eM1 Flashcards
The first e-module on medlearn
What is a research question?
It is a question that provides the frame for the entire research project. Everything read, experimented and discussed should relate back to it. They can be quite broad and non specific or narrow/ specific.
What are some examples of a (i) broad and (ii) narrow research question using osteoporosis and exercise?
Ex. Does exercise prevent osteoporotic fractures? Ex. Can walking 4 miles a day prevent osteoporotic fractures in postmenopausal women over 55?
What is a hypothesis?
A statement that predicts a new finding before the answer is known. Stated at the outset of a piece of work and tested with statistical tests after results.
Why do we use a hypothesis?
To evaluate a research question- may be used in conjunction with significance testing
What are the two types of hypothesis?
- Null hypothesis 2. Alternative (or several) hypotheses
What is a null hypothesis (H0)?
There is no dependent relationship between two variables. Often not stated but is tacit.
What is an alternative hypothesis?
There is a specific and reproducible relationship between variables.
Should we use directional hypotheses?
It may be helpful but remember to check for changes in both directions when analysing data to avoid missing an effect.
How do I reject a null hypothesis?
Once a hypothesis is stated and results collected we can use a statistical test to test its validity and can determine whether to reject or fail to reject the null hypothesis. NOTE: we do not accept- only fail to reject
What is a p-value for?
The p stands for probability. It refers to the likelihood that the observed difference was observed by chance. It is part of an output in statistical tests.
What can a p-value be?
Between 0-1. It calculates the probability that your observed results were real (i.e. this small group reflects the population).
How do we use a p-value?
To reject or fail to reject the hypothesis. This is done by comparing the calculated probability to the significance level.
What is an alpha value?
The significance level, denoted by (alpha sign). It is the probability of rejecting a null hypothesis when it is true (i.e. observing a difference when it is not there)
What does a significance level of 0.05 mean?
A significance level of 0.05 indicates a 5% risk of concluding that a difference exists when there is no actual difference. That is to say rejecting the hypothesis when it should not be rejected.
What is the boundary for significance level?
Often, authors say that the p-value indicates a significant difference when alpha=0.05 (or less). The p-value is highly sensitive when alpha=0.001 (or less)
