Teaching · Related Services

Reliable enough for this decision, or not.

Enter the coefficients from the manual and what the score will decide. The review compares each with the usual minimum and shows how wide the error band really is.

Review a testWhat you get
$5one-time · every device you sign in on
Buy · $5

Included with Caveat Classroom and Whole-school licenses. Or get it in the Related Services toolkit, $99 for 96 items.

The reliability review

One review per test score and date. The example is made up.

Loading the form. If nothing appears, turn on JavaScript.

What you get

Teaching · Related Services$5 · included with Classroom and Whole-school licenses

What you finish with
Each coefficient against the minimum for the decision, the standard error of measurement, a 95% band and the cautions to report.
Who it is for
School psychologists, educational diagnosticians, speech-language pathologists and evaluation teams choosing or interpreting tests.
Format
Browser form. One saved review per test score and date; printed with the summary or as a blank form; CSV; export and import a file.
Access
$5 one-time, yours on every device you sign in on; the tool opens once you are signed in with it. Included with Caveat Classroom and Whole-school licenses. Also in the Related Services toolkit ($99) and the Teaching Suite pass ($199), and 1 more bundle. Everything you type stays in this browser; nothing is sent to Caveat.
Where it lives
The Related Services space in Caveat Teaching.
Support
Questions about an order, access or a correction: team@caveatacademic.com.
every score is a range
Drawn for this page. Reliability tells you how wide that range is before you make a decision with it.

How to use it

  1. Name the test and the score you will use.
  2. Choose what the score will decide.
  3. Enter every reliability coefficient the manual gives for that score.
  4. Mark the four checks on the manual.
  5. Report scores with their band and the cautions.

Review a test

How the review works

The minimums

A common rule of thumb asks for reliability of about .90 when a score helps decide something about one student, .80 for screening and .70 for group or research use. They are guides from the measurement literature, not legal standards.

The standard error

SEM = SD × √(1 − r). With a standard deviation of 15 and reliability of .84, the SEM is 6 points, so a 95% band runs about 12 points either side of the score.

What this is not

A validity review. A reliable score can still measure the wrong thing.

Sources

  1. American Educational Research Association, American Psychological Association, & National Council on Measurement in Education (2014). Standards for Educational and Psychological Testing, chapter 2.
  2. Salvia, J., Ysseldyke, J. E., & Witmer, S. (2017). Assessment in Special and Inclusive Education (13th ed.). Cengage: reliability standards by type of decision.
  3. Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric Theory (3rd ed.). McGraw-Hill.
  4. Individuals with Disabilities Education Act regulations, 34 CFR 300.304(c)(1)(iii): assessments used for purposes for which they are valid and reliable.

The example review is made up. Support: team@caveatacademic.com.