Exam Flashcards
Confidence intervals mean
We are 95% confident that the sample mean difference will between these two value, which are two standard deviation away from the the mean.
When is it is experiment or observational?
Observation
Experiment, pre and post outcome measure with the implementation of an intervention.
Central limit theorem
says that distribution of sample means is approximately normal is the sample size is sufficient. Since t test and ANOVA based on normal distribution, this mean s that we can use these methods, even if the data os slightly skewed, particularly if sample size is large.
What is a residual
The difference between the observed value and the value predicted by the model.
What does omitting an outlier do?
Omitting an outlier reduce the residual variability and would likely increase the strength of evidence of the difference.
Why might grouping data in one-way anova preferable to a linear regression model using individual number
Improves power since relationship may not be linear.
List three assumptions of anova and comment on validity
Independent group ie random sampling
Equal variances –> within one ie IQR
Normality –> median in centre
Assumptions about linear regression
independent observations
Iinear association
normal variability
Equal variances
Why do you look at change in something rather than the value itself
Baseline and post intervention values are not independent
why use parametric over non-parametric
if assumptions are ratified then parametric tests are more powerful than non-parametric. Parametric also prove direct estimate for effects, including confidence intervals.
What is meant by sampling variability in a statistical studyy
Sampling variability refers to the process whereby statistics, such as the sample mean, would give different results if the random sampling process was repeated. We thus need to account for sampling variability when making any conclusions from our data.
