Writing an alternative hypothesis

The commonest failure is the aim in disguise — music affects memory. It names the topic, not the testable variables, so nothing can be measured. A hypothesis is usable only when both variables are operationalised: name both conditions of the IV (not music but with music versus in silence) and the exact DV (not memory but words recalled out of twenty).

Asked for a directional hypothesis, candidates write there will be a difference — which gives no direction and is non-directional. Asked for a non-directional one, they slip in a comparative such as more and make it directional. Test: is there a comparative — more, less, fewer, higher, lower? If yes it is directional; if no, non-directional.

A directional hypothesis is appropriate only when previous research or theory points one way. Choosing it where the evidence is mixed, or where the question gives no prior research, is a guess dressed up as a prediction. With no basis to commit, a non-directional hypothesis is safer because it does not claim to know the direction. Read the stem for prior research first.

Because the null and the alternative are taught together, candidates asked for the alternative sometimes write there will be no difference — which is the null. The alternative always predicts there will be a difference or relationship; the null predicts there will not. Slipping a no into an alternative turns it into its opposite and loses the mark.

An aim states a study's purpose — to investigate whether music affects memory — while a hypothesis predicts a specific outcome. Candidates often hand back the aim, reworded, as their hypothesis; it reads as a plan, not a prediction, so nothing can be tested. Turn the purpose into a concrete prediction, naming both IV conditions and the exact DV.

A hypothesis that simply re-states the stem — this study will show whether music affects memory — adds nothing testable and reads as the aim again. Repeating the question is not creditworthy. Convert the topic into a concrete prediction: name both IV conditions, name the DV measurement, and predict a difference, committing to a direction only if it can be justified.

Students often think a proper hypothesis must say which way the results go, and that predicting only a difference is incomplete. It is not: a non-directional hypothesis is a fully valid alternative, predicting that a difference exists, without a direction. The word alternative means the alternative to the null, not a back-up.

Two errors weaken answers. First, treating the hypothesis as the result: it is written before data are collected — the prediction, not the finding. Second, saying a study proves a hypothesis: GCSE psychology avoids proof, so a study gives evidence for one prediction over the other. Write that results support it, never prove it.