Academic and formal language (B2 Vocabulary)

These formal academic nouns keep the plural from Greek or Latin, so the ending shifts instead of just adding -s. A criterion is a standard you judge or decide something by, and its plural is criteria. A phenomenon is an observable fact or event that can be studied, and its plural is phenomena. An analysis is a careful, detailed examination of something, and its plural is analyses (say "-seez"). Because these plurals don't look like plurals, learners often add an extra -s, which sounds off to an examiner.

Greek and Latin plurals

Abstract nouns for ideas

Formal writing leans on abstract nouns to name ideas precisely. Significance means the importance or deeper meaning of something. An implication is a consequence or result that follows from something. A perspective is a viewpoint or angle you look at a question from. They often sit in fixed pairings, such as serious implications or from multiple perspectives. All three are neutral-to-formal in register and fit academic writing well.

Methodology and framework

Two nouns describe how a study was set up, and they are easy to swap by mistake. A methodology is the system of methods a study uses to gather and analyze its data, so it answers what the researchers actually did. A framework is the set of ideas that organizes the thinking, so it answers how they make sense of what they found. A paper describes its methodology in the methods section and states its framework when it explains its reasoning.

Assumption

An assumption is something researchers take as true without testing it, so it sits underneath the work rather than coming out of it. The fixed phrase is make an assumption, and a study is often said to rest on an assumption. Note the spelling: the noun from assume is assumption, with -mption.

What the study found

Two more nouns report results. A correlation is a measured relationship between two things that change together, so it can be strong or weak and positive or negative. Validity is whether a study really measures the thing it claims to measure. Note the warning that always travels with the first one: a correlation on its own never proves that one thing caused the other.