The objectives are the shortest part of a piece of work and the one that constrains it most. The analysis, the results, the discussion and the conclusions all hang from them: if they are poorly written, everything after carries the problem. And they are usually poorly written for a simple reason: they are drafted early, before it is quite clear what will be done, and never looked at again.
There is a misunderstanding at the start. Many people mix in one sentence what they want to find out, what they will do and what they expect to find, and out come objectives that are both vague and long. Separating the question, the objective and the hypothesis solves much of the problem, because each has a different job.
This guide explains how they differ, how to write each, which verbs work and which do not, and how to check that your objectives fit the analysis and the conclusions. It works for an article and for a thesis.
Although they share the same page, they do not do the same thing, and it helps to have them clear before writing any.
In an article the objective and hypotheses are usually enough, or the objective alone if the study is exploratory. The question appears mostly in qualitative work, systematic reviews and theses. What should never be missing is the objective, and what should never be surplus is a hypothesis you cannot test with your data.
The general objective describes the overall purpose of the study, in one sentence. The specific ones break it into steps that can be checked separately, and together they should cover the general one without adding anything not in it. In a journal article the norm is one main objective and, at most, two or three secondary ones; in a thesis there is room for more, but the rule is the same: each specific objective should correspond to a specific analysis.
An example with the full structure. General: “To evaluate the efficacy of a brief cognitive restructuring programme for reducing anxiety about oral exams in humanities students.” Specific: “(1) To compare anxiety after the programme between the intervention and waitlist groups; (2) To examine whether the effect holds at three months; (3) To explore whether the effect differs by academic year.”
Notice that each specific objective announces an analysis (a between-group comparison, a follow-up, a moderation effect) and that the third is marked as exploratory. That marking matters: a reviewer tells what you decided to test from what you looked at out of curiosity, and values your saying so.
The verb decides whether the objective can be checked. The ones that work describe an action whose outcome can be shown:
A good hypothesis names the variables, states the expected direction and can be false. “Anxiety influences performance” is too vague: almost any result would confirm it. “The higher the exam anxiety, the lower the oral-test grade, controlling for prior grade” says what, in which sense and with what control.
Two cautions. First: write them before seeing the data, or declare that they were formulated afterwards. Presenting as a prior hypothesis a result that emerged while exploring is what is called HARKing, and reviewers detect it through the excessive coherence between introduction and results. Second: justify in the introduction why you expect that direction. A hypothesis with no support in the literature looks like a bet.
If you have no expected direction, do not invent one. An exploratory objective, declared as such, is perfectly legitimate and reads better than a forced hypothesis.
The quickest way to know whether your objectives are right is to build a table with one row per objective and five columns: the objective, the hypothesis, the analysis that tests it, the result where it is reported and the conclusion drawn. If any cell is empty, there is a problem.
An empty analysis cell means an objective the study cannot answer. An empty result cell, an analysis never reported. A conclusion with no objective, something you claim without having set out. Reviewers build this table in their heads, and comments such as “the results do not answer the objective” or “the conclusions do not follow from the data” almost always come from a cell that was empty.
Make the table once the analysis is finished, and adjust the objectives (not the results) until it fits: if one analysis answers no objective, or an objective that is truly answered is missing, the legitimate fix is to rewrite the wording and declare what was confirmatory and what exploratory.
There are six, and a reviewer recognises them at once. There is more on how objectives link to the rest in the guides to the introduction and to the conclusions, which should answer, one by one, what you set out to do.
Paste the end of your introduction, the paragraph where you state the general aim, the specific objectives and the hypotheses, and you will see the first objection a reviewer would raise, with your exact sentence and what it would take to fix it. It is the reading this guide proposes: whether each objective follows from what you have just argued, whether the hypotheses are testable and directional when they should be, and whether what you promise fits the design you describe next.
It returns one objection, the one that weighs most. It needs at least 350 characters, so include the background paragraph that justifies the aims. What you paste is not stored, and if deciding something would require seeing the method, it says so separately.
Those that correspond to a specific analysis. In an article, normally one main and two or three secondary at most; in a thesis, a few more. If an objective has no analysis to answer it, it is surplus.
The question is what you want to know, in interrogative form; the objective is what you will do to answer it, in the infinitive and with a measurable verb. They say the same from different angles, and many works use only one of the two.
No. In exploratory, descriptive or qualitative studies the objective is enough. What is not advisable is to invent a hypothesis to meet the structure: an exploratory objective declared as such is more honest and reads better.
You can adjust the wording to reflect what you actually did, but you must declare what was decided before and what after seeing the data. Presenting an exploratory result as if it had been the objective from the start is HARKing.
One that describes an action whose outcome can be shown: compare, test, estimate, evaluate, determine, identify, examine. Avoid “to know”, “to understand” and “to study”, which do not say what result you expect.
At the end of the introduction, closing the justification, and the same content is summarised in the abstract. In a thesis they usually have their own section right after the introduction or the theoretical framework.