Stats Flashcards
What is inferential stats?
Make generalisations about the population.
Expressed as a p-value
What is univariate data?
Only answers simple questions.
1 variable.
What is bivariate data?
Answers how,why,when.
2 catergorical, 2 continuous or 1 categorical or 1 continuous.
What are residuals?
Unexplained variance.
Closer to 0 the less prediction error.
What does linear regression explain?
How well one variable predicts another
What do scatterplot show?
Orientation and pattern of data.
How much/type of relationship.
What does multiple testing result in?
One of the tests becoming significant by chance.
What are parametric statistical models?
Make assumptions about the input they receive.
What predictor is linear regression?
Continuous
What predictor is ANOVA/ t-tests
Categorical
What does multiple r squared show?
X100
Shows the % variance explained by the model.
What are independent t tests?
Establishes whether 2 means collected from independant samples differ significantly
What are paired sample t-tests
Where her mean difference between two sets of observations is 0.
What are paired observations ?
Data collected from same participant for both sets of observations.
What’s an ANOVA?
Analysis of variance
Estimated significant differences between a set of means.
Compared variance within and between groups
When can issues with multiple testing happen?
Unplanned analysis.
Analysing before data collection is finished.
Looking for differences among groups using a number of outcome measures.
Why use inferential data analysis?
Make important decisions about theories, interventions-need to be sure data is accurate
Statistically significant.
Reliability.
How do you choose which inferential test to use?
Comparing 2+ groups- ANOVA or t-test.
Relationship between variables- Pearson’s correlation or linear regression.
When should you use a t-test?
2 groups /conditions, between design= independant t test
Within design =paired t test.
When to use a correlation?
Continuous predictor, association with the relationship.
When to use linear regression?
Continuous predictor, prediction of relationship.
When to use ANOVA!
Categorical predictor, 3+ groups
What’s the bull hypothesis?
no difference between certain characteristics of a population
What do we conclude if the p value is bigger than 0.05?
We fail to reject the null hypothesis. The evidence is inconclusive.
How do you work out df for groups in anova?
Number of groups -1
How do you work out the dd for residuals in ANOVA?
Number of observations - groups.
How do you work out the mean square values in ANOVA?
Sun of squares- divided by df.
