This is the research companion to When to Push Anyway. Read the main post first.
This series has spent a lot of words telling you to stop pushing, so we owe you the caveat. Sometimes you push. The trick is that two very different things both get called pushing. Pushing a child through a hard patch in something they chose is one thing. Pushing a child toward something you chose for them is another. The first builds the muscle. The second builds resentment. Five sources anchor why the line sits where it does, and every one of them carries the same honest warning: hold the line warmly, with a reachable next step, or you get the opposite of what you wanted.
The Dip Is a Phase, Not a Verdict
Crone, E. A., & van Drunen, L. (2024). Development of self-concept in childhood and adolescence: How neuroscience can inform theory and vice versa. Human Development, 68, 255-271.
DOI: 10.1159/000539844
What they found: This review synthesizes longitudinal brain-imaging and self-report studies and describes a nonlinear pattern in how young people rate themselves. Positive self-evaluation is high in childhood, dips in mid-adolescence as social comparison replaces the child's earlier, rosier self-view, and then recovers in late adolescence. The dip is most pronounced in the academic domain. The authors read it not as something breaking but as the self-evaluation system getting more sophisticated: a previously simple, overly positive self-model is temporarily destabilized while a more realistic, more differentiated one comes online.
Why this matters for you: When your kid suddenly decides they are bad at the thing they used to love, that low is often the phase, not the truth. A parent who treats every hard day as proof the child should quit is reading a temporary dip as a permanent verdict. Knowing the pattern lets you respond with patience instead of alarm, and hold a steady bar while the wobble passes.
What it doesn't answer: This is a narrative review, mostly of the authors' own Dutch and European samples, and the brain measures it leans on have low test-retest reliability. It describes an average trajectory, not your specific child, and it does not tell you which particular hard patch is worth pushing through.
How Deep the Dip Goes, and How Steady the Rest Stays
Robins, R. W., & Trzesniewski, K. H. (2005). Self-esteem development across the lifespan. Current Directions in Psychological Science, 14(3), 158-162.
DOI: 10.1111/j.0963-7214.2005.00353.x
What they found: Robins and Trzesniewski pooled large cross-sectional and longitudinal datasets to chart how self-esteem moves across a whole life. The shape is a long, gentle arc. Self-esteem is high in childhood, declines through adolescence to a low point in the mid-teens, then climbs again across adulthood. The adolescent dip is real and normative, it shows up across many samples, and it tends to run deeper for girls. Just as important, self-esteem is fairly stable in rank — the children who start relatively higher tend to stay relatively higher even as the whole group dips, so the dip is a change in level, not a scrambling of who a child is.
Why this matters for you: This is the population picture behind your kid's rough stretch. A mid-adolescent low in self-regard is the expected reading on the chart, not a sign something went wrong in your home. And because rank order holds, a temporary dip does not erase the child underneath it. Knowing the arc bends back up is what lets you hold a steady bar through the low point, rather than rescuing them out of every hard thing while their confidence is briefly underwater.
What it doesn't answer: How large the adolescent decline really is remains debated. The biggest longitudinal meta-analysis to date (Orth, Erol, and Luciano, 2018, Psychological Bulletin, DOI 10.1037/bul0000161) finds self-esteem steadier through the early teens than the folk story suggests, with the clearer gains arriving later. So treat the dip as a real but modest and domain-specific wobble in how a kid rates their own ability, not a global collapse of self-worth. And like the rest here, it is an average curve, not a rule for which particular hard patch is worth pushing through.
Sample Wide Early, Commit Deep Later
Epstein, D. (2019). Range: Why Generalists Triumph in a Specialized World. Riverhead Books.
ISBN: 978-0735214507
What they found: Epstein gathers research across sport, music, science, and knowledge work and finds a recurring pattern: people who end up excellent often pass through a sampling period first, trying many things before they specialize. Match quality, the fit between a person and their pursuit, usually improves by testing and adjusting rather than by picking early and grinding. The book also reframes struggle itself as a desirable difficulty, a sign of learning rather than a sign the child lacks the ability.
Why this matters for you: This is the front end of the push question. Early on, breadth is the goal, so let them try and drop things freely. The point is to collect signal about what actually fits. Later, once something has been genuinely chosen, that is when it is worth helping them stay through the hard patches. Wide early, deep later. And when the deep phase gets hard, remember the struggle can be the work, not a red flag.
What it doesn't answer: Range draws heavily on elite athletes, musicians, and adult knowledge workers, not typical childhood, and much of its evidence is observational and biographical rather than a controlled parenting trial. It makes a strong case for sampling, but it does not pin down the day a hard patch stops being useful signal and starts being the thing worth pushing through.
Belief in Ability Moves the Needle, but Only Where the Kid Can Act
Yeager, D. S., Hanselman, P., Walton, G. M., Murray, J. S., Crosnoe, R., Muller, C., ... Dweck, C. S. (2019). A national experiment reveals where a growth mindset improves achievement. Nature, 573, 364-369.
DOI: 10.1038/s41586-019-1466-y
What they found: A national randomized experiment with about 12,000 US ninth graders across 65 schools tested a short online exercise teaching that ability can grow. It raised grades for lower-achieving students (about a tenth of a grade point) and made students more likely to take a harder math course the next year. The effect was not uniform: it showed up where the school's peer norms and course options actually let a student act on the new belief, and it was flat where they did not. The authors call it a seed that needs the right soil.
Why this matters for you: This is the evidence for holding the line warmly with a reachable next step, rather than forcing. Telling a child you believe they can grow does something real, but only when there is an actual harder step within reach and a setting that lets them take it. Two other large trials point the same way. A two-country replication across the US and Norway found the effect held up (Rege, M., et al., 2021, American Psychologist, DOI 10.1037/amp0000647), and a four-year field experiment in disadvantaged French schools found a small, durable, cheap gain of about 0.05 standard deviations that plateaued rather than growing with more of the same (Huillery, E., et al., 2025, The Economic Journal, DOI 10.1093/ej/ueaf015). Belief plus a reachable challenge moves the needle. Belief alone, or pressure with no path, does not.
What it doesn't answer: The effects are modest and concentrated among students who were struggling and in supportive settings, and they plateau. This was measured on school grades, not at a kitchen table, so the parent version of holding the line is reasoned from the school evidence rather than directly tested there.
Whose Goal Is This
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68-78.
DOI: 10.1037/0003-066X.55.1.68
What they found: Self-determination theory holds that durable motivation depends on three needs being met: autonomy, a sense of ownership over what you do; competence, a sense you can grow at it; and relatedness, connection to people who matter. Persistence at a pursuit a person genuinely owns tends to be durable and self-sustaining. When the same behavior is driven mainly by outside control, motivation shifts from internal to external and tends to erode once the pressure lifts.
Why this matters for you: This is the spine of the whole caveat, and it is one question: whose goal is this. Pushing a child through a hard stretch of something they chose supports the autonomy their motivation runs on, because the goal is already theirs. Pushing a child toward something you chose for them does the opposite: it makes the pursuit yours, and the wanting tends to leave with the pressure. Same word, pushing, opposite result, and the difference is who owns the goal.
What it doesn't answer: This is a synthesis of the theory and its supporting research, not a single experiment, and it names the need without settling the daily calibration. Real limits still apply, such as safety, family obligations, and a young child who would choose dessert for dinner. The theory tells you autonomy matters. It does not tell you where your particular line goes tonight.
Coming Up
That was almost the close of the pushing arc. One post remains, because every test in this series assumes a parent can see what a demand costs, and for neurodivergent kids the cost is the invisible part. Next time, what pushing looks like when the same ask costs one kid a dollar and another kid four, and what fair means in a house with differently wired kids.
Every Functional Child Development post comes with a Dive Deeper companion. Subscribe to get both, every week.
