EXAM stuff Flashcards
What is chi square
- Test for association or randomness or homogeneity
- Non-parametric test to compare observed results to expected results on the basis of a null hypothesis
- Samples must be random
- Expected frequency must exceed 5
- Calculates a c²-statistic that is compared to a statistical table to determine whether there is a statistically significant difference between the observed and expected frequencies
Null hypothesis for chi squatred
There was no significant difference in observed and expected values for L. obtusata of each colour
OR
There was no significant difference in frequency of occurrence between the three colours
t tests
- Parametric test to compare the means of two sets of continuous data
- Can be used with large and small sample sizes
- Calculates a t-statistic and P-value that allow us to determine whether there is a statistically significant difference in the two means
data screening t test
- Data must be continuous
- i.e. measurements or on an uninterrupted scale
- Data should be normally distributed
- Use plot or normality test (e.g. Kolmogorov Smirnov)
- Variances should be similar between samples
- Use F-test for equality of variance
p vlalue t test
- If your P-value is less than 0.05 then there is a statistically significant difference in the means (at P = 0.05) and we can reject a null hypothesis
- If your P-value is greater than 0.05 then there is no statistically significant difference in the means (at P = 0.05) and we can accept a null hypothesis
what is a mann-whitney u test?
- Non-parametric test to compare the medians of two sets of discontinuous data
- Can be used with as few as 4 observations in each sample
- Samples can be unequal
- Distribution free so can be used with non-normal data
p value mann whitney u
- If your P-value is less than 0.05 then there is a statistically significant difference in the medians (at P = 0.05) and we can reject a null hypothesis
- If your P-value is greater than 0.05 then there is no statistically significant difference in the medians (at P = 0.05) and we can accept a null hypothesis
what is an anova?
- One-way analysis of variance
- Parametric test to compare the means of three or more sets of data
- Samples can be unequal
- Calculates a F-statistic and a P-value that allow us to determine whether there is a statistically significant difference between the three (or more) means
p value anova
- We determine significance using the P-value
- If your P-value is less than 0.05 then there is a statistically significant difference in the means (at P = 0.05) and we can reject a null hypothesis
- If your P-value is greater than 0.05 then there is no statistically significant difference in the means (at P = 0.05) and we can accept a null hypothesis
what is a krustal wallace?
- Non-parametric test to compare the medians of three or more sets of data
- Can be used with as few as 5 observations in each sample
- Samples can be unequal
- Distribution free so can be used with non-normal data
krustal wallace p value
- We determine significance using the P-value
- If your P-value is less than 0.05 then there is a statistically significant difference in the medians (at P = 0.05) and we can reject a null hypothesis
- If your P-value is greater than 0.05 then there is no statistically significant difference in the medians (at P = 0.05) and we can accept a null hypothesis
What do corralations establish?
- Need to establish
- If there is a relationship
- Nature of the relationship
- If relationship is significant
What is a pearsons corralation coefficient?
•Also called Product Moment Correlation Coefficient
•
•Parametric test to determine if there is a linear relationship between two variables
•
•Calculates a r-statistic and a P-value that allow us to determine whether there is a statistically significant relationship between two variables
p vlaue pearsons corralation coefficient
- If your P-value is less than 0.05 then there is a statistically significant relationship between the two variables (at P = 0.05) and we can reject a null hypothesis
- If your P-value is greater than 0.05 then there is no statistically significant relationship between the two variables (at P = 0.05) and we can accept a null hypothesis
what is a spearmens rank corralation coefficient?
•Non-parametric test to determine if there is a relationship between two variables
•
•Use if both variables or one variable is discontinuous
•
•Converts x and y variables into ranks
•
•Calculates a rs-statistic and a P-value that allow us to determine whether there is a statistically significant relationship between two variables