The aim of this essay is to look into the correlation and difference test. It will also explain how to analyse both the testing methods and how each test is conducted. SPSS was used to conduct both the correlation and difference tests and the tables and graphs that are used are also from this programme. Wagner (2007) says that using SPSS for social statistics and research methods indicates the many elements of the programme and how to use it successfully. Correlation Test
Correlation is a key part of statistical data and is used to describe the degree of relationship between two variables. This includes the strength and direction of the relationship (Bryman and Cramer, 2009). When carrying out this test a scatter graph is plotted so that it can then be analysed. A scatter graph allows a large range of values to be used and both variables can be shown on the same graph at the same time (Argyrous, 2005). The Y-axis is the dependant variable, this is the affected variable. Then the X-axis is the independent variable which affects the other variable (Argyrous, 2005). Once the scatter graph has been plotted the direction, correlation and the strength of the correlation can be analysed. Analysis becomes a lot more intensive after the data has been collected (Thomas, J.R, 2005). The correlation can be positive or negative and is represented as a numerical figure. This decimal figure ranges from -1 to +1 (Hosker, 2008) When the figure is 0 this means that there is no relationship between the two variables. A positive correlation is where one variable increases at the same time as the other increases. A negative correlation is where one variable increases and the other decreases. (Downing, D and Clark, J. 2010) The strength of the scatter graph shows how close the results are. It is important to include a line of best fit which then shows clearly the strength of the scatter graph. If all the points are close to the line of best fit this equals a strong positive correlation between the two variables. If however the points are all spread out then there is a weak correlation. Sometimes however there is a strong correlation but also a few anomalies. These are results that do not quite fit the pattern of results. Variables
The two variables that are being used for data are resting heart rate and exercise hours. The resting heart rate would be the dependant variable with this changing based on the amount of exercise taken part in by that specific person. This is on the Y – axis, and then the independent variable is the exercise hours on the x – axis.
When looking at the results on the initial look at the scatter graph, it looks like there is a weak negative correlation. It shows that the more exercise taken part in the lover the resting heart rate. I have now come up with the hypothesis below.
When conducting the correlation test using Pearson’s product moment both a null and alternative hypothesis was written. These are shown below Null hypothesis - There will be no significant relationship between exercise hours and resting heart rate. Alternative hypothesis - There will be a significant relationship between resting heart rate and exercise hours. Pearson’s Product Coefficient
Pearson’s product moment is an example of correlation coefficient, it is a measure of the linear association between two variables that have been measure on a ratio scale. Bryman and Cramer (2009) states that, when one of the variables is a ratio or interval variable, the most reliable correlation test to use is the Pearson’s product r. The values can be anywhere between -1 which is perfect negative correlation and 1 which is perfect positive correlation (McDaniel and Gates, 1998). The value of r is indicated by the strength and direction of the correlation. When this has been worked out it can easily be seen from the table below what the correlation is between the two variables. 0.0 – 0.2
= Weak or no relationship
0.2 – 0.4
Please join StudyMode to read the full document