 articles and has authored , co-authored or edited over 50 books or curriculum programs on reading education. He is the author of several bestselling books on reading fluency.His research on reading has been cited by the National reading Panel and has been published in many high profile journals.