1/134
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
What starts with the general case (population or model) and then predicts what would happen in many samples?
Probability
What starts with one set of data (sample) and makes inferences about the overall population or model?
Statistics
What variable is the experimental/predictor variable that can be manipulated in an experiment?
Independent variable
What variable is defined as the outcome variable?
Dependent variable
Blood typing is an example of what kind of variable?
Nominal
What variable is defined as categorical outcomes with more than 2 possible outcomes?
Nominal
What variable is defined as categorical outcomes with 2 distinct possible outcomes?
Binomial
Yes/No is an example of what kind of variable?
Binomial
What variable expresses ranks and is influenced by order?
Ordinal
A pain scale measures what kind of variable?
Ordinal
What variable represents data capable of possessing any value in a given range?
Continuous
Blood pressure is an example of what kind of variable?
Continuous
IQ is an example of what kind of variable?
Interval
T/F: The zero point is NOT considered meaningful in an interval variable
True
T/F: The zero point is NOT considered meaningful in ratio variables
False
Temperature in Kelvin is considered what kind of variable?
Ratio
What type of variable can you NOT calculate medians and percentiles?
Nominal
What types of variables can you add or subtract?
Ratio and interval
What types of variables can you calculate median, SD, and SEM?
Ratio and interval
What is the ONLY variable that you can calculate a coefficient of variable?
Ratio
Which of the following is NOT influenced by outliers?
Median
T/F: The mean is influenced by outliers
True
T/F: The median is influenced by outliers
False
What is the benefit of a trimmed mean?
Removes influence from outliers
Which of the following is FALSE about the mode?
Not useful with variables that can only be expressed as integers
T/F: Bias is only caused by the preconceived notions of the experimenter
False
Error is used to refer to _________
Variability
Biased measurements tend to result from what?
Systematic errors
Which of the following terms is defined as the variation among values expressed in the same units as the data?
Standard deviation
1 SD is equivalent to what percentage?
68%
2 SD is equivalent to what percentage?
95%
3 SD is equivalent to what percentage?
99%
What does it mean when the Gaussian distribution curve is called a standard normal curve?
Mean= 0, SD= 1
T/F: The larger the sample size, the smaller the SEM
True
Which of the following statements is FALSE?
The SEM is a measure of the spread of the data
The confidence interval (CI) of a mean is computed by what 4 values?
Sample mean, SD, Sample size, Degree of confidence
The width of the CI is ______ to the sample SD.
Proportional
The CI is ________ to the square root of the sample size.
Inversely proportional
The CI is ________ to the degree of confidence.
Proportional
5 assumptions are accepted when interpreting a CI of a mean. If one of the assumptions is violated, what is the effect?
The true CI will be wider than the calculated CI, the true Cl will be too optimistic
The median is equal to the ______ percentile.
50th
T/F: The SD can be negative.
False
T/F: All bell-shaped curves are Gaussian
False
T/F: The Gaussian distribution is bell-shaped
True
T/F: A 99% CI is wider than a 90% CI
True
What 2 values affect the SEM?
Sample size and SD
T/F: The SEM is always smaller than the SD
True
What is the effect of increasing the sample size on SD?
No effect
Which of the following does the CI size (width) NOT depend on?
Sample median
What term is defined as the two ends of a confidence interval (CI)?
Confidence limits
Which of the following is an example of censored survival data?
All of the above
What method accounts for censored patients when creating a survival table?
Kaplan-Meier
What is the preferred method for creating a survival table, unless the sample size is very large?
Kaplan-Meier
T/F: Pooled data is independent
False
T/F: Patients who die before treatment should not be removed from a survival analysis
True
Why is the median survival time reported instead of the mean survival time?
All of the above
A p-value of ______ is almost universally regarded as the cutoff level for statistical significance.
0.05
T/F: The smaller the p-value, the stronger the evidence to dispute the null hypothesis
True
When the null hypothesis is rejected, the result is said to be __________
Statistically significant
If the p value is less than 0.05, you ________ the null hypothesis
Reject
T/F: The p value is the probability that the null hypothesis is true
False
T/F: The p value is the same as the Type I error rate
False
What error is the incorrect rejection of the true null hypothesis?
Type I
What error is also called a false positive?
Type I
What error is incorrectly retaining a false null hypothesis?
Type II
What error is a false negative?
Type II
T/F: The p value conveys information regarding the size of the observed effect
False
The _____ variables or endpoints in a study, the more likely one of them will come up statistically significant by chance alone
More
T/F: A result with p=0.004 is more statistically significant than a result with p=0.04
False
If you set an α very low, you will make _____ Type I errors
Few
If you set an α very low, you will make _____ Type II errors
More
If the CI for the difference between two means does not include _____ (the null hypothesis), then the result must be statistically significant
0
If the CI for the ratio of two proportions does not include _____ (the null hypothesis), then the result must be statistically significant
1
T/F: Not statistically different means no difference
False
T/F: A high p value proves the null hypothesis
False
If sample size is increased by a factor of 16, CI is expected to narrow by a factor of _____
4
As the power increases, the chances of a Type II error _______
Decreases
T/F: P values can be negative
False
T/F: The concept of statistical hypothesis testing is about making decisions
True
Which of the following is NOT computed from the data?
α value
According to the Familywise Error Rate, more comparisons will lead to a larger chance for a ________
Type I error (false +)
What is the Bonferroni correction?
Divide the α value by number of comparisons
The Bonferroni correction increases the risk for a ________
Type 2 error (false -)
What type of tests quantify how much a data set deviates from the expectations of a Gaussian distribution?
Normality tests
T/F: A normality test can prove the data were sampled from a Gaussian distribution
False
What would fail a normality test?
Small p value
Which of the following should you do if the data fail a normality test?
All of the above
What test is used for non-normal, more robust distributions?
Non-parametric test
What is the simplest robust statistic?
Median
T/F: A normality test asks whether a particular set of data is Gaussian
False
T/F: Outliers may be removed in some instances
True
What test compares the observed and expected numbers of each subject?
Chi square test
What term is defined as the rate of new cases of disease?
Incidence
What term is defined as the fraction of the group that has the disease?
Prevalence
What test compares the prevalence of disease between 2 groups?
Cross sectional study
2 groups of subjects (one with exposure & one without) are observed overtime to determine the incidence rates. What type of study is this?
Prospective study
A single sample of subjects are randomly divided into 2 groups and each group gets a different treatment. Outcomes/Incidences are observed. What study is this?
Experimental study
Two groups of subjects are selected: one with the disease and the other without. Investigators look back to determine possible risk factors. What study is this?
Case control study
How do you calculate the number needed to treat (NNT)?
Take the reciprocal of the absolute risk reduction
Are higher or lower NNT preferred?
Lower