PCM: Biostatistics 2 Flashcards
Name the two phases to statistical analysis.
Descriptive statistics and statistical inference
______ statistics are used to describe key study variables. Whereas ____________ is employed to make generalizations or to draw inferences about larger populations based on study data.
Descriptive; statistical inference
The simplest study has how many primary outcome variables?
One
For analytic purposes, variables are classified as ______ or discrete.
continuous
_____ variables, sometimes called quantitative or measurement variables, assume any value between a theoretical min and max. For example protein expression levels and serum biomarkers.
Continuous
_______ variables can be thought of as having a fixed number of categories. For example the categories male or female.
Discrete
Descriptive stats for continuous variables should always include a measure of location (i.e. the mean), and a measure of _________ (i.e. the standard deviation).
Variability
What is the appropriate measure of a typical value in the presence of extremes that may inflate the SD or Mean?
Median (i.e. the value that holds 50% of the values above it and 50% of the values below it)
For a normal distribution, the range (min-max) is approx. equal to the mean plus or minus ______ standard deviations. Any variables beyond that are considered extremes (outliers).
3 SDs
Limit the number of decimal places in tables to ______ decimal place(s) more than the original unit of measurement, so as not to overstate the precision of the analysis.
one
In relation to graphical displays, __________ plots are most appropriate for continuous outcomes.
Box-and-whisker plots
In relation to graphical analysis, ________ are the most appropriate for ordinal outcomes.
Histograms
In relation to graphical analysis, ________ are the most appropriate for categorical outcomes.
Bar charts
________ plots are appropriate when both variables are continuous.
Scatter plots
The ____________ refers to the entity on which measurements are made. For example, an individual.
Unit of analysis
_________ groups are physically separate and are comprised of distinct sampling units (e.g. different experimental units assigned to the active drug versus placebo).
Independent groups
_________, matched, or paired groups are often produced when teh same sampling units are measured twice (e.g. before and after an exposure) or when the sampling units are paired (e.g. siblings, litter mates).
Dependent groups
Name the two general areas of statistical inference.
Estimation and hypothesis testing
In ________ of statistical inference, we generate confidence interval population parameters based on sample data appropriately accounting for sampling variability.
estimation
In ______________ of statistical inference, we formally compare population parameters based on sample data, again accounting for sampling variability. We set up competing hypotheses (null and research hypotheses).
Hypothesis testing
The null hypothesis represents:
That there will be no difference or no effect between two outcomes based on separate (although possibly identical or non-identical) treatment.
The research hypothesis is:
States the anticipated or hypothesized difference or effect outcome between two identical or non-identical treatments.
_________ is the probability of observing a test statistic as large or larger than that observed if the null hypothesis were true.
The p value
A small p value, <________, would suggest that there is less than a 5% probability of observing a difference as large or larger than that observed in the study sample.
p value <0.05
A p value <0.05 would likely lead to rejection or acceptance of the null hypothesis in favor of the research hypothesis?
Rejection, in favor of the research hypothesis.
__________-sample studies are most useful when investigating new techniques or technologies.
One-sample
Is confidence interval proportional to sample size?
No, inversely proportional. As sample size goes up, 95% confidence intervals tend to shrink in size.
When there are more than two independent groups, the procedure to test for differences in means is _______.
Analysis of variance (ANOVA)
T/F: in ANOVA, there are >2 independent groups, variances among groups are assumed to be equal and data are assumed to follow a normal distribution.
True
If your number of independent sampling units is 5 and you are going to take 6 measurements of each over the period of the study, you will use this type of analysis:
Repeated-measures ANOVA
When there are two or more independent groups and the outcome is dichotomous, the procedure to test for differences in proportions is the ________ test.
Chi-square test
