Statistics 4 Flashcards
If we are comparing two datasets where one is parametric and the other non-parametric then what type of tests do we need to carry out and why?
Non-parametric: this is because non-parametric data cannot carry out parametric tests whereas parametric data is capable of carrying out both, we therefore need to carry out tests that ensures they are on a level foundation
What does a one-sample t-test compare?
The difference between a sample and the population mean values to determine if the sample is representative of the population
How do we interpret a one-sample t-test in SPSS?
If the “sig.”, otherwise known as the p value, surpasses the significance threshold then we can be confident to that degree that we should reject the null hypothesis. i.e. the null hypothesis states that there is no statistical significance, but because we reject it we can say that there is statistical significance between the sample and the population
What reasons could there be for statistical significance between a sample and a mean?
lack of units from the population, or misrepresentative units (i.e. biased),
What does a two-sample t-test compare?
The difference between two sample’s middle values to see if there is a statistically significant difference between for example two variables within a population
How do we interpret a two-sample t-test in SPSS?
If the “sig.”, otherwise known as the p value, surpasses the significance threshold then we can be confident to that degree that we should reject the null hypothesis. i.e. the null hypothesis states that there is no statistical significance between the two samples, but because we reject this then we know that there is statistically significant difference between them.
What does SPSS do and produce for a two-sampled t-test that we do not consider yet?
It compares the variances of the two samples beforehand to see if they validly can be compared through their means. It produces a table that consists of two rows; one producing a level of significance for not carrying out the variance test beforehand and one for where it has. We interpret both but mostly the one without
What does an ANOVA test compare?
Compares the mean between more than two samples. By looking at the behaviour within and between them. It is looking to see if there is a statistically significant difference between them
What does relational statistics do?
It tests to see if there is correlation between two variables that are both parametric
What does the range of outcomes stretch between?
-1 and +1
What is the chart to determine the strength of correlation between variables?
- 0-0.19: very weak
- 2-0.39 weak
- 4-0.69 moderate
- 7-0.89 strong
- 9-1.0 very strong
What is a relational statistic?
Product Moment Correlation Coefficient
What is the relational statistics represented as?
r
What is important to remember about correlation?
It does not always mean causation between the two variables
