United States
Hong KongWhat a scaled score actually means
A scaled score is a position on a common reporting scale, not a share of questions answered correctly. Raw scores from different forms of the exam are equated and converted onto that scale so the same number means the same standard on every form. TDI's FY2025 mean was 69.75.
Your result on this exam arrives as a number that looks like a percentage and is not one. That single confusion is responsible for most of what is written wrongly about the Texas insurance exam, so it is worth taking apart slowly.
Raw score, then scaled score
| Raw score | Scaled score | |
|---|---|---|
| What it is | How many of the 130 scored questions you got right | A position on a common reporting scale |
| Depends on your form? | Yes | No, that is the point |
| Reported to you? | No | Yes |
| Comparable across candidates? | Only if they sat the same form | Yes |
| Is it a percentage? | It can be expressed as one | No |
You answer questions, so a raw score exists. It just is not the thing anyone reports, because on its own it does not answer the question the Department is asking.
Why different forms need reconciling
Exams are assembled to a specification. The content outline fixes what proportion of the paper covers each of the 13 sections, so every form has 100 general questions and 30 Texas ones distributed the same way. What the specification cannot fix is difficulty.
Two questions can both test nonforfeiture options, both be fair, both be correctly keyed, and still differ in how many competent candidates get them right. Build a hundred-question form out of items like that and small differences accumulate. One form ends up a shade harder than another.
Now imagine the standard were a flat percentage of questions correct. A candidate who drew the harder form would need more knowledge to reach it than an identical candidate who drew the easier one. The bar would move with the luck of the draw, which is the exact unfairness the whole apparatus exists to prevent.
Equating, in plain terms
Equating is the measurement that tells you how much harder one form was than another. It works by comparing performance on questions the two forms share, since any difference in how a group performs on the rest of the paper, when they perform identically on the shared part, is evidence about the forms rather than the candidates.
Once you know how much harder a form was, you can correct for it. That correction is what turns a raw count into a scaled score.
- The problem
- Forms differ slightly in difficulty
- The measurement
- Equating, using questions common to forms
- The correction
- Scaling raw scores onto one reporting scale
- The result
- The same scaled number means the same standard on any form
- The cost
- You are not told your raw score or your percentage
That last row is a genuine loss to the candidate and it is worth saying so rather than presenting scaling as pure gain. You lose the diagnostic satisfaction of knowing you got a specific number right. You gain a standard that does not depend on which Tuesday you booked.
What this makes true, and what it makes false
It makes one thing true: there is no fixed number of the 130 you must answer correctly. On a harder form, slightly fewer will do. On an easier one, slightly more.
It makes a whole genre of internet advice false. A specific count out of 130. A confident percentage you can aim at and be safe. Both describe a threshold that does not exist in the form they state it, and both are common enough that you have probably already read one.
The numbers TDI does publish
In its FY2025 examination report, the Department publishes a mean scaled score of 69.75 with a standard deviation of 14.51, against a first-time pass rate of 57.7 percent from 26,427 candidates.
The standard deviation is the interesting one and nobody quotes it. 14.51 on a scale whose mean is 69.75 is a wide spread. It says the candidate population is not clustered tightly around a common level of preparation; it is spread out, with a long tail of people who arrived badly under-prepared and a top end who were never in doubt.
That shape matters for how you read the 57.7 percent pass rate. It is not a knife-edge exam where everyone is near the line and luck decides. Most candidates are some distance from the standard in one direction or the other, which is a more encouraging picture than the raw pass rate suggests.
Why the mean sitting near 70 is not proof of anything on its own
A mean of 69.75 next to an internet pass mark of 70 percent looks like confirmation. It is not, and the reasoning is worth having straight.
Scales are chosen. A testing program can put its reporting scale wherever it likes, and plenty of programs set the passing standard at a round number by construction. The mean landing near 70 is consistent with a standard near a scaled 70, which is a statement about a scale. It says nothing about the share of questions correct.
Where this site says the evidence puts the cut near a scaled 70, that is our cross-check against the published mean, standard deviation and pass rate. TDI has published no passing score and we are not converting our inference into a percentage of questions.
How to set a target anyway
Use percentages on your own papers, because that is the only measurement you control, and set the number high enough that form difficulty cannot reach you. Ours is 78 percent on our own mocks. It is a target of ours, not a pass mark of the Department's, and the gap between it and the plausible standard is the safety margin doing its job.
Common questions
What is a scaled score on the Texas insurance exam?
A position on a common reporting scale, produced by converting your raw count of correct answers so that results from different forms of the exam mean the same thing. It is not a percentage of questions answered correctly.
Why is a raw percentage not reported?
Because it would not be comparable between forms. Two forms built to the same content outline still differ slightly in difficulty, so the same percentage would represent different levels of knowledge depending on which form a candidate drew.
What is equating?
The statistical measurement of how much harder one form of an exam was than another, using questions the forms have in common. It is what makes the correction from raw to scaled score defensible rather than arbitrary.
What does TDI's mean scaled score of 69.75 tell you?
That the average first-time candidate lands close to where the standard plausibly sits. Paired with a standard deviation of 14.51, it also shows a wide spread of preparation rather than a field clustered on the line.
Is a scaled 70 the same as 70 percent?
No, and treating them as the same is the root of most misinformation about this exam. A scaled score is a scale position; a percentage is a share of questions. The two numbers being close is a property of the scale, not a conversion.