Failure of the Change of Sign Method

Even Number of Roots: If an interval contains an even number of roots (e.g., 2, 4, 6), the function will cross the x-axis and then cross back again. This results in $f(a)$ and $f(b)$ having the same sign, falsely suggesting that no roots exist in the interval.

Undetected Odd Roots: While a sign change guarantees at least one root, it does not reveal the presence of multiple roots if there is an odd number greater than one (e.g., 3 roots). The method only identifies that 'some' roots exist, but fails to distinguish between a single root and multiple crossings.

Solution through Subdivision: To avoid missing roots, the interval should be systematically subdivided into smaller regions until the sign change behavior clearly isolates individual solutions.

Asymptotic False Positives: In functions with vertical asymptotes, such as $f(x) = \frac{1}{x}$, the graph may jump from a positive value to a negative value without crossing the x-axis. This produces a sign change that mimics a root but actually represents a point of non-existence.

Verification Strategy: When a sign change is found, it is crucial to verify that the function is defined and continuous throughout the interval. Evaluating the function at points very close to the suspected root can help determine if it is approaching zero or infinity.

Geometric Intuition: On a graph, a true root involves a continuous line intersecting the x-axis, whereas a failure due to discontinuity involves the graph 'disappearing' off the top of the grid and 'reappearing' at the bottom.

No Sign Change at Turning Points: If a root occurs at a local minimum or maximum that lies exactly on the x-axis (a repeated root), the function 'touches' the axis but does not pass through it. In this case, $f(x)$ remains positive (or negative) on both sides of the root.

Method Failure: Because there is no sign change, the standard numerical method will completely fail to detect this root, even if the interval is extremely small.

Identification Technique: These points can be identified by looking for local extrema where $f(x) = 0$ or by observing that the derivative $f'(x)$ is also zero at the suspected root

Check for Continuity FIRST: Always look at the function's denominator or domains (e.g., logs or square roots) to identify potential asymptotes before applying sign change logic.

The 'Small Interval' Rule: Examiners often test this by giving an interval like $[1, 2]$ where two roots exist. Always check the midpoint (e.g., $x=1.5$) to see if the function has crossed and returned within the bounds.

Rounding and Accuracy: When using bounds to prove a root to a specific number of decimal places, evaluate at the upper and lower bounds (e.g., $x.x5$ and $x.(x+1)5$). A sign change here rigorously proves the root's location to that accuracy, provided no discontinuity exists.

Common Mistake: Assuming that no sign change means no root. In exams, if a function looks like a perfect square, such as $(x-2)^2$, remember it will touch the axis at $x=2$ without a sign change.