## Choosing a Hypothesis

 Getting Started General Instructions | Introduction to Your Study | Experimental Design | Stating a Hypothesis Descriptive Statistics Histograms | Central Tendency | Standard Deviation | Confidence Intervals Comparing Two Samples Samples and Populations | Choosing a T-Test | Paired T-Test | P-Values and T-Tables Important Concepts The Normal Distribution | Z Scores | Probability Distributions Levels You are currently on Stating a Hypothesis at level 2. Level 1 | Level 2 | Level 3 Next Topic Experimental Design | Frequency Histograms

### Explanation

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.
• 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.
Using the example of height from the list above, you could phrase the research hypothesis as this follows:

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.

### Exploration

You can construct a clumsy version of the research hypothesis using a sentence like the one below:

Units in the first sample of independent variable have a difference average dependent variable than units in the second sample sample of independent variable.

Once you have the sentence like this, you can tidy it up to make it read better.

### Application

Your tutor provided the following research hypothesis for the experiment that produced your data:

The number of items correctly recalled will decrease when the recall interval is increased

Try to reproduce a similar research hypothesis using the method described in the section above. Click the [Generate Hypothesis] to see how your hypothesis reads.

in the    sample of    have a    average    than    under the    sample of