.png)
In order for a Speech-Language Pathologist (SLP) to diagnose a person with a speech and/or
language disorder, they must first administer a formal assessment. There are many different
assessments that Speech-Language Pathologists can use to observe speech and language, but in order for results to be valid, each assessment must contain the proper psychometric properties.
Psychometric properties refer to the components of the test that make it valid and
reliable to properly assess and diagnose speech and/or language.
These components are extremely important to the acceptance and use of the test by
SLPs. Because of this, creators of speech and language assessments must include details about what makes their test acceptable in the assessment handbook. The psychometric properties that any SLP (and parent!) must look for in an assessment handbook include the following:
Clear administration and scoring criteria
Because the results of an assessment can identify a disorder, it is imperative that
the test is administered and scored the way it was originally intended.
Assessment handbooks should include information about how SLPs are to read
questions/present material, if any additional help can be given, and how correct
and incorrect answers or productions should be recorded. This means that
everyone who is administered a particular assessment will be tested and scored
the same way.
Validity
Validity refers to "the degree to which an item is measuring what it’s actually supposed to be measuring." (leadersproject.org)
There are two types of validity that are common among formal assessments, construct and content validity.
Construct validity is how well the test accurately measures what it claims to measure.
Content validity is whether test items measure the intended skills.
Validity: Sensitivity and Specificity
A test's sensitivity and specificity can be used to prove its validity. Both sensitivity and specificity are expressed as either a decimal or a percent.
Sensitivity refers to a test's ability to identify a disorder or delay. For example, a test with a sensitivity of .83 shows that 83% of individuals were correctly diagnosed with a disorder or delay.
The specificity of a test is its ability to identify an individual who does not have a disorder or delay. A test with a specificity of .80 indicates that 80% of normally developing test-takers were identified as not having a disorder or delay.
Reliability
Reliability is the "degree of consistency of measurement in a test. A test has a high degree of reliability if it produces similar results consistently under similar conditions." (leadersproject.org)
The two types of reliability that speech and language assessments utilize are interrater and test-retest reliability.
Interrater reliability includes the opinions of more than one professional. Usually, there is one professional who administers and scores the test while one or more professionals observe and score as well.
Test-retest reliability addresses how consistent a test is by testing the same individual on different occasions and comparing the scores. If a test has good reliability, the scores during a test-retest should be similar.
Reliability: Standardization and Demographics
According to the American Speech-Language-Hearing Association (ASHA), "a standardized test is one that requires all test takers to answer the same items/questions in the same way and that is scored in a standard or consistent way, thus making it possible to compare the relative performance of individuals or groups of individuals." A standardized assessment can either be norm-referenced or criterion-referenced.
Norm-referenced tests are standardized tests designed to compare and rank test takers in relation to one another.
Criterion-referenced tests are standardized tests that measure an individual's performance against a set of predetermined criteria or performance standards.
A very important component of standardized assessments is how inclusive their normative or criterion samples are.
Samples are often based on age, sex, geographic region, race/ethnicity, and primary caregiver’s education level. If your child is not accurately represented in these samples, then their performance in speech and language will NOT be accurately assessed!
There is A LOT that goes into providing a valid and reliable speech and language
assessment for your child. We encourage you to be in the know about how your child is being tested for a disorder or delay and advocate for them/ask questions if you are unsure about the assessment that has been chosen for them.
Helpful sources:
https://www.asha.org/practice-portal/clinical-topics/late-language-emergence/assessment-
tools-techniques-and-data-sources/ (ASHA article)
https://www.leadersproject.org/2013/03/01/types-of-validity-in-testing/ (Validity)
https://www.leadersproject.org/2013/03/01/reliability-of-tests/ (Reliability)
https://www.voxlingue.com/post/speech-and-language-assessment-pediatric-standardized-
tests (Testing process and common tests given to children for speech and language)
Gracie Lee is a student at the University of Alabama studying Communicative Disorders with a minor in Business. She plans to become a Speech-Language Pathologist and specialize in aural rehabilitation. In her spare time, Gracie likes to read, bake, and go to concerts with her friends.
Skyrocket Pediatric Therapy Foundation (Skyrocket) does not provide medical or legal advice or services. Skyrocket provides general information about developmental disabilities and developmental therapies as a service to the community. The information provided on our website is not a recommendation, referral or endorsement of any resource, therapeutic method, or service provider and does not replace the advice of medical, legal or educational professionals. Skyrocket has not validated and is not responsible for any information, events, or services provided by third parties. The views and opinions expressed in blogs on our website do not necessarily reflect the views of Skyrocket.