Someone says the thing. That was exactly right. We could not have done it without you.
Something sorts it before the sentence has finished arriving. They are being kind. They did not see the first three drafts. That is what managers say in reviews. The sorting is fast, it does not feel like a decision, and that is most of the reason it never gets looked at.
What confuses everybody standing on the other side is the arithmetic. Praise looks like the obvious remedy for a person who cannot see what they have done, so more of it gets offered, with more emphasis, and the effect of each round is smaller than the last.
266 studies
asked which way people file their own results
d = 0.96
average pull toward crediting the self in those studies
38%
of feedback interventions in a 131-experiment review made performance worse
63%
of one first-generation grade gap closed by a single session about belonging
The question that cannot be answered, and the one that can
Whether a person is good enough is not settleable from inside that person, and it is certainly not settleable by a questionnaire. The experience has a name in the literature, impostor phenomenon, and the term's history, the 1978 study it came from and the prevalence figures that get quoted alongside it are all laid out in the entry on impostor phenomenon. That is not the subject here.
What is left once the terminology is set aside is a rule. Results arrive and get assigned somewhere. Assignment is a procedure. Procedures can be written down, and a written procedure can be checked against a particular Tuesday.
That is a much smaller question than whether somebody deserves their job. Unlike the larger one it has an answer.
The books are kept two different ways
Taken one row at a time, every line of that is defensible. Timing is real. Reviews do catch things. Rooms vary. Nothing in the left column is a lie, which is exactly why it survives inspection for years.
The asymmetry only shows up when the columns are set beside each other, and what shows up is not modesty. It is a filing system running one rule for credits and a different rule for debits, which guarantees the balance never moves regardless of what gets posted to it.
Attribution research has a vocabulary for this. Bernard Weiner's 1985 theory in Psychological Review sorts causes along three axes: inside the person or outside, stable or temporary, controllable or not. The axes matter because they decide what an outcome is permitted to teach. A result attributed to something external, unstable and uncontrollable teaches nothing about the person, by construction rather than by accident. Luck is the purest case of all three, and it is the row this pattern reaches for first.
The direction most people file in is the other one. Amy Mezulis and colleagues pooled 266 studies and 503 effect sizes in Psychological Bulletin in 2004 and found an average pull of d = 0.96 toward taking personal credit for successes and locating failures outside. In US samples it ran at 1.05.
- US samples: 1.05d
- Other Western samples: 0.7d
- Asian samples: 0.3d
The same review found the pull shrinking where self-criticism runs high: 0.46 in samples with anxiety, 0.21 in samples with depression. Nothing in that is a diagnosis of anybody reading it. The point is narrower and more useful. The direction of the filing is a measurable variable, it moves, and it can be established without anyone first having to settle whether the results were any good.
Fewer signals, and an absent signal is not a malfunction
Two things need saying before any of this hardens into a verdict about a person.
The first is that none of it is a diagnosis and none of it is a defect waiting for a cure. No manual contains it, there are no criteria, there is nothing to treat. It is a habit of accounting, and habits of accounting are learned somewhere.
Where they get learned is the part usually left out. This pattern turns up more in people from minority groups, in first-generation university students, in people who changed field in their thirties, and in anyone who looks unlike the rest of the room. The familiar reading is that such people arrive with thinner confidence. The likelier reading is that such rooms issue fewer signals that a person belongs in them, and that registering the absence of a signal is accurate perception rather than distortion.
There is hard evidence on the signal rather than on the confidence, and it is some of the most replicated work in the area. Greg Walton and Geoffrey Cohen reported a randomised trial in Science in 2011 in which first-year students spent one session with material framing social adversity at university as common and temporary. Three years later the Black students who had received it were carrying a cumulative grade average of 3.60 against 3.34 in the comparison group, which halved the achievement gap at that institution, and they were reporting better health and fewer visits to a doctor. Nicole Stephens and colleagues ran a different version in Psychological Science in 2014 with 168 incoming students, where first-generation students were graduating the year 0.30 grade points behind their peers. One hour of hearing senior students describe how their own backgrounds had shaped what they ran into closed 63 per cent of that gap.
Neither study went anywhere near anybody's ability. What moved was the reading of the room, and the reading of the room turned out to be carrying the weight.
A signal that was never sent cannot be retrieved by trying harder to believe it arrived.
What happens to a compliment on the way in
William Swann and colleagues ran the experiment that explains the texture of this, published in the Journal of Personality and Social Psychology in 1987 under a title worth keeping: the cognitive-affective crossfire. People holding positive and negative self-views were given favourable or unfavourable feedback about themselves, and their reactions were measured along two separate channels, what they felt and what they concluded. The channels came apart. Mood tracked how flattering the feedback was. Judgements of its accuracy tracked how well it matched the self-view already in place.
So the warm twenty seconds and the verdict that it was not really true are not a contradiction, and nobody involved is being insincere. Two systems reach different answers about the same sentence. Only one of them is allowed to write in the ledger.
| How praise gets reclassified | The sentence it becomes | What it ends up being evidence about |
|---|---|---|
| Incomplete information | They did not see the first three drafts | Their view, which is based on less than the full picture |
| Social obligation | That is what gets said at the end of a project | The format of the occasion |
| Wrong unit | They are judging the output, not what it took | Something other than the person |
Each exit leaves the praise intact as a social event and empty as information. Nothing has been argued with, nothing has been rejected out loud, and nothing has been credited either.
Which is why volume is the one variable that cannot help. A rule that reclassifies the first compliment reclassifies the fifth, and by the fifth the reclassification has acquired a second job, because now there is a person who has gone to some trouble and who has to be handled. The predictable endpoint is praise becoming a thing to get through.
More preparation is what this runs on
The over-preparation works. That is the difficulty with it.
Every time something goes well on the back of three times the necessary effort, the result is available to be filed under the scaffolding rather than under whoever built the scaffolding. The assumption that the person alone was not sufficient comes out of the week confirmed, with one more entry supporting it, and the dose required next time is now set by the last success, which is on the record as a measurement of how much support was needed.
Run that loop for six years and the usual outcome is not exposure. It is a large quantity of unbilled work, maintained indefinitely, insuring against a scene that keeps not happening.
Write the actions, not the verdict
There is one practical move in this piece and it is deliberately narrow. After something lands, write down what was done. Actions only. No assessment of them.
The entries should read flat. Sent the revised scope to the client on Tuesday before anyone asked for it. Rewrote the third section after reading the transcript twice. Phoned the supplier instead of emailing. Noticed the figure in the appendix did not match the summary and went back to check it.
What is excluded is the entire register of assessment. Not it went well. Not I handled that badly. Not I think they were happy with it. An evaluation is a sentence about a person, and sentences about a person are precisely what the rule is built to process. An action is a sentence about a task, and the rule has no procedure for reassigning a phone call made on a Tuesday to good luck.
Support for that distinction arrives from an unexpected direction. Avraham Kluger and Angelo DeNisi reviewed 131 experiments, 607 effect sizes and 23,663 observations in Psychological Bulletin in 1996 and found that feedback raised performance on average and lowered it in 38 per cent of cases. Their explanation of the failures is the part that matters here. Feedback degrades performance when it pulls attention away from the task and onto the self. A weekly written self-assessment is a feedback intervention aimed at the self, administered by the self, on a schedule. An action record points the other way.
There is a second reason for the format, and it has nothing to do with feeling better. Memory keeps verdicts and discards procedures. Eight months on, what survives of a good quarter is a vague impression and no specifics, which leaves the next conversation about contribution to be conducted entirely out of the impression. That is the wrong material for the purpose, and it is the only material most people have.
When it is not this at all
The second honest thing is a distinction the whole subject tends to blur, and getting it wrong wastes time in both directions.
Being in a role that is ahead of current ability is not imposture. It is a learning period, it is how anybody arrives at being able to do anything, and it responds to entirely different moves: working beside somebody who has done it before, shortening the feedback loop so answers come back in days instead of quarters, and naming the specific missing thing out loud to a person who can supply it. All three require being visible, which is exactly what a fear of exposure spends its budget avoiding.
| The difference | A learning period | The accounting habit |
|---|---|---|
| What it names | Specific gaps, which can be listed | Nothing specific. A general verdict about a person |
| When the task gets done | The doubt goes down | The result is reassigned and the doubt holds |
| What it responds to | Instruction, repetition, a shorter loop | Not those, and not more preparation either |
| What it asks for | Help with a named thing | Cover |
No questionnaire can separate those two, including the one on this site, because both produce the same answers to questions about confidence. The person holding them can separate them, by checking whether the doubt produces a list.
Four things that arrive in the same week
Self-criticism rarely shows up as a single mechanism, and four of them get collapsed into whichever word is nearest to hand.
| Pattern | What it is actually about |
|---|---|
| High-functioning anxiety | The gap between the anxiety and the functioning. Nothing visible has gone wrong, and nothing does |
| Perfectionism | The standard itself, and the belief that below it there is no worth |
| Impostor feeling | The attribution. The work went out, and it does not count as the person's own |
| People-pleasing | The ordering. What somebody else needs is known before what the person needs is remembered |
All four in one person is the ordinary case rather than a complication. What they do not share is the handling, which is the only reason the distinction earns its space. The imposter syndrome check here scores the third one across three facets: where results get filed, how much attention goes on not being found out, and what becomes of evidence pointing the other way. Fifteen questions, reading the accounting rather than the ability, and diagnosing nothing.
What is left
Nothing above argues that the ledger has reached the wrong total. It argues that the ledger is being kept under two rules, that the rules are written in a hand a person can learn to read, and that all of this can be established without settling a question that was never going to be settled from the inside.
The next compliment will still slide off. It is not built to stick, and no amount of goodwill from the person offering it changes the routing.
What can be kept is a record of moves, in a register the rule has no instructions for. Nobody has to agree with it for it to be a record.
Sources
- Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4).
- Mezulis, A. H., Abramson, L. Y., Hyde, J. S., Hankin, B. L. (2004). Is there a universal positivity bias in attributions? A meta-analytic review of individual, developmental, and cultural differences in the self-serving attributional bias. Psychological Bulletin, 130(5).
- Swann, W. B., Griffin, J. J., Predmore, S. C., Gaines, B. (1987). The cognitive-affective crossfire: when self-consistency confronts self-enhancement. Journal of Personality and Social Psychology, 52(5).
- Kluger, A. N., DeNisi, A. (1996). The effects of feedback interventions on performance: a historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119(2).
- Walton, G. M., Cohen, G. L. (2011). A brief social-belonging intervention improves academic and health outcomes of minority students. Science, 331(6023).
- Stephens, N. M., Hamedani, M. G., Destin, M. (2014). Closing the social-class achievement gap: a difference-education intervention improves first-generation students' academic performance and all students' college transition. Psychological Science, 25(4).