Research is

Concept — research is defined by its method, not by its result. Standard methodology definitions all fix the same pair of features. Clifford Woody describes research as defining and redefining a problem, formulating a hypothesis, collecting and evaluating data, and testing whether the conclusions fit that hypothesis; Kerlinger calls it a systematic, controlled, empirical and critical investigation of hypothetical propositions; Redman and Mory call it a systematised effort to gain new knowledge. Each of these fixes (i) a scientific, systematic procedure — the method — applied to (ii) establishing what is true about a stated problem — the object. A definition that supplies only one of the two is incomplete.

Application — test each offered phrase for both features.

• "Searching again and again" reads the word re-search literally and supplies repetition only. Repeating a look adds no systematic procedure and asserts no truth about a defined problem — a storekeeper who counts the same stock twice has repeated a search and has done no research.

• "Finding solution to any problem" names the object loosely and leaves the method open. A solution reached by asking a colleague, following a manual, or trial and error is still a solution, so the phrase would admit any route at all; it also stops at solving rather than at establishing what is true, which is why exploratory and descriptive studies that solve nothing still count as research.

• "Working in a scientific way to search for truth of any problem" carries both features at once: "in a scientific way" is the systematic, controlled, empirical procedure, and "to search for truth of any problem" is the pursuit of verified knowledge about a stated problem.

• "None of these" is a closed alternative that would hold only if no listed phrase carried both features.

Cross-check — run the definition backwards. Take any accepted piece of research — a controlled trial, a sample survey, a field study. Each one follows a stated procedure and reports what it found to be true about a defined problem, satisfying both features. Now take the near-misses: mere repetition satisfies neither, and a solution obtained anyhow satisfies at most the second. Only the phrase that names the scientific procedure together with the pursuit of truth survives the test, which also rules out the closed "none" alternative.

Result. Research is working in a scientific way to search for the truth of any problem.