A proper fraction g(x)f(x) (degf<degg) decomposes uniquely into simpler pieces once g(x) is factored into linear and irreducible-quadratic factors:
- each linear factor (x−a)k contributes x−aA1+(x−a)2A2+⋯+(x−a)kAk;
- each irreducible quadratic factor (x2+ax+b)k contributes x2+ax+bB1x+C1+⋯+(x2+ax+b)kBkx+Ck.
Finding the constants. Clear denominators, then either substitute each linear factor's root directly (the cover-up rule -- instantly isolates that factor's constant, since every other term vanishes), or -- required whenever a quadratic or repeated factor is present -- expand and match coefficients of like powers of x on both sides; a convenient extra substitution (often x=0) frequently speeds up the coefficient-matching step.
Improper fractions (degf≥degg) are first reduced by polynomial long division to a polynomial part plus a genuinely proper remainder, and only the remainder is decomposed into partial fractions.
This machinery is the standard first step before integrating a rational function in calculus -- each partial-fraction term integrates far more simply than the combined original.