How to Score the San Diego Quick Assessment Properly
The San Diego Quick Assessment is a timed oral reading inventory. You hand a student a list of words and start a stopwatch. They read each word aloud as fast as they can while staying accurate. You stop the clock when they finish a page, note errors, and move on. That's the whole exercise. The scoring itself is straightforward arithmetic, but the way people administer it is where things usually go wrong. Here's how you actually calculate scores. Time is recorded in seconds for each page. Words per minute comes from dividing the total words on a page by the time in minutes. If a page has 50 words and the student takes 42 seconds, that's 50 divided by 0.7, which equals roughly 71 words per minute. Accuracy is calculated separately. Every word read incorrectly, skipped without substitution, or substituted with a word that doesn't fit counts as an error. Self-corrections do not count as errors if the student arrives at the correct word within three seconds of the initial mistake. Anything longer than three seconds and you log it as two errors — one for the miss and one for the delayed fix. I've seen a lot of people mess this up. The most common error I caught during my first year doing these assessments was timing. Someone would start the clock after the student began reading instead of before the first word. That inflates the time, deflates the WPM, and produces a score that looks worse than it actually is. My workaround was simple: I hold the stopwatch in my hand before calling the student forward, press start the instant I say "begin," and stop it the instant the student's mouth stops moving on the last word. No exceptions. Consistency matters more than speed when you're comparing scores over time.
Level placement works differently than you might expect. The SDQA uses a series of word lists organized by difficulty level. The rule is you start at a level you think is appropriate based on prior data, then go down one level if there are more than five errors on that page. If there are fewer than two errors, you go up one level. You keep adjusting until you find the instructional level, which is defined as the level where the student reads with 95 to 97 percent accuracy. That's your target for teaching. Below that is the frustration level. Above that is the easy level, which tells you what the student can read independently. Here's something most training materials don't emphasize enough: the word lists are not linear. A jump from level H to level I isn't the same increment as level F to level G. The later levels pack in significantly more multi-syllabic words and less frequent vocabulary in a shorter span. When you're placing a student who scores well at the higher levels, you need to be careful about assuming linear growth. A student who reads level J at 120 WPM is not on the same trajectory as a student reading level E at 80 WPM in terms of what that gap means for their instructional needs. The metric still works, but the interpretation shifts. Another thing people overlook is the cutoff rule. If a student scores below 90 percent accuracy on any page, you do not continue. You stop and record the highest level they passed at. There's no point in going further because anything above that is already at frustration level. I used to push through to the end out of habit. One of my supervisors made me stop and I won't pretend I liked it at the time, but it saved me roughly ten minutes per assessment and kept the data honest. You're measuring a range, not hunting for a perfect number.
The fluency score — that's the WPM combined with the accuracy percentage — is what you report. A student reading at 95 percent accuracy and 90 WPM at grade level is performing within expected bounds. Below 90 WPM with under 95 percent accuracy is a red flag that warrants further investigation. The assessment doesn't tell you why. It tells you that something is off. That's its job, and it does that part fine. One limitation worth being blunt about: the SDQA only measures word-level decoding. It says nothing about comprehension, prosody, or reading stamina across extended text. A student can score well on this and still struggle enormously with paragraphs or chapters. I've had cases where a kid hit the upper grade levels on the word lists but couldn't summarize a two-paragraph passage afterward. The assessment is still useful, but it's narrow by design. If you need a fuller picture, pair it with a passage comprehension measure like the DIBELS oral reading fluency probe or the Aimsweb Plus administration. The other downside is that it's norm-referenced but the norms are older. The original standardization dates back to the late 1970s and early 1980s. More recent editions have attempted updates, but the baseline hasn't shifted dramatically. In my experience, this rarely causes practical problems in a classroom setting, but if you're using these scores for eligibility decisions or placement in a funded intervention program, check whether your district accepts the current edition or if they require a more recently standardized tool. Some districts have moved away from the SDQA entirely for that reason.
Get the Full Details
Download links for the assessment materials themselves are typically available through the publisher, McGraw-Hill Education, or through your school district's purchasing office. You can't legally distribute the actual test booklets online without a license, so don't bother looking for free PDFs. What you can find freely are scoring sheets, administration guides, and practice word lists from university education departments and state education agency websites. Those are legitimate resources and they help you prepare before you bring the official materials into a room with a student. Bottom line: score it consistently, stop at the right cutoff, interpret the numbers within the tool's actual scope, and supplement it when you need information about comprehension. The math is simple. The judgment calls around administration and follow-up are what separate a useful assessment from a checkbox exercise.
