Review of Feifer Assessment of Reading and Math

Introduction:

The Feifer Assessment of Reading and the Feifer Assessment of Math are tests authored by Steven Feifer.

For use in states and school districts still using the historical intelligence/achievement discrepancy methodology to identify children with a specific learning disability, these tests depart from standard practice for nationally normed tests by offering only grade-based norms with age approximations.  Also, there are no monthly, trimester, or semester norms.  The standard score for any one raw score remains the same throughout the entire grade, September through June.

Achievement scores based on grade norms cannot validly be compared to ability scores based on age norms.  (Also see “Age Norms vs. Grade Norms.”) Although the FAR and FAM are aimed in their  advertising at schools using a response to intervention methodology to identify SLD, they have an advertised average administration time of sixty and fifty minutes.  This in this reviewer’s opinion  makes them ineffective from a time/cost effective perspective for meeting the 2006 Part B Regulations’ (Section 309) requirement for repeated assessments of achievement at reasonable intervals.  These requirements are more readily met by using normed curriculum based reading and math probes, e.g., AIMSweb products..  Another limitation with respect to using the tests as a pre-post assessment is that there is only one form of the full test and only one form of the screener.

The identification of a child as having a learning disability regardless of methodology requires schools to provide:

Data-based documentation of
repeated assessments of achievement at
reasonable intervals, reflecting formal
assessment of student progress during
instruction, which was provided to the
child’s parents  (Section .300,309 2006 Part B Regulations)

Providing data-based documentation of repeated assessments,  however, was and is not the purpose of these tests.  With respect to the FAR, for example, the test was

“Designed to examine the underlying cognitive and linguistic processes that support proficient reading skills … [to help] users determine not only the presence of a reading disorder but also the specific dyslexia subtype.”

In this reviewer’s opinion, that might make the FAR more relevant in states that required screening and evaluating for dyslexia than for assessing students for SLD under the IDEA.

At the time of this editor’s review (June, 2018), credible independent reviews had not yet been published on the Internet, nor were they available from Buros.  (When they become available in the 21st edition in 2020,  the cost per review will be $15.)

The lack of independent critiques supporting or criticizing the FAR/FAM in 2018 could be a significant factor in an adversarial proceeding where the school’s decisions regarding services were largely based on the data gleaned from the FAR/FAM when the parent experts were relying on more extensively documented tests such as the WJ IV, the WIAT III, the KTEA 3 or  in an RTI environment on normed probes such as AIMSweb..  ( (Reliability and validity studies on CBM probes have been extensive.)

Readers who wanted to conduct their own individual review of this test (or any new test) should consult Ron Dumont’s and John Willis’s Test Information Outline page (found under Test Information).

Administration and Interpretation:

As is always the case, end users are strongly encouraged to read the manual  before attempting to administer these tests or to interpret the test results.

In addition to information provided for end users in the manual, John Willis has   provided a more than 90 slide PowerPoint presentation that provides examiners with additional suggestions regarding test administration and interpretation of the Feifer Assessment of Reading.  For those considering but who have not yet purchased the FAR, the PowerPoint address norms, reliability and validity based on the Professional Manual more extensively than the publisher does on-line.  The information provided was taken directly from a presentation made at a November 29, 2016 workshop.   Some of the slides at that workshop were removed due to copyright concerns.  Nevertheless, readers needing additional information for citation purposes are referred to the FAR Registration which can be accessed by clicking on the preceding link.  As noted on the form itself, the workshop was NOT an official PAR training..

The three links provided below are all to the same content.  Only the program required to open them is different.

Some Statistical Information Regarding the Feifer Assessment of Reading (FAR) by Steven G. Feifer (Lutz, FL: PAR, 2015).

Some Statistical Information Regarding the FAR  (PowerPoint Presentation)

Some Statistical Information Regarding the FAR   (PDF File)

Some Statistical Information Regarding the FAR  (Word File)

(Elaine Holden also contributed)

Comparisons

Technical Manual 

As of July 1, 2020, neither technical manual was available as a free PDF download.   The FAR and FAM manuals were available from the publisher for a total cost in excess of $200.   RiversideInsight previously made its Technical manual for the WJ IV available as a free PDF download but deleted the link on or about June 22, 2020.  There were no listings for the CD or technical manual on their website (although the CD had been available previously on Amazon.)  The manual continued to be included with the test kit itself. If someone was interested enough to spend approximately $150 on the manual before buying the kit, ordering information could be obtained by contacting RiversideInsight directly.

Normative Sample:

FAR  1,074 examinees in prekindergarten to college were tested.
WJ IV:   7416 examinees preschool through college

Reliability

FAR:  “The FAR subtests have median reliability coefficients that range from .67 to .95, the majority
of which are in the upper .80s and .90s, indicating an overall high degree of internal
consistency”
WJ IV:  Broad Reading (Mean) .97  Basic Reading Skills:  .95   Reading Comprehension:  .93  (Note:  The Technical Manual also provides a breakdown of composite reliabilities by age; at some ages, the reliability of course is higher, at others, lower.)

Validity:

FAR:   “Predictive validity analysis of learning disability demonstrated a significant discriminant
function, with an overall classification rate of 96.7%.”  No concurrent reliability studies were reported by PAR.
WJ IV:  Concurrent validity coefficients were reported with the KTEA-II and the WIAT III.  Correlations with the KTEA II reading tests ranged from .65 to .83

PAR has a section on reliability, validity and the normative sample for the FAM on its web page.  The normative sample for the FAM consisted of only 1061 examinees from prekindergarten through college according to PAR.  Internal reliability as reported for the Indexes appeared satisfactory,  However, no concurrent or predictive or, for that matter, any type of validity studies were reported.

Theoretical Foundation for the FAR;

Dr. Feifer’s test purports to provide end users with a way to identify one of four subtypes of dyslexia through the index scores.  They are

• Dysphonetic dyslexia
• Surface dyslexia
• Mixed dyslexia
• Reading comprehension deficits

Some brief (and oversimplified) online definitions of the first three types follow:

Dysphonetic dyslexia:  “Dysphonetic Dyslexia or Auditory Dyslexia:is when a learner struggles with the decoding and or spelling of words because he or she has great difficulty associating sounds with symbols (also known as phonemic awareness).  These learners tend to have good visual processing skills, but they have deficits in auditory processing as well as linking a sound to a visual cue..”  (Learning Specialist and Learning Materials)

Surface dyslexia:  “patients with this kind of disorder cannot recognize a word as a whole due to the damage of the left parietal or temporal lobe. Individuals with surface dyslexia are unable to recognize a word as a whole word and retrieve its pronunciation from memory. Rather, individuals with surface dyslexia rely on pronunciation rules. Thus, patients with this particular type of reading disorder read non-words fluently, like “yatchet”, but struggle with words that defy pronunciation rules (i.e. exception words). For example, a patient with surface dyslexia can correctly read regular words like “mint”, but will err when presented a word that disobeys typical pronunciation rules, like “pint”. Often, semantic knowledge is preserved in individuals with surface dyslexia.  (Wikipedia)

Mixed dyslexia:  If a child can’t read (decode) and spell (encode) words either eidetically or phonetically, this is known as Mixed Dyslexia or Dysphoneidetic Dyslexia. This is the severest form of dyslexia because it involves both types of coding functions.  Dyslexia Institute of America.

Some initial observations:

1.  This is really a three pronged model for dyslexia;  while difficulties in reading comprehension can entitle a child to a diagnosis of specific learning disorder or a classification of specific learning disability, it is not considered a marker in and of itself  for dyslexia.  (For a more detailed definition of dyslexia, see our Dyslexia page.)
2. Reading comprehension aside, there is no consensus as to actual number of variants that may be considered entitling a child to a diagnosis of dyslexia. Understood.org posits five different types.  The Reading Well discussed 9 different types.  Others only posit three.
3.  With respect to the three types addressed by the FAR, consensus among the various experts  on dyslexia as to the value of differentiating between surface and phonological types of dyslexia has not yet been achieved, as is evident from the article referenced below.

David A Kilpatrick, an associate professor with the State University of New York, Cortland, and a school psychologist with 28 years experience (18 full time, 10 two days a week) , concluded:

Conclusions

It seems that the enterprise of diagnosing phonological and surface subtypes of dyslexia lacks empirical support. This notion is common in the neuropsychology literature, yet articles published in that area frequently display no interaction with the broader enterprise of reading research mentioned above. As a result, the conventional dual-route based subtype model does not reflect research advances in the last 20-30 years regarding word-reading development, orthographic skills, and the role (or non-role) of visual skills in reading, nor does it take account of the research literature on word-reading intervention effectiveness. Despite its recent popularity in the field of school psychology, practitioners should not feel the need to establish a dyslexia subtype when evaluating students who struggle in word-level reading.

Note: A fuller discussion of this and related issues will appear in a book chapter by David A. Kilpatrick in
the upcoming fourth edition of Contemporary Intellectual Assessment edited by Dawn Flanagan (2018).

According to the author, the linked article from which the excerpt above was taken was extensively rewritten and is basically an expanded footnote from the chapter in the book below which was scheduled for publication in August, 2018.  The book is also now available for preorder from Guilford Press for $93.50 in hardcover or e-book format.

This paper has also been posted on the NASP Members Exchange.

PAR