ExplanationChoosing Your Hypothesis
In this section, you will learn how to choose and phrase a hypothesis for a study or experiment. A hypothesis is generally a single sentence that describes what your experiment sets out to test. Here are the key points to remember.
Using the example of height from the list above, you could phrase the research hypothesis as this follows:
- The hypothesis should identify the following:
- What the experimental units are (are they people, soil samples, bacteria cells?);
- What is being compared, i.e. the dependent variable (is it height, acidity, life-span?);
- What separates the things being compared, i.e. what is the independent variable and what two values does it take (is it male/female gender, or clay/sandy soil, or anti-bacterial gell/placebo treatment);
- Whether a direction is expected (do we expect height to be greater in males, or do we just want to know if there is a difference in either direction?). For reasons that will be explained later, experiments where a direction is expected are called one tailed and experiments where a change in either direction is expected are called two tailed.
- Include these things in a sentence, not a list;
- Be as precise and specific as you can.
People of the male gender have a larger average height than people of the female gender.
Point at any of the highlighted words in the sentence above to see which of the points above the word covers. You could re-write the sentence to make it less clumsy, for example, 'Males are taller, on average, than females', but you are assuming that who ever reads your hypothesis can work out that gender is the independent variable and that you are measuring people, not some other animal.
The null hypothesis is usually just the opposite of the research hypothesis, for example 'People of the male gender DO NOT have a larger average height than people of the female gender'. Notice how we keep a reference to the direction in the null hypothesis too.