Almost every method people use has the same defect: the report is written by the thing being reported on. Here is what survives that, and what to do instead.
Ask a person how much they have grown this year and they will answer you honestly. The answer will still be close to worthless as a measurement, because the person answering is the same person being measured. That is not a character flaw. It is a structural problem, and it is the reason so many people do the work for years and genuinely cannot tell whether anything happened.
Almost nothing written about this names the problem. The advice is to journal and reflect. But a journal entry about how much you have grown is authored by the very thing under examination, on a day when you had already decided how the story goes. So start with the defect, because everything useful comes from designing around it.
You want a verdict. Nobody sits down to assess their own change from a neutral posture. You arrive already hoping, and hope edits evidence quietly. On a good week you find proof of transformation. On a bad week the same twelve months read as wasted. Neither reading is about the twelve months.
You cannot reach the earlier version. You are not comparing yourself to who you were. You are comparing yourself to your memory of who you were, and that memory is maintained by the person running the comparison. A large study published in Science in 2013 found the same pattern at every age: people report having changed enormously in the past ten years and expect to change very little in the next ten. Researchers named it the end of history illusion. The present always looks like the finished version, which means your sense of how far you have come is calibrated against a reference point that keeps quietly moving.
The ruler moves with you. This is the one that ruins most tracking habits, and it is worth its own section.
Say last January you rated your patience a 6 out of 10. This January you rate it a 6 again. Flat year, nothing gained.
Maybe. Or maybe you grew a great deal, and the behavior you now call a 6 would have looked like a 9 to the person who made last year's rating. The score held still because the scale moved underneath it. Program evaluators run into this so reliably that it has a name, response shift bias, and it is the reason a before number and an after number from the same person are often not readings from the same instrument at all.
It cuts both ways, and the second direction is crueler. People who make real progress frequently rate themselves worse afterward, because growth raises your standard faster than it raises your behavior. You did not get more anxious. You got better at noticing anxiety, and now you are grading a stricter exam.
A number that means something different this year than it did last year is not a measurement. It is a mood with a decimal point.
There is a clean fix, borrowed from evaluation research: rate both points today, from today's ruler. Do not dig up last year's score. Sit down now and answer two questions in one sitting. How patient am I today? And how patient was I last January, judged by what I understand about patience now? One ruler, two marks. It still runs through memory, so it is not perfect, but it removes the moving scale, which is the larger of the two errors.
Everything below is designed around the same principle: get yourself out of the instrument. The less your current mood can touch the number, the more the number is worth. None of this requires buying anything.
The strongest evidence you own is anything you wrote and cannot now edit. Sent emails. Old messages. A document with a date on it. A journal entry from an ordinary Tuesday, written before you had a thesis about yourself.
Pull one from a year ago and read it cold. Do not read it for content. Read it for voice: what that person was certain about, what they kept circling, what they needed from the reader, what they could not say directly. Then write something today about the same subject and set them side by side. The gap is not a feeling. It is text. You can run this by hand with two printed pages and a highlighter, and it costs nothing.
One warning. Do not use the entries where you were assessing yourself. Those are performances for a future reader, which is you. The ordinary ones measure better.
Paste something from a while ago and something recent. It names what actually moved and what held underneath. Free, no signup.
Nothing you paste is stored, and no account is created.
That is two dated points, taken by hand. A reading starts a record that keeps taking them, so the next comparison is not something you have to remember to run.
Turn the vague thing into a countable one, and set the threshold before you start so you cannot move it later. Not "I want to be more assertive" but "the number of times I said an actual number out loud in a money conversation." Not "I want to be healthier" but "days between deciding to book the appointment and the appointment existing."
Counts have a property that ratings do not: they are indifferent to how you feel about them. Four is four in a triumphant month and a bleak one. Write the definition down on day one and do not renegotiate it, because the urge to renegotiate is itself the most interesting data you will collect.
If you are only going to track one thing, track this. Growth almost never arrives as a smaller reaction. The spike is largely wiring, and it stays roughly the size it has always been. What changes is the tail.
The argument that used to cost three days now costs an afternoon. The rejection that used to end the week ends the hour. The mistake you used to carry for a month you put down by Thursday. Recovery time is countable, it is unusually sensitive to real change, it responds within weeks rather than years, and it is very hard to flatter. Note the date of the hit and the date you were actually functional again. That is the whole practice.
Outside witnesses see the parts of you that never make it into your own account. But "have I changed?" gets you a kind yes and no information, so ask something they cannot answer politely.
Ask the same two people every six months. Keeping the observer constant matters more than picking the perfect observer, because you are trying to measure movement in you, not variance in your panel. Write down what they say verbatim, including the part you want to argue with. The urge to argue is the reading.
The one thing nobody can forge is elapsed time. Everything else about a self assessment can be talked into a better shape after the fact; a dated artifact cannot, because it was made before you knew how the year would go.
So make the measurement now and read it later. Write a page today about what you are afraid of, what you are avoiding, and what you would call a good year. Seal it. Put a note in your calendar twelve months out. Do not reread it in the meantime, because rereading is how the past gets quietly revised. If you want more structure for what to put on the page, these prompts go past the surface layer.
If you change the questions, you have changed the measurement, and any difference you see afterward could be you or could be the new questions. Whatever you use, ask it the same way every time. The whole value of a repeated measure comes from the part that stays still.
This is also the honest case against retaking a personality test to see if you have grown. A test freezes you on the day you took it. Taking it again gives you a second frozen day and no line between them, and because it scores your description of yourself, most of what moves between the two sittings is mood and self image rather than you. That ceiling is real and it is worth understanding before you trust any result.
Personal growth, self growth, self improvement: the words differ and the measurement problem is identical every time. The moment the assessment can be authored by today's mood, you are measuring the mood.
Everything above works and all of it depends on you remembering to run it. That is the practical failure mode. People do the year-one measurement, mean it, and never do the year-two one. The instrument that beats a good habit is the one that keeps the dated points for you.
That is what LUX is. Not a personality test and not a type. It is a living record: you answer six questions, about eight minutes, and it opens a record dated that day. From then on the record accumulates what you add to it, folds that pile down into a shorter and truer reading when enough has gathered, and grows again from there. Grow, compress, grow. Old readings are never overwritten, so the earlier version of you stays on the record as evidence rather than as memory.
The measurements live on the back of your card, and it is worth being exact about what each one is, because vague numbers are the thing this page is against.
Alongside the card there is drift. The record compares your most recent reading against the one before it, and when something has genuinely moved, it tells you once, and at most once every three weeks. When nothing has moved you hear nothing at all. Silence is a real result, and producing silence is something most self improvement products structurally cannot do.
The whole argument is time. A test freezes you on the day you took it. A record moves when you do, and every week it exists it becomes harder to copy, because it has a beginning and no one can go back and start earlier. The first reading and your word are free. The room where the record lives is on the pricing page, yearly or monthly, and you can cancel any time.