###### Representing Data

##### Tables

For Psychologists, tables are a very important thing. The **title** should be informative with the units stated and the
**column titles** should allow you to know exactly what the results show.

Below is a table with the results from an experiment where students who had had breakfast and not, did a maths test.

##### Graphs

Graphs are a good way of visually representing your findings in a study, the first one will be a **histogram**. They are different to bar charts because their scale is continuous (1,2,3,4,5 rather than red, blue and green), so the bars are put next to each other. In a 'proper' histogram there is frequency density but for the purposes of Psychology we will only look at **frequency** which means the 'number of'.

Another type of graph that you may encounter is a **scatter graph** this is done by plotting one thing against another, and will show a **correlation**. As well as a visual representation there are a number of methods that we can use to calculate correlation numerically.

A correlation of +1 would be perfect positive correlation and a correlation of -1 is perfect negative. And having a correlation of 0 would say there is no relationship at all. The diagram below outlines all of this.

##### Averages and Distribution

Another term for average is measures of central tendancy. So an average gives an indication as to the most typical result; there are three main types, outlined in the table below.

Mode | The most frequent piece of data. i.e. the one that appears the most. |
---|---|

Median | All of the data is put in order, and the middle one is selected. If there is an even amount of data the two middle are added together and halved. |

Mean | All of the data is added together and then divided by the number there was. |

The **distribution** of some data can also be called its **spread**. The easiest method is called the **range** and you simply subtract the smallest value from the largest. However this is not too accurate.

This is why statisticians have devised a measure called **standard deviation**. Put simply this measure tells you how much the data **deviates** (is smaller or larger than) the mean; on average. It is relative to the size of the mean.