Why Factoring Out Before Integrating Can Wreck Your Integral in Calculus

The rule being borrowed here is the constant multiple rule. If $f$ is integrable on $[a,b]$ and $c$ keeps the same value at every point of the interval, then

Test it with $c = 5$: $\int_0^2 5x\,dx = 5\int_0^2 x\,dx = 10$, which direct integration confirms. The same condition governs the indefinite case, $\int c \cdot f(x)\,dx = c \int f(x)\,dx$, where there are no bounds to catch you.

Everything hangs on one word. "Constant" does not mean "the factor out front." It means a quantity that does not change as $x$ sweeps the interval. As $x$ runs from $0$ to $2$, the number $5$ stays $5$; the quantity $x$ does not, and nor does $x+1$, or $\sin(x)$, or anything else with an $x$ in it. The test is not what the factor looks like; it is whether it holds still.

Here is the move carried to the end:

Stop there. The bounds are numbers, so the integral is a number. The $x$ inside never takes a value of its own: it runs from $0$ to $2$ and is used up. The $x$ now parked out front is a free variable the bounds never reach, so an $x$ surviving to the final line proves the integration never happened.

Now do the same thing to the other factor:

One integral, two shortcuts, two different answers. $4x$ and $2x+2$ agree only at $x=1$; at $x=2$ one says $8$, the other $6$. A legitimate step cannot depend on which factor you grabbed. This is not a rule with an exception; it is not a rule.

Multiply the product out, then integrate term by term:

Check it against the picture: between $0$ and $2$ the height $x(x+1)$ never exceeds $6$, so the area is under $2 \times 6 = 12$, and $\frac{14}{3}$, about $4.67$, fits. The discarded $4x$ cannot face that check.

The move feels right because it is right nearly everywhere else: pulling a common factor to the front is a reflex built over years of algebra. My own read, as judgment and nothing firmer — the rule was learned; its condition was not.

That shape of failure has been studied, though not in calculus. More than a thousand children in grades two through eight sat a written arithmetic test and reached for proportion where it plainly did not apply^[1]. Those errors grew more common as their skill with proportional reasoning grew, peaking around grade five, and a considerable number were still slipping in grade eight^[1]. These are children doing word problems, so it is a suggestive parallel, not proof about calculus^[1].

Closer to home, undergraduates who had already passed a full year of calculus were asked five questions about integration, and as a group did not show a satisfactory grasp of what it means^[2]. The ones who understood the idea were also the ones who carried out the steps well, which is as much as that study can say^[2]. A year of correct answers is no guarantee that a picture of the operation came with it.

Experienced calculus students working on pure mathematics and physics problems drew on several mental images of the integral, and adding up many small pieces looked especially useful in multivariable and physics settings^[3]. That image also makes the error easy to see — my argument, not the study's. If the integral totals many thin strips, each with its own $x$, moving an $x$ out in front of the whole sum leaves nothing to say which strip it came from.

Sometimes the variable out front is the hinge the problem turns on. Take

The $2y$ depends on $y$, so it cannot leave. Let $u = y^2$, so $du = 2y\,dy$; as $y$ goes from $0$ to $1$, so does $u$, and the integral becomes

The factor you wanted to discard is what made the substitution work. Sometimes it is the only reason a problem can be finished: $e^{x^2}$ has no antiderivative you can write with the usual functions, so in $\int_1^3 x e^{x^2}\,dx$ that lone $x$ is everything.

Showing the broken derivation, and marking where it dies, was deliberate.

In two experiments with students learning probability problems, a wrong solution shown beside a correct one helped learners who already had a good grounding when they moved on to new problems, while those with little background did better from correct solutions alone^[4]. A review pooling 40 studies across subjects agreed in part: examples with a mistake in them can help, but how much depends on whether the mistake is explained or highlighted, and how students work with it^[5].

So, the exercise. Before you compute $\int_1^3 x e^{x^2}\,dx$, say out loud, in a full sentence, why the $x$ cannot come out front. Then find the substitution that uses it, carry the bounds across, and check you land on $\frac{1}{2}\left(e^9 - e\right)$. A review of students learning mathematics found a small lift from prompting them to explain material to themselves, measured straight afterwards; its authors recommend asking learners to explain why a common mistake is wrong^[6]. The sentence, not the answer, is what I want you carrying to the next problem.

The mathematics above needs no study behind it; it is proved on the page. The research is here for two narrower questions — why this slip happens, and what helps — and on both it leaves gaps. Nothing cited is about this error: no study tested the constant multiple rule, or what happens when a variable is pulled outside an integral sign^[1][2][3]. The closest work measures how well calculus students grasp integration in general^[2] and which mental images they use^[3]; neither asked whether naming a rule's condition out loud prevents the mistake^[2][3]. And the over-reaching finding comes from children doing arithmetic word problems, so it says nothing about how often this slip happens in first-year calculus^[1].

The work on learning from wrong examples disagrees with itself: the two experiments found it helped only learners who already had a grounding in the topic^[4], while the larger review called the evidence on prior background mixed and asked for more research^[5]. So I cannot promise the broken derivation helps someone meeting integrals for the first time^[4][5]; if that is you, work the correct path until it is steady, then come back for the wreckage. The explain-it-back exercise has its own hedge: that lift showed up in immediate testing, and the evidence that it lasts, or survives an ordinary classroom, is much weaker^[6].

Before moving anything across an integral sign, ask whether it changes as the variable runs between the bounds. If it does, it stays. And if the variable is still in your answer, you have not integrated yet.