Exam 1 Part 3 Flashcards
Summarize a sample of data using graphs and numerical measures such as means or percentages
Descriptive Statistics
Use the sample to reflect back on the population the sample represents
Inferential statistics
What are the two main forms of statistical inferencing?
Confidence intervals, hypothesis testing
Data is collected separately for both groups.
Independent samples
Individuals are measured twice (e.g. before and after treatment), or individuals are matched in pairs (e.g. identical twins).
Dependent/Paired/Matched Samples
The proportion of times an event will occur in the long run. Gives us the chance, or likelihood of that event occurring.
Probability of an event
If the p-value is less than .05, what do we say?
There is sufficient evidence to support that there is a mean/median difference between ____ (Ha). We will reject H0.
If the p-value is more than .05, what do we say?
There is insufficient evidence to support that there is a mean/median difference between _____ (Ha). We fail to reject H0.
Error that occurs when we reject H0 when in fact H0 is true (i.e., conclude there is a significant result when in fact there is not).
Type I
Error that occurs when we fail to reject H0 when in fact Ha is true (i.e., conclude the result is not significant when in fact a significant difference or relationship exists).
Type II
The probability of rejecting H0 when in fact Ha is true (i.e., correctly conclude there is a significant difference).
Power of a test
Name an example of a parametric test
Paired t-test
Name an example of a non-parametric test
Wilcoxon signed rank test
What statements are true about a parametric test?
First choice for interval/ratio data, must assume a normal population, sensitive to outliers
What statements are true about a non-parametric test?
Our choice for I/R data if the normality assumption is not met, Do not need to make any assumptions about the population, Are based on ranks; not affected by outliers
How are the hypotheses written for a paired t-test?
H0: The population mean difference is zero (there is no difference, on average)
Ha: The population mean difference is not zero (there is a difference, on average)
How are hypotheses written for a Wilcoxon Signed Rank test?
H0: The population median difference is zero (no difference between before & after)
Ha: The population median difference is not zero (there is a difference between before & after)
How would you write the hypotheses in the following situation (if it was a paired t-test): Was there a significant mean change in BMI from admission to discharge?
H0: There is no mean difference in BMI from admission to discharge.
Ha: There is a mean difference in BMI from admission to discharge.
How would you write the hypotheses in the following situation (if it was a Wilcoxon signed rank test): Is there a significant (median) difference in duration of middle-ear effusion between breast-fed and bottle-fed babies
H0: There is no median difference in duration of middle-ear effusion between breast-fed and bottle-fed babies.
Ha: There is a median difference in duration of middle-ear effusion between breast-fed and bottle-fed babies.
How are hypotheses written for a two sample t-test?
H0: The means of the two populations are equal
Ha: The means of the two populations differ