Introduction to Biostatistics in Epidemiology Flashcards
What are the three notes about human studies?
- Data will be collected on desired variables
- Comparisons are commonly made
- inferences will be made about the sample derived data and their comparisons
What is a null hypothesis?
research perspective which states there will be no (true) difference between the groups being compared
What are the 3 key attributes of data measurement?
- order/magnitude
- consistency of scale/ equal distances
- rational absolute zero
What are the 3 levels of data?
nominal
ordinal
interval or ratio
Describe nominal data
named categories
no order or magnitude
no consistency of scale or equal distances
simply labeled variables without quantitative characteristics (or dichotomous/binary)
Describe ordinal data
ordered
yes order and magnitude
no consistency of scale or equal distances
Describe interval/ratio data
interval: arbitrary zero value
ratio: absolute zero value
yes order and magnitude
yes consistency of scale or equal distances
After data is collected, we can appropriately go _____ in specificity/detail of data measurement, but never ____
down; up
What is variance?
the average of the squared differences in each individual measurement value and the groups mean
What is standard deviation?
square root of variance value
Graphical representation shows ____ of data
Shape
Normally distributed= _______
symmetrical
What are stats test useful for normally distributed data called?
parametric test
Describe a positively skewed graph.
asymmetrical distribution with one tail longer than another
mean is higher than median (tail points to right)
A distribution is skewed anytime when __________
median differs from the mean
Describe a negatively skewed graph
asymmetrical distribution with one tail longer than the other
mean is lower than median (tail points to left)
What is skewness?
a measure of asymmetry of a distribution
What is kurtosis?
a measure of the extent to which observations cluster around the mean
What is positive kurtosis?
more cluster
What is negative kurtosis?
less cluster
What are the required assumptions of interval/ratio data?
- normally distributed
- equal variances
- randomly derived and independent
How do you handle data that is not normally distributed?
- use a statistical test that does not require the data to be normally distributed (non parametric test)
- transform data to a standardized value ( z score or log transformation)
What is a type 1 error?
Not accepting the null hypothesis when it is actually true, and you should have accepted it, false positive
What is a type 2 error?
Accepting the null hypothesis when it is actually false, and you should have not accepted it, false negative
What is power (1-B)?
the statistical ability of a study to detect a true difference, if one truly exist between group comparisons and therefore the level of accuracy in correctly accepting or not accepting the null hypothesis
The larger the sample size, the ______ the likelihood of detecting a difference if one truly exists.
greater
How do you determine a sample size?
- minimum difference between groups deemed significant
- expected variation of measurement
- Type 1 and 2 error rates and confidence interval
What is the p value test?
statistical test determine possible error rate or likelihood of chance in comparing difference or relationship between variables
A p value is obtained, based on the probability of observing, due to _______ alone, a test statistic value as extreme or more extreme than actually observed if groups were ______
chance; similar
If the p value is ____ that the pre selected alpha value (usually 5%), then it is statistically ______
lower, significant
If the p value < alpha percentage risk of error, we ____ the null hypothesis
reject
What are the require assumptions of interval/ratio data?
normally distributed
equal variances
randomly derived and independent
What are confidence intervals based on?
variation in sample
sample size
How do you interpret a 95% confidence interval?
We are 95% confident that the true difference (0) or relationship (1) between the groups is contained within the confidence interval range
If the confidence interval crosses 1.0 or 0.0 it is _______
not significant
What do you always ask when reviewing the findings of a study?
does “statistical” significance actually confer meaningful, “clinical” significance?
What four questions do you ask when selecting the correct statistical test?
- What data level is being recorded?
- what type of comparison/assessment is desired?
- How many groups are being compared?
- is the data independent or related (paired)?
What is correlation?
provides a quantitative measure of the strength and direction of a relationship between variables
Nominal correlation test = _______ coefficient
contingency
ordinal correlation test = _____ correlation
spearman
interval correlation test = ______ correlation
Pearson
p> .05 for a pearson correlation just means there is no ______ correlation, there still may be _______ correlations
linear; non-linear
What type of data is gender?
nominal
What is a survival test?
compares the proportions of events over time, or time to events, between groups
What is a survival test commonly represented by?
kaplan-meier curve
What are regressions?
provide a measure of the relationship between variables by allowing the prediction about the dependent, or outcome, variable knowing the value/category of independent variables
What does a chi square test do?
both tests compares group proportions and if they are different from that expected by chance
Are multiple chi square test acceptable?
never
What does the Bonferroni test of inequality (Bonferroni correction) do?
adjusts the p value for # of comparisons being made, very conservative
What determines if a test is related/paired?
comparison from past to present
pre vs post
before vs after
What do the Mann-whitney and kruskal- Wallis test do?
compares the median values between groups
What does the Wilcoxon signed rank test and friendman test compare?
the median values between groups
What does the student- Newman keel test do?
compares all pairwise comparisons possible, all groups must be equal in size
What does the Dunnett test do?
compares all pairwise comparisons against a single control, all groups must be equal in size
What does the Dunn test do?
compares all pairwise comparisons possible
useful when all groups are NOT equal in size
what does a student t-test and analysis of variance (ANOVA) compare?
compares the means of all groups against a dependent variable
What does an analysis of covariance (ANCOVA) do?
compares the means of all groups against a dependent variable while also controlling for the co variance of confounders
What does a paired t-test and repeated measure ANOVA compare?
the means of all groups of related data against a dependent variable
What does a repeated measures ANCOVA compare?
the means of all groups against a dependent variable while also controlling for the co variance of confounders
What does a tukey or scheffe test compare?
all pairwise comparisons possible, all groups must be equal
What is a kappa statistic?
a correlation test showing relationship or agreement between evaluators
What does a +1 kappa interpretation mean?
the observers perfectly classify everyone the same way
What does a 0 kappa interpretation mean?
there is no relationship at all between the observers classifications, above the agreement that would be expected by chance
What does a -1 kappa interpretation mean?
the observers classify everyone exactly the opposite of each other