Understanding your results
How your exam results are interpreted
Your Memory Exam score is a raw score, which is the number of points you earned on the assessment. On its own, that number tells only part of the story.
To put your results into context, your score is compared with results from people with similar characteristics, such as age, education, and gender. This comparison is expressed as a percentile, which shows how your performance compares with that reference population.
Your results are one part of a larger clinical picture. A neurologist or neuropsychologist considers your medical history, your pattern of results across the assessment, and other relevant information when determining whether further evaluation is appropriate. They may also recommend additional testing.
If your score falls below the 16th percentile on any individual test, we recommend following up with a neurologist or neuropsychologist.
Converting a raw score to a percentile
Converting a raw score to a percentile uses basic statistical methods. Converting it using normative data involves comparing your test score to the scores of a representative sample of people who have taken the same test, in order to determine where your score falls in relation to the sample.
To do this, the raw score (the total number of points earned on the test) is first compared to the average score for the normative sample, which is typically represented as a mean and standard deviation. The standard deviation indicates how much variation there is in the scores of the normative sample.
Once the raw score has been compared to the mean and standard deviation of the normative sample, a percentile rank can be calculated. The percentile rank represents the percentage of people in the normative sample who scored at or below your raw score.
For example, if your raw score on a test is 75, and the mean score for the normative sample is 65 with a standard deviation of 10, then your score is one standard deviation above the mean (i.e., a z-score of 1).
Looking up the corresponding percentile rank for a z-score of 1 in a standard normal distribution table, we find that the percentile rank is about 84%. This means that the individual scored higher than about 84% of the people in the normative sample.
Converting raw scores to percentile ranks using normative data is a common way of interpreting test scores because it provides a meaningful way of comparing an individual's performance to that of a larger population.
The real question is: compared with whom?
A percentile is only meaningful when you know who you're being compared with. That's where correct interpretation makes the difference.
Most brief memory screens use a fixed cut-off, applying the same threshold to nearly everyone, sometimes with a small adjustment for education. This can lead to errors in both directions: an older adult with less formal education may be flagged despite having no impairment, while someone whose memory has declined significantly from a high baseline may still score above the threshold and receive false reassurance.
Your results on Memory Exam are read against 26 age bands and 12 education bands, built from published reference data and from our own study of nearly 1,000 adults ages 18 to 100. The same score can mean very different things for a 48-year-old college graduate and an 82-year-old who left school at 15 — and it should be interpreted accordingly.
What a percentile doesn't tell you
A percentile can be useful, but it's easy to misinterpret. Three things are worth understanding.
- It's not a percentage score. Scoring at the 40th percentile doesn't mean you answered 40% of the questions correctly. It means you scored higher than approximately 40% of people in your comparison group.
- It's not a diagnosis. A low percentile is a reason to look more closely, not a conclusion about your cognitive health. A poor night's sleep, a heavy cold, test anxiety, or even a missing hearing aid can affect your performance on a given day.
- It's not a permanent label. A percentile reflects one assessment at one point in time. Your next result may offer a different perspective—and comparing results over time can tell you more than any single score.
What a single score can't tell you
Even when compared with the right reference group, a single result answers only one question: how does your performance compare with that of people like you? It cannot tell you what may have changed over time.
Two people can receive the same percentile and be in very different positions. One may have scored consistently at that level for years. The other may have experienced a significant decline. A single score cannot distinguish between them.
That's why having an earlier result matters — ideally, a baseline established before you have concerns about your memory.
Read more about why two results are worth more than one.