Statistics Flashcards
What are the 5 steps in the Scientific Method?
True or False?
The research question is always about the sample size.
False
The research questions are always about the population.
Why do we use Inferential Statistics?
We are unable to test our whole population or do the same test multiple times. Inferential Statistics (P Value) allows us to determine the probability of an effect on a population without testing the whole lot.
What are examples of a nominal set of data?
- Gender
- Hair colour
Characteristics
- Must be unordered - not higher or valued more than any other
- Must be categorical - distinct groups
What are examples of an ordinal set of data?
- Level of education - High school, tafe, uni
- Karate levels - White, brown, blue etc.
Characteristics
- Ordered categories
What are examples of an interval set of data?
- IQ results
Characteristics
- No meaningful 0 point
What are examples of a ratio set of data?
- Age - 1, 2 , etc
Characteristics
- Has an absolute 0
- Must be numerical
What is the difference between a univariate and a bivariate summary?
Univaritate - One variable
Bivariate - Comparing two variables.
What is a qualitative variable?
It is categorical (written)
e.g. hair colour, type of pet etc.
What is a quantitative variable?
It is a numeric variable
e.g. age, height, weight etc.
What does this picture represent?
It shows sampling variability.
There will be variability across samples. BUT if we take the average of all these scores, it will be a representation of our population
If your p-value is 0.8, would you reject the H0?
Yes, the further away from 0.05, reduces the likelihood that H1 is supported
What are the five steps to testing a hypothesis using a computer program?
- Decide on alpha
- Calculate test statistic
- Calculate p-value from test statistic
- Compare p-value with the alpha level
- If the p-value is:
- Less than 0.05 (alpha) reject H0
- Greater than 0.05 (alpha) don’t reject H0
What are the 6 types of statistical tests we have learnt?
- 1-sample t-test
- Independent samples t-test
- Paired samples t-test
- Pearson’s correlation
- Chi-squared goodness of fit test
- Chi-squared test of independence
If you have one numeric variable and you are comparing the mean of this variable to another known mean, what statistical test would you conduct?
One-sample t-test