# Section 8 Research methods and statistics

**Chapter 36 Why do we need research methods?**

The question on most psychology students' lips as they discover that they have to do 2 years of research methods is 'Why do we need research methods?' On the face of it, it might seem like we just like torturing psychology students by making them learn lots of complicated statistics and the like. Actually, that's not the reason: most of us are nice really. The reason is that psychology is a scientific discipline that tries to understand human behaviour by using research to test and develop ideas about what drives our thoughts and behaviours. This chapter attempts to describe the research process in general terms to give you a background as to why we need research methods, and some of the important issues that we need to consider when conducting psychological research. We start by looking at how theories and research are intrinsically linked, before looking at how we can test theories (and why it's important to measure things). We then look at how we can try to identify cause and effect relationships and how we analyse data.

**Chapter 37 Collecting data**

In the last chapter we saw the importance of collecting data to inform theories and to try to explain human behaviour. This chapter looks at how we conduct research. We begin by looking at a very important aspect of research: ethics. We overview some important issues to consider in conducting ethical research such as informing people about what you're doing, deceiving people, debriefing participants, confidentiality and the right to withdraw from the experiment. We then turn our attention to experimental designs. We learnt a bit about cause and effect in the previous chapter and we build on this to find out why control conditions and randomisation are important in experimental research. We then look at different ways to measure and manipulate variables and overview some commonly used experimental designs. We then turn our attention to non-experimental designs before considering how we actually find people to test.

**Chapter 38 Summarising data**

Imagine you have just collected your first set of data. What on earth are you going to do with it? This chapter explains the basics of summarising and presenting your data. To begin with we look at histograms, which are a way of graphing raw data to see the distribution of scores. We'll discover some of the properties of histograms and find out about a special kind of distribution known as the** **normal distribution. It's important as scientists to summarise your data so that others can quickly and easily see what you have discovered. The chapter looks at some of the ways we can summarise data both in terms of what a typical score is (the mean, mode and median) and how diverse your scores are (the** **range, interquartile range and variance). Next, we look at some common ways of graphing these values, in particular bar charts, line charts, boxplots and scatterplots. Finally, we end with a few comments about how not to draw graphs!

**Chapter 39 Going beyond your sample**

This chapter brings together everything you've learnt so far to try to explain how we use statistics to test hypotheses. The first step is to see whether summaries of our data (like the mean) are representative of our population and we can do this using the standard error and confidence intervals. We then look at the rationale behind fitting statistical models to test hypotheses. You will discover that the experimental and null hypothesis can be conceptualised in terms of a statistical model that is fitted to our data. The exact model depends on what data you have and what your hypotheses are, but in general terms every model throws out a statistic with known properties. Based on these properties, we can work out whether a value as large as the one we have would be likely if the null hypothesis is true. The final part of the chapter is spent giving you a brief overview of a range of statistical models and giving examples of when they should be used.