General Statistics
Is the test appropriate?
Measurement Scales
Study Designs
Review
100
Statistical tests that determine the likelihood or probability that the null hypothesis is true
What are inferential statistics
100
Hypothesis: Older people sway more than younger people Test used: Unpaired t-test
Yes
100
This type of measurement scale is concerned with categories by labels or numbers. Can be nominal or ordinal.
What is non-parametric
100
Why is study design important when considering statistical analysis
By considering the study design you can determine if the statistical test used was appropriate. Using the appropriate statistical test ensures the validity of the p-value and determines the probability that the hypothesis is true.
100
What does PICOT stand for?
Population Intervention Comparison Outcome Time
200
This p-value indicates a high probability that the null hypothesis is true, also known as the critical p-value
What is >.05
200
Hypothesis: People sway more with eyes closed than with their eyes open Test used: unpaired t-test
No- the comparison is between two variables in one group, should use a paired t-test
200
This type of measurement scale is concerned with numbers only. Can be ratio or interval.
What is parametric
200
This type of test is appropriate when comparing two conditions in the same group
What is a paired t-test
200
A hypothesis in which there is no difference or change
What is a null hypothesis
300
The probability of committing a type I error
What is alpha
300
Hypothesis: Body sway will differ among young, healthy-old, and at-risk old adults Test used: multiple t-tests
No- potential for error increases 5% with every test, therefore you should use one-way ANOVA
300
This test is used for a nominal measurement scale to compare two independent groups
What is chi-square
300
This type of study determines an association between 2 or more variables
What is an association study
300
When the confidence in a cause and effect relationship is compromised in a study
What is internal validity
400
This determines an association between two or more variables
What is a correlation coefficient
400
Hypothesis: the blue bag of trail mix will have more M&M's than the red bag Test used: One-tailed t- test
Yes
400
This type of test is used for parametric measurement scales that compare two independent groups
What is an unpaired t-test
400
This type of test is used to compare three or more conditions within the same group
What is repeated measures one-way ANOVA
400
The population actually being used in a study
What are research samples
500
This test is used for a non-directional hypothesis
What is a two-tailed t-test
500
Hypothesis: Young, old and at-risk old people sway differently with eyes closed, open and with feet together and apart Test used: One-way ANOVA
No- comparing different conditions across independent groups, mixed design, should use 3x4 ANOVA (two-way ANOVA)
500
A red flag concerning measurement scales
Using parametric tests for non-parametric scale measures
500
What happens when you use a paired t-test instead of unpaired or vice versa
You get a totally different p-value
500
Why is it important to have a hypothesis prior to experiment?
1. ensure the experiment is designed well 2. to avoid vague, unhelpful, and unfocused investigation 3. to provide direction
M
e
n
u