Font size
WorksheetsIPP3 BIOSTAT DR. HIGGINS
Total questions: 67
Worksheet time: 3hrs 52mins
The nature of the OUTCOME
DEPENDENT variable
INDEPENDENT variable
The number of groupings of the
DEPENDENT variable
INDEPENDENT variable
If the p-value from a study result was 0.002 (<0.05)? Select all that apply
Shows a statistically significant difference between the two groups
REJECT the null hypothesis
ACCEPT the null hypothesis
Does NOT show a statistically significant difference between the two groups
If the p-value from a study result was 0.06 (>0.05)? Select all that apply
REJECT the null hypothesis
Shows a statistically significant difference between the two groups
ACCEPT the null hypothesis
Does NOT show a statistically significant difference between the two groups
P-value by itself do not provide a good measure of evidence regarding a model or hypothesis
True
False
Which of the following are parameters for Clinical Significance? Select all that apply
Assess importance of difference
Expert/personal opinion
Extent of change in clinical practice or change in quality-of-life measures
P <0.05
Which of the following are parameters for Statistical Significance? Select all that apply
Detect non-chance differences
P <0.05
Sample size, patient variability, accurate/reliable measures
Expert/personal opinion
Parametric test makes assumptions on? Select all that apply
Normality
Independence
No outliers
Equal variance
What are the advantages of Non Parametric? Select all that apply
More statistical power when assumptions of parametric tests are violated
Assumptions of normality do not apply
Small sample sizes are OK
Used for all data types
Used with data outliers
The test statistic generated from the ANOVA is ?
T
F
In order to determine which groups, the difference is between, must complete a subgroup t-test analyses in 3 or more groups.
ANOVA
T-test
In ANOVA test can do a t-test called a Bonferroni correction to correct this issue.
True
False
Which of the following is TRUE regarding Chi-Square test? Select all that apply
More than two categories
outcome is NOMINAL
Do NOT have an intrinsic order
Two groupings of independent variables
Chi-Square calculated and converted to a p-value of 0.025.
Chi-Square test is extremely sensitive to large sample sizes AND can only determine whether two variables are related.
True
False
If, in any one cell, there are less than 5 counts which of the following I'm going to use in place of the Chi-Square test?
Fisher's Exact test
Bonferroni correction
Which of the following test Results could have been a Type 1 (alpha) error and we REJECT the null hypothesis?
chi-square analysis
Fisher’s Exact test
Which of the following test Results could have been a Type 2 (beta) error and we ACEPT the null hypothesis meaning the results of the chi-square test are correct.?
Fisher’s Exact test
chi-square analysis
Similar to the t-test EXCEPT it is used in situations where subjects serve as their own controls (Before and After).
Paired t-test
Fisher Exact test
Bonferroni correction
T-test
The main change is that we are comparing two medications, for example:
• Provide them with 30 days of drug A, then test their cholesterol level (First grouping)
• Have them stop taking drug A and wait a period of time to allow drug A to be eliminated from their body (washout period)
• Provide the same person with 30 days of drug B, then test their cholesterol level (second grouping)
paired t-test
t-test
Chi-square test
Wilcoxon Mann-Whitney test
Which of the following tests is represented in this image?
T-test
Mcnemar Test
Paired t-test
Wilcoxon Mann-Whitney test
W-M-W TEST LIMITATIONS does not explain why there is a difference AND becomes less accurate when the sample size is very small or very large
True
False
The test statistic generated from the K-W test is "H" and is then converted into a p-value for inference purposes
True
False
The advantage of this type of test is a statistically powerful outcome when ASSUMPTIONS are met.
Parametric
Non Parametric
Type of significance that is determined through change in quality of measures.
Clinical significance
Statistical significance
Chi-Square Test
T-test
Paired t-test
Mann whitney test
rejecting null hypotesis
accepting null hypotesis
ANOVA
Mann Whitney Test
Kruskal Wallis Test
T-test
Bonferroni correction
Fisher Exact Test
ANOVA
Kruskal Wallis Test
Chi-Square Test
Mcnemar Test
Underlying disease risk
Cell source
Pretransplant
Donor relationship
Kruskall Wallis Test
Wilcoxon Mann Whitney Test
Chi-Square test
T-test
Bonferroni correction
Fisher Exact test
T-test
Degree of freedom
T-test
Paired t-test
Chi-Square test
ANOVA
Parametric
Discrete
Continuous
Quantitative
p value
t value
f value
u value
Ordinal
Continuous
Quantitative
Parametric
Nominal
Independence
Dependence
normal distribution
positive skew
negative skew
Hint: Ranks
paired t test
t-test
Wilcoxon Mann Whitney Test
Kruskal Wallis Test
Bonferroni correction
T-test
Fisher Exact test
Chi-Square test
T-test
Chi-square test
continuous
nominal
ordinal
Hint: interval outcome
ANOVA
Chi-square test
Kruskal Wallis test
Hint: nominal, before and after
Mcnemar test
T-test
Paired t-test
Mann Whitney test
Chi-Square test
T-test
Wilcoxon Mann-Whitney test
Paired t-test
T-test
Chi-Square test
Wilcoxon Mann-Whitney Test
Type 1 (alpha error)
Type 2 (beta error)
ANOVA
T-test
Paired T- test
Chi-Square test
outlier
defense
Chi-Square test
Mcnemar test
Paired t-test
Wilcoxon Mann Whitney Test
ANOVA
Chi-Square test
T-test
Wilcoxon Mann-Whitney test
What is the dependent variable (outcome)?
Cost of respiratory medications per hospital stay
X Axis
What type of variable is the outcome?
Discrete
Continuous
How many groups of the independent variable are there?
2
1
3
4
What is relation of the independent variable to one another?
Independent since same patients are not in each group
Dependent since same patients are not in each group
What test would you utilize?
Paired t-test
T-test
Chi-Square test
Is the result statistically significant?
**Hint: p value
Yes
No
Based on the result what would you do with the null hypothesis?
hint: p value
Accept
Reject
P-value set a priori as < 0.05 allows for what type of error?
Type 1 error of less than 5%.
Type 2 error of less than 5%
Type 1 error of less than 50%
Type 2 error of greater than 5%
Which result would you accept the null hypothesis with alpha set as 5%
P value < 0.001
P value 0.029
P value 0.063
P value 0.042
Which is NOT an assumption of a parametric test?
Normality
Outliers
Independence
Equality of variance
A study compares atorvastatin, rosuvastatin and pravastatin for 1000 patients with hypercholesterolemia to determine cholesterol lowering effect on average cholesterol 3 months after starting the designated drug. What test would you use?
hint: 3 variables
cholesterol levels are a ratio (continuous)
T-test
Chi-Square
Wilcoxon-Mann-Whitney
ANOVA
A statistically significant difference is found for atorvastatin vs rosuvastatin vs pravastatin but what test is needed to determine between which groups the difference occurred?
Hint: Ratio, 3 or more independent
Dunn test
Bonferroni correction
Fisher’s exact
Kruskal-Wallis test
T-test
ordinal
nominal
dependent
continuous
Chi Square
Continuous
Nominal
Ordinal
Dependent
McNemar test
Discontinuous
Ordinal
Dependent
Kruskal-Wallis test
Ordinal
Continuous
Nominal
Dependent
