Testing for DifferencesAnalyses of Variance (ANOVAs) are used to determine statistical differences between treatments. There are two simple 'types' of ANOVA you can use; one-way or two-way. Before you conduct an ANOVA, however, make sure your explanatory variable is discrete and your response variable is continuous. This analysis should only be used if your data is parametric. For non-parametric data you can either transform the data or use the Kruskal-Wallis ANOVA. ![]() One-way ANOVA This is used to determine differences between two or more factors (treatments). For example, you could be researching the effects of temperature on the number of cells of Mycobacterium leprae (leprosy) by exposing the bacteria to different incubation temperatures (20°C, 25°C, 30°C, 35°C and 40°C). The data would be analysed using a one-way ANOVA because your experiment has one independent variable (temperature). The results of the ANOVA are presented as means of the treatments, with distribution graphs, standard deviation, standard error and variance. The P-value obtained from the statistical software will tell you whether or not there is a significant effect of your treatments. Two-way ANOVA The two-way ANOVA is used to determine differences and/or relationships between two factors (treatments). Mycobacterium leprae growth is dependant upon the optimum temperature and acidity (two independent variables). Given this extra information, you could modify the above experiment to look at the combined effect of temperature and pH on the growth of the bacteria. This would be analysed using a two-way ANOVA. The results of the analysis would be presented as means of treatments (temperature and pH), distribution graphs, standard deviation, standard error and variance. P-values will be quoted for differences between treatments and the relationship between temperature and pH.
|
||
| <<< Previous Page >>> |
| © Copyright 2007, Centre for Excellence in Teaching and Learning in Applied Undergraduate Research Skills (CETL-AURS), University of Reading, UK. All rights reserved. If you wish to apply for permission to use any materials found on the ENGAGE website, please contact us at engageinresearch@reading.ac.uk |