Statistical Tests

Statistical Tests

Cut off points

>0.05 statistically significant

<0.0001 Statistically significant

Anywhere in between may require another statistical test to confirm the hypothesis or the null hypothesis

Parametric

Makes an assumption on the parameters of the data, the defining properties of where the data comes from and its population distribution.

Used when data follows a normal distribution, bell-shaped curve

Non-Parametric

Makes no assumptions about the data, taking only into account the numbers themselves.

Used when the data follows a non-normal distribution, such a linear or bolic

Shapiro Wilks Test

This is a simple test to see if your data has normal or non-normal distribution.

If normal - use Parametric statistical tests to analyse potential differences between groups / levels / means

If non-normal - use Non-parametric statistical tests to analyse potential differences between groups / levels / means

http://sdittami.altervista.org/shapirotest/ShapiroTest.html

Comparing means

Parametric

T-Test

The T-test is a measure of variance for two variables, it compares the mean of two normally distributed data sets and finds a statistically significant difference

Used when needed to compare the means of two populations in a normal distribution to find a significant difference

1-Way ANOVA

Compares the variance of two or more variables, usually used when the t test isn`t, compares a normal distribution of data

Used when needed to compare the means of three populations in a normal distribution to find a significant difference

Non-Parametric

Mann Whitney U-Test

Looks at the null hypothesis such that a randomly selected value from one sample will be less than or greater than a randomly selected value from another sample

Used for continuous data that is not normally distributed. It is a rank order test that compares how data is spread out between two variables

Wilcoxon

A Wilcoxon signed-rank test is a nonparametric test that can be used to determine whether two dependent samples were selected from populations having the same distribution

Similar to the U-test, sees how data is spread out

Kruskal-Wallis

is a rank-based nonparametric test that can be used to determine if there are statistically significant differences between two or more groups of an independent variable on a continuous or ordinal dependent variable.

Used when there are more than two samples of data, with different n numbers

Catagorical

ANCOVA

Analysis of covariance is used to test the main and interaction effects of categorical variables on a continuous dependent variable, controlling for the effects of selected other continuous variables, which co-vary with the dependent.

Used when the data`s trend needs to be analysed and not just it`s mean

Comparing Data in a contingency table

Non-Parametric

Fisher s exact test

Fisher`s exact test is a statistical significance test used in the analysis of contingency tables. Although in practice it is employed when sample sizes are small, it is valid for all sample sizes.

Looking at relationships

Parametric

Pearson Linear Regression

It is used to determine the extent to which there is a linear relationship between a dependent variable and one or more independent variables

Non-Parametric

Chi Squared Test

A chi-square test, also written as test