Unit 3 Flashcards
What are examples of parametric tests?
-F ratio-test (Bartlett’s test)
-1 sample t-test
-Paired t-test
-2 sample t-test
-One-Way ANOVA
-Factorial DOE with one factor and one blocking variable
What are examples of non-parametric tests?
-Leven’s Test
-1 sample sign
-1 sample Wilcoxon
-Mann-Whitney test
-Kruskal-Wallies
-Mood’s median test
-Friedman test
How do you calculate the standard error of the mean (SEOM)?
Standard deviation / square root of sample size
What is the recommended minimum sample size for t-tests to run properly?
When sample sizes are greater than 30
When do you use a z-test?
When sample sizes are greater than 100
What does a One sample t-test do?
Assesses whether the sample (one sample) mean differs significantly from a particular value (Usually a theoretical value)
When is an independent sample t-test used?
When there are two means and two groups that are independent of one another
When is pooled independent t-tests used?
When equal variance is assumed
When is non-pooled independent sample t-tests used?
When unequal variance is assumed
When is paired t-tests used?
-Dependent data (related samples)
-Before and after experiments on the same dataset
What is the p-value for significance?
<0.05
What is the p-value for highly-significant?
<0.01
What is the p-value for very high level of significance?
<0.001
What is the probability of 5% significance of being wrong?
Less than 1 in 20 chance of being wrong
What is the probability of 1% significance of being wrong?
Less than 1 in 100 chance of being wrong
What is the probability of 0.1% significance of being wrong?
Less than 1 in 1000 chance of being wrong
What does a one-tailed directional hypothesis predict?
Predicts the nature of the effect of the independent variable on the dependent variable.
What is an example of a one-tailed directional hypothesis?
In whether the levels of pollution are significantly larger than a maximum threshold marked by legislation.
What does two tailed non-directional hypothesis predict?
-Predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect is not specified
What is an example of a two tailed non-directional hypothesis?
Whether differences in temperature are significantly different either above or below a certain value
What does the left tail in a one tail hypothesis predict?
-Looks for a negative effect
-Whether there is a reduction in the number species / individuals etc
What does the right tail of a one-tailed hypothesis predict?
-Looks for a positive effect
-Whether there is an increase in number of species/ individual etc.
How is the significance level distributed in a two-tailed test?
-Half in left and right tail
-Left 0.025
-Right 0.025
-Makes two tailed tests less powerful than one-tailed t-tests
How is Cohen’s effect size calculated?
CES = Mean difference (mean of group 1 - mean of group 2) / standard deviation
What is cohen’s effect size (Cohen’s d)?
Is an effect size used to indicate the standardised difference between two means
What does a larger cohen’s effect size indicate?
A stronger relationship between two variables
What does a cohen’s effect of 0.2 indicate?
Small
What does a cohen’s effect of 0.5 indicate?
Medium
What does a cohen’s effect of 0.8 indicate?
Large
What is a non-parametric equivalent to a one sample t-test?
Wilcoxon Signed Rank Test
What would be an example of null hypothesis for a Wilcoxon Signed rank test?
The population median (n) is equal to the hypothesised median
How do you calculate the confidence interval?
Standard Error of the mean * t-critical
How do you calculate the standard error of the mean? (SEOM)
Standard deviation / square root of sample size (n)
How do you find the t-critical value?
Using degrees of freedom and a t-table.
You get your degrees of freedom by n-1, or you minus 1 from the number of samples you have
What are two tests that compare variances?
-Fisher’s test
-Levene’s test
Why does the variance need to be compared?
-Before conducting parametric testing to ensure that the right assumptions are met
Is Fisher’s test normal or non normal?
Normal
Works best on parametric data
-Can only assess variances of two groups
Is Levene’s test normal or non-normal?
Non-normal
Used for non-parametric datasets
-Can assess variances of more than two groups
How is variance calculated?
Standard Deviation^2
What does this symbol mean?
<
Less than
What does this symbol mean?
>
Greater than
When is the p-value not significant?
P > 0.05
Greater than 0.05
When is the p-value significant?
P < 0.05
Less than 0.05
What a is method of comparison of two groups that are non-parametric?
Mann-Whitney U test
What are the assumptions of the Mann-Whitney test?
-Data is independent random samples
-Data must be measured at least at the ordinal or continuous level
How do you get a Man-Whitney test in SPSS?
-Analyse
-Nonparametric tests
-Independent samples
-Settings
-Choose tests
What do you use for man whitney non-parametric if the distributions are not the same in SPSS?
-Non parametric tests
-Legacy dialogs
-2 independent samples
What is the non-parametric version of a paired t-test?
-Wilcoxon matched pairs test
When do you reject the H0 in a Wilcoxon matched pairs test?
-If (smaller) T value is lower than or equal to the critical value ( from T-tables)
How do you perform Wilcoxon signed rank test in SPSS?
-Nonparametric tests
-One sample
-Settings
-Choose test
