It's a typo. That said, you are incorrect on many things about this problem. You failed to provide critical context, making your post almost unintelligible and impossible to answer. It was only when that context was provided via a link in the comments to the text itself that I was able to puzzle this out.
This is one of a multi-part sequence of exercises meant to show that if $\mathbf{F}$ contains a primitive $p$-th root of unity, then $x^p-a$ either has a root in $\mathbf{F}$ or else is irreducible over $\mathbf{F}$.
The preamble to the set of exercises explicitly states that $\omega\in\mathbf{F}$ is a primitive $p$th root of unity. That means that it is false that "$\mathbf{F}$ can be $\mathbb{Q}$ or $\mathbb{R}$ or anything". In fact, it can be neither $\mathbb{R}$ nor $\mathbb{Q}$ (which contain no primitive $p$-th root of unity for any prime other than $p=2$), nor can it be many finite fields.
Second, the first part of the problem tells you to assume that the field does not contain any roots of $x^p-a$, but that $x^p-a$ is reducible; the goal is to establish a contradiction at the end, so that you will conclude that $x^p-a$ either has a root in $\mathbf{F}$ or else is irreducible over $\mathbf{F}$. Note that this means that you cannot take $a=1$, since that violates the assumption that $x^p-a$ has no roots in $\mathbf{F}$.
This is where the typo comes in. You are expected to explain why you can factor $x^p-a$ into a product $p(x)f(x)$ where both factors have degree at least $2$ ("$\geq 2$" instead of "$\leq 2$").
At which point, you are in the territory of the duplicate indicated above.