Statistics Flashcards
(14 cards)
What is ordinal data?
Categorical data where the categories have a natural, ordered ranking, however, the distance or interval between the categories is not necessarily equal or quantifiable
E.g. pain scale, letter grades, positions in a race
What is nominal data?
Categorical data where there is no inherent order or ranking. Names are distinct, but one isn’t “more” or “less” than another
E.g. eye colour, gender, types of pets
Can only calculate mode from this data
What is a parametric test?
A type of statistical test that makes certain assumptions about the population from which your data is sampled. These assumptions typically involve the shape of distribution in the population and the knowledge of certain population parameters
What are the key assumptions of parametric tests?
- Normality of the data
- Homogeneity of variance (homoscedasity): if you are comparing two groups, the spread (variance) of data in each group is roughly equal
- Independence
- Interval or ratio data: the data being analyzed is measured on an interval or ratio scale (means the differences between values are meaningful and consistent, and there may be a true zero point for data)
What are some common examples of parametric tests?
T-tests: used to compare the means of one or two groups
ANOVA (analysis of variance): Used to compare the means of three or more groups
Pearson correlation coefficient: used to measure the linear relationship between 2 variables
Linear regression: used to model the relationship between a dependent variable and one or more independent variable
Why are the assumptions in Parametric tests important?
To ensure validity and reliability of their results. If these assumptions are significantly violated, the conclusions drawn from the test might be inaccurate. May need a non-parametric test in this instance.
What is a one-tailed test?
A type of hypothesis test used to determine if there is a statistically significant difference or relationship in a specific direction.
Requires a null hypothesis to be set up
In a one-tailed test the alternative hypothesis specifies the direction of the difference or relationship (e.g. the mean of group A is less than the mean of group B)
When should you use a one-tailed test?
Appropriate only when you have a strong, a prior reason to believe that the effect will occur in one specific direction. This belief should be based on prior research, established theory or a very clear understanding of the phenomenon you are studying.
E.g. testing if a new drug reduces blood pressure (you are only interested in a decrease, not an increase)
What is a two-tailed test?
The alternative hypothesis simply states that there is a difference between groups or relationship between variables, without specifying the direction of that difference or relationship.
What is the null hypothesis?
States that there is no effect or difference with what we are testing and this is studied with the alternative hypothesis that there is an effect of difference
What is the definition of the p-value?
Essentially tells you whether or not it is likely that your results occurred purely by chance or not
E.g. a p-value of 0.01 tells you that there’s only a 1% chance of seeing your results if your intervention did nothing
It is a measure of the evidence against the null hypothesis
What is a type 1 error?
Essentially a false positive. You conclude there is a difference when in fact there is none.
Denoted by alpha, typically the p-value which you set. If your p-value is less than alpha, you reject the null hypothesis, accepting the risk of a type 1 error.
What is a type II error?
Occurs when you fail to reject a false null hypothesis (false negative)
E.g. you conclude there is no effect or difference when there was one
What is the relationship between type I and type II errors?
Inverse relationship.
