Which half do candidates fail on?

We went looking for this number. It is not published, and after reading TDI's examination report we are fairly confident it does not exist in any public form.

Why the number does not exist

Two published facts close the door. The general and Texas portions are sat in one appointment and scored as one result, so Pearson does not produce a portion-level pass or fail in the first place. And TDI's annual report gives one pass rate, one mean scaled score and one standard deviation per examination, with no sub-scores.

You will still see the claim, usually phrased as "most candidates fail on the state law section". It is a plausible-sounding sentence with nothing behind it. When you meet it, ask which document it came from, because the two documents that could contain it do not.

What can be reasoned from the weights

Weight is not the same as failure, but it is what you have.

A perfect Texas portion is worth less than a quarter of the paper. It cannot rescue a weak general portion, and that is arithmetic rather than opinion. The reverse is much more forgiving: a strong general portion absorbs a lot of damage on Texas statute, because there are only 30 marks there to lose.

So even without the failure data, the risk is asymmetric. If you are going to be weak somewhere, be weak on the smaller thing.

The distribution hints at something, but only faintly

The one indirect signal in the published data is the standard deviation, which is 14.51 against a mean of 69.75. That is a wide spread. Wide spreads happen when a population contains genuinely different groups rather than one group with a bit of variation.

Our reading is that the spread is mostly about preparation rather than about portions: Texas requires no pre-licensing course, an appointment is $49, and some people sit this examination having read very little. That is a hypothesis. The report cannot confirm it, and we are labeling it as ours so nobody quotes it back as TDI's.

The only version of this question you can actually answer

Not "which half do candidates fail", but "which half am I failing". That one is answerable in an evening, and it is the useful question anyway.

• Sit the two portions separately and time them. Score the 100-question general side and the 30-question Texas side apart from each other, even though the real paper will not
• Weight what you find. A gap of ten points on the general side matters more than the same gap on Texas, because the general side is more than three times the size
• Re-measure after each block of study, not once at the start. A single reading tells you where you were, not whether anything is moving

Our own diagnostic splits the result exactly this way, which is the honest reason it exists. Ten questions will not diagnose you precisely. They will tell you which side of the split your problem is on, and that decides how the next month of your studying is organized.

What we would ask TDI for

Mean scaled sub-scores by portion, and the eventual pass rate for candidates who fail once. Neither is in the report. The first would settle this page permanently, and the second would tell a candidate who has just failed something far more useful than a first-attempt rate does.

Until either appears, the responsible answer to which half people fail is that we do not know. That is a worse page than the one where we tell you confidently. It is the only one we can source.