4. Design your experiment

By this time you have chosen a research question to study and have reviewed the research literature to determine what is already known about your question. Now it’s time to start thinking through the details of how you will collect and anlyze your data.

At the end of Task 1, you started to think ahead about how you will formulate the details of your experiment in such a way that the research will be feasible or doable with the limited time and resources that are available. In this Task, you’ll spell out some of the details that will help you plan data collection and data analysis, as well as evaluate how feasible the experiment is.

Homework:
- Hand in your experimental design
- Be ready to explain the pros and cons of your methodological decisions

What you need to do:

• 4.1 Read about the different kinds of experimental and non-experimental research to understand why you're doing an experiment
• 4.2 Choose the levels for your factors
• 4.3 Identify the experimental conditions in your design
• 4.4 Decide how to test your factors, whether between subjects or within subjects
• 4.5 Check your design for internal validity

Remember to ask as many questions (using this web site!) as you need to understand everything clearly.

Why do you need to do this?
Developing an experimental design is part of the broader process of planning your research. Before you proceed with your experiment, you need to plan in great detail exactly how your data collection will happen: you need to design your experiment to decide on the details well before collecting data. This planning will help you to do several things:

• Think more about the details of your research problem;
• Produce the right data for your analyses;
• Produce more relevant and more reliable data;
• Avoid stress and confusion during data collection and analysis;
• Use your time more effectively;
• Finish on schedule;

Planning is the key to successful research (and to other complex activities). It is particularly important to plan your analyses before you start to collect data.

Your experimental design is an important part of a plan for how you will collect and analyze your data. This plan becomes the Method section of the research paper. For many scientists, the clarity and precision of the Method section helps them decide whether or not they will believe your results.

Keep in mind that you are specifying your experimental design so that you know exactly what you will do at every step of your data collection and analysis. As with other plans, however, when they’re put into practice, reality has a way of forcing you to change them. Planning ahead helps prepare you to react to these changes.

Statistical analyses. You should build your experimental design with a particular statistical analysis technique in mind. The experimental design is also a plan for how you will do your statistical analyses after the data has been collected. This way, you plan ahead so that you collect the right data to fit the statistical analyses that you will do.

This course focuses on the statistical technique called Analysis of Variance (ANOVA). It’s been chosen ahead of time to simplify parts of the teaching process. It’s also one of the most flexible and widely used statistical techniques in experimental psychology. ANOVA, however, is only appropriate when you want to study the effects of discrete or categorical grouping independent variables (ex: male vs. female; old vs. young) on continuous dependent variables (ex: number correct). You need to use other techniques, such as regression analysis, when you study the effects of continuous independent variables (ex: age, IQ, etc.) or when you have categorical dependent variables.

ANOVA is used to study the effects of discrete or categorical independent variables on continuous dependent variables.

Given that you’re going to use ANOVA, you have to know several things to use the statistical analysis software:

• which factors are in the design;
• how many levels there are of each factor;
• how you want to test each factor;

You need to identify which experimental conditions you will use to help you understand this information.

It is very important to note that specifying your experimental design will sometimes make you reformulate or modify your initial research problem if it turns out to be too difficult to study. This means that you will also have to make changes to your Literature Review so that it matches the changes in your research problem.

Read this topic next: Different kinds of Research