Before you write a single line of algebra, notice the shape of the question. When a problem asks "for what values of x does [some condition] hold?", the answer is almost never a single number. It is almost always a region of the real line — an interval, or a handful of intervals glued together with \cup symbols.
This is the recognition skill: the moment your eyes land on "for what values of x," you should already be reaching for interval notation. Your pen should be ready to write (, [, \cup, \infty. You are not solving for a value. You are describing a set.
The pattern
Almost every question in introductory algebra and calculus that begins "for what values of x..." resolves to one of these four shapes.
| Shape | Example | Answer |
|---|---|---|
| Single interval (bounded) | x^2 - 4x + 3 < 0 | (1, 3) |
| Single interval (unbounded / ray) | 2x - 7 \ge 0 | [3.5, \infty) |
| Union of two or more intervals | x^2 - 4 > 0 | (-\infty, -2) \cup (2, \infty) |
| All of \mathbb{R} or \varnothing | x^2 + 1 > 0 | (-\infty, \infty) |
Notice what is not on that list: "x = 5." A single number is the answer to an equation (2x - 7 = 3), not to a "for what values" question. If you find yourself writing just x = 5 as an answer to a "for what values" question, pause — you probably misread the problem, or the problem is one of the rare exceptions we discuss at the end.
Why this shape shows up everywhere
"For what values of x..." is the natural phrasing for three kinds of problem, and each one produces a region.
1. Inequalities. The solution to an inequality is a set of points where one side exceeds the other. On a graph, that set is the portion of the x-axis where one curve sits above another. Curves are continuous, so the crossing points partition the axis into finitely many pieces, and on each piece the inequality either holds or fails. The solution is the union of the pieces where it holds — an interval, a ray, or a disjoint union.
2. Domain questions. "For what values of x is f(x) defined?" asks where the formula makes sense. Forbidden operations (dividing by zero, taking a square root of a negative, feeding a non-positive number into \log) exclude a handful of points or intervals from \mathbb{R}. What remains is a union of intervals.
3. Convergence / validity questions. "For what values of x does this series converge?" or "For what values of x is this expansion valid?" almost always produces an interval — the interval of convergence — often symmetric around a centre, with behaviour at the endpoints decided case by case.
In all three cases, the answer is a subset of \mathbb{R} that is a union of intervals. That fact flows directly from the continuity of the underlying functions: continuous functions change sign only at isolated points, so the regions where they are positive, negative, or defined are unions of intervals.
A worked example: domain
Problem. For what values of x is f(x) = \dfrac{\sqrt{x - 2}}{x - 5} defined?
Before any algebra, forecast the answer. This is a "for what values" question about a domain. The expected shape is a union of intervals.
Now do the work. The square root demands x - 2 \ge 0, so x \ge 2. The denominator demands x - 5 \ne 0, so x \ne 5. Intersecting: x \in [2, 5) \cup (5, \infty).
A union of two intervals, exactly as forecast. If you had ended up with x = 7 or x = 2, you would know something went wrong — the shape would not fit the question.
A worked example: inequality
Problem. For what values of x does x^2 \le 9?
Forecast: inequality, quadratic. The answer will be a single interval (the "inside" of a parabola condition) or a union of two rays (the "outside"). Since x^2 \le 9 is the "inside" version — values of x where x^2 is small — expect a single bounded interval.
Rewrite as |x| \le 3, which says x is within 3 of 0. That is [-3, 3].
A single bounded interval, exactly as forecast.
A worked example: convergence
Problem. For what values of x does the geometric series \sum_{n=0}^{\infty} x^n converge?
Forecast: a convergence question. Expect an interval, likely centred at 0.
The series converges iff |x| < 1, which is the interval (-1, 1). At the endpoints x = \pm 1, the series becomes 1 + 1 + 1 + \cdots or 1 - 1 + 1 - \cdots — neither converges — so the interval stays open.
A single bounded interval, exactly as forecast.
The real exceptions
There are honest exceptions. Spot them early so they do not derail your reading of the problem.
Finite set. "For what values of x is x^2 = 9?" The answer is \{-3, 3\} — two isolated points, not an interval. But notice: the question uses =, not an inequality or a domain condition. The word "equation" in the problem is usually a signal.
Empty set. "For what values of x is x^2 + 1 < 0?" No real x works, so the answer is \varnothing. This counts as a degenerate interval. The empty set is often the result of a "no solution" problem designed to test whether you check feasibility.
All of \mathbb{R}. "For what values of x is (x - 1)^2 \ge 0?" Every real x works, so the answer is (-\infty, \infty) = \mathbb{R}. Also a degenerate "interval" — one that covers everything.
Discrete set from a divisibility or integer condition. "For what integer values of n does n^2 < 10?" gives \{-3, -2, -1, 0, 1, 2, 3\}. The restriction to integers (not real x) breaks the interval pattern. Watch the word "integer" or "natural number" — it changes the shape of the answer.
Even with these exceptions, the default assumption for a "for what values of x" problem over the reals is: the answer is an interval or a union of intervals. Carrying that expectation into the problem shapes how you set up the work — you reach for a sign chart, a number-line sketch, interval notation — instead of trying to solve for a single number.
How to use this recognition
When you read "for what values of x":
- Have interval notation ready. Leave space for \cup and \infty. Don't write "x =" at the top of your working.
- Sketch a number line. See Sketch the Number Line First, Solve Second — this reflex pairs directly with the recognition here.
- Expect at most a handful of pieces. A quadratic produces one or two. A cubic produces up to two or three. Rational functions with vertical asymptotes produce one extra piece per asymptote. If your algebra produces fifteen disjoint intervals, something has gone wrong.
- Check the shape against the problem type. Strict inequality on a quadratic with two real roots? Expect either a bounded open interval (parabola below the axis between roots) or a union of two open rays (parabola above the axis outside the roots). If your answer doesn't have that shape, re-check the sign.
The recognition saves time because you stop fighting the problem. You are not hunting for a specific x; you are carving a region out of \mathbb{R}, and the tools — inequalities, sign charts, interval notation — are all designed for exactly that task. If you want the toolkit itself, Intervals and Inequalities Preview walks through every bracket type and every inequality move you need. For the absolute-value family specifically, |x − c| < r Rewrites as an Interval Instantly shows the one-line pattern that turns disguised distance problems into intervals on sight.
Once you expect an interval, you stop being surprised by the answer. You start the problem already looking for it.