-
hypothesis is..
speculation about how to explain facts or observations
-
To avoid biases, what type of hypothesis do you test?
null hypothesis
-
What type of question must be posed with every experiment?
research question
-
A hypothesis (best guess) is always written as a
statement
-
A research question tends to be..
A general inquiry that is considered "nondiectional" (does not say whether there is an adverse effect through association)
-
A hypothesis should always be __________ rather than just relying on basic associations
"nondirectional"
-
What are the five steps to testing a null hypothesis for statistical significance
- 1) Develop a null and alternate hypothesis
- 2) Establish the appropriate alpha level
- 3) perform a suitable test of statistical significance on appropriately collected data
- 4) Compare the p value from the test with the alpha level
- 5) Fail to reject the null hypothesis
-
What are the six steps to the scientific method?
- 1) Observe an event
- 2) Develop a model (hypothesis) that makes a prediction
- 3) Test the prediction
- 4) Compare the p value from the test with the alpha level
- 5) Reject or fail to reject the null hypothesis
- 6) Repeat as needed
-
What is biostatistics?
A science that deals with the collection, organization, analysis, interpretation and presentation of information that can be stated numerically with regard to medical, public health and biological applications
-
What are the 4 main activities of an epidemiologist?
- 1) Design Studies
- 2) Collect and organize data,
- 3)summarize results using appropriate statistical tests
- 4) critically interpret findings
-
To be a responsible epidemiologist you should know to not rely on ________ with a statistican
collaboration
-
To be a responsible epidemiologist you should know how to distinguish between _________ data and routine __________
-
A statistic is..
an estimate based on a sample of an unknown numerical quantity in the population
-
If information is based on a ______ of the population, we want to use this information to make inductive inferences about the _________
sample population
-
When data is collected from a sample of a population, what should you do with the data?
make inferences about the underlying population
-
What's the main "equation" of biostatistics?
(Good sample data representative of the population) + (Good application of satistical methods) = (Good estimates of relevant parameters) + (Good decisions about whether or not certain hypotheses are tenable)
-
What are the five things (issues) an epidemiologist must do?
- 1) Have to clearly define the research question
- 2) Must choose the most appropriate study design
- 3) Must select a representative sample for study
- 4) Must collect data appropriately
- 5) Must summarize using appropriate statistical methods in order to make the correct inferences
-
If a study was incorrectly planned or executed, no amount of statistical expertise can _______ the results
salvage
-
What are the 4 main statistical needs in the early stages of research?
- 1) Recommend study designs for a given research question
- 2) estimate the number of subjects (sample size) needed to achieve study objectives
- 3) Develop efficient data collection methods
- 4) Recommend ways to monitor the quality of data as it is collected
-
What are the three statistical needs after the data is collected?
- 1) Recommend most appropriate statistical methods for analysis
- 2) Interpret the findings in understandable terms
- 3) Review and contribute to presentations and publications
-
Why do we need EEh 505?
Because you should have a good understandingstatistics to aid all future collaborations
-
The higher the "impact factor" of a publication the _______ it is
better
-
The impact facotr reflects what?
The number of times articles are cited
-
The impact factor is not necessarily an indicator of _________
quality
-
What are the three ways research is judged?
- 1) It's perceived "impact factor"
- 2) Subsequent frequency of citation in new peer-reviewed research
- 3) Reports of consistent (or inconsistent) findings [whether the results can be replicated or not]
-
"Inference" is another word for..
Prediction (conclusions)
-
Our goal is to make the strongest possible ________ from limited amounts of _____
conclusions (inferences) data
-
What do frequency distribution list?
Each value in the data and the frequency with which it occurs
-
Descriptive statistics are used to..
Summarize the information in a collection of data
-
Inferential statistics are used to..
Provide predictions about a population, based on a sample of the population
-
What are the four things a good scientific study includes?
- 1) Clearly stated objectives
- 2) Clearly stated hypothesis
- 3) Specific effect to be estimated
- 4) Specify defintions of terms essential to the study
-
The inferences we wish to make are limited by what two things?
Study design and selection of subjects.
-
What do we use study units for?
To make inferences about the population
-
Study units make up the _________
Population.
-
Increasing the number of measurements taken on a study unit is not the ________ to increasing the number of study units.
-
The population is...
the total set of subjects of interest in a study
-
The sample is...
the subset of the population on which the study collects data
-
What do we call the quantity computed from the sample data?
statistic
-
What is a parameter?
A quantity that is characteristic of the population
-
If a sample is representative of the population, _____________ will give accurate impressions of the corresponding __________ of the population
descriptive statistics parameters
-
What is a target population?
The whole group of study units to what we are interested in applying our conclusions
-
Target population is also known as..
sampling frame
-
What is the study population?
The group of study units to which we can legitimately apply our conclusions
-
The study population and the inaccessible study units together make up..
target population
-
The accessible study units solely make up the..
study population
-
What are the two main types of epidemiologic studies?
Observational and Experimental
-
Observational studies are where the researcher..
observes the study units, makes measurements and make inferences that are limited to descriptions and associations
-
Experimental studies are where the researcher..
intervenes (experiments) in some way to affect the manner in which the study units(experimental units) respond
-
In a _____ study, you follow exposed and non-exposed individuals over time
Cohort (prospective)
-
In a ______ study, you identify a sample of individuals with the disease and without
Case-control (retrospective)
|
|