Module 6: Quantitative Research Flashcards
Data, numbers, “quantity” - Is what type of research?
QUANTITATIVE RESEARCH
Feelings, emotions, thoughts, opinions, “narrative” - Is what type of research?
QUALITATIVE RESEARCH
Which variable is the one being manipulated?
INDEPENDENT VARIABLE
Which variable is the one being changed or influenced by the other?
DEPENDENT VARIABLE
What variable is an external variable that is either not controlled, OR is not recognized during measurement, that also AFFECTS the outcome of the dependent variable?
EXTRANEOUS/CONFOUNDING VARIABLE
What is a Type I Error?
Rejecting the null when it is true
So there is no difference but you are seeing one, a FALSE POSITIVE
What is a Type II Error?
Accepting the null when it is false
So there is no change or difference but that is false or incorrect because there IS A CHANGE, a FALSE NEGATIVE
How do you remember Type I and Type II Errors?
RAAR
REJECT the null when you should be ACCEPTING, is TYPE I
ACCEPT the null when you should be REJECTING, is TYPE II
How do you control for extraneous variables?
Use a control group, or have inclusion/exclusion criteria
What is repeated measures mean?
Testing is done at specific intervals throughout the trial
What is Quasi-experimental study designs?
includes an intervention or treatment, but does NOT randomize participants
What does an R-value of 0.7 to 0.8 or higher mean? **
0.7 ++ = STRONG CORRELATION
What does an R-value of 0.4 to 0.7 mean? **
0.4 to 0.7 = MODERATE CORRELATION
What does an R-value of less than 0.4 mean? **
> 0.4 = WEAK CORRELATION
Define Reliability.
Consistency in measurement (the closeness of the initial estimated values to the subsequent estimated values)
Is it repeatable? Will repeating it produce similar results?
CLOSENESS (RC)
Define Validity.
The extent to which the conclusions drawn from a statistical test are ACCURATE
Are the results accurate and well-founded? Are the instruments used measuring what they are intended to measure?
aka PRECISE
ACCURATE (VA)
Define Internal Validity.
The extent to which the independent variable, rather than some extraneous variables, actually caused the outcome of the study.
What are some threats to Internal Validity?
- History (specific events which occur b/t the 1st & 2nd measurement)
- Maturation (process within subjects which act as a function of the passage of time)
- Selection of Subjects
- Mortality/Attrition (loss of subjects)
- Testing
- Instrumentation
- Regression to the Mean (AKA Statistical Regression)
Define External Validity.
The extent to which the results of your study can be generalized to another place, population, and/or time point.
What are 3 threats to External Validity?
- People
- Place (city vs. rural)
- Time (effects of the pandemic)
Define Statistical Conclusion Validity.
The extent to which relationships among the variables produced by the statistical analyses are accurate.
What are 5 threats to Statistical Conclusion Validity?
- Low statistical power
- Violation of the assumptions of statistical threats
- Fishing (have no statistical analysis plan, looking for answers)
- Reliability of measures
- Reliability of treatment implementation
What is the extent to which researchers are actually measuring the theoretical concepts or constructs in the study?
OR it looks at how well the dependent variable as well as the intervention represent the underlying construct in an experiment.
CONSTRUCT VALIDITY
What are 3 threats to Construct Validity?
- Inadequate pre-operational explication of constructs
- Mono-Method Bias (refers to measurements or observations, not your programs or causes. AKA the researcher has only chosen one method to measure/evaluate).
- Expectation Bias (AKA Researcher expectancies)
Define ERROR.
The difference between the true value of a measurement and the actual recorded measurement.
Define random error.
Due to chance, due to normal variability that you find in people or items, cans be minimized by having a large enough sample size to compensate for that random error.
Define systematic error/bias.
An error not due to chance meaning that multiple replications of the study, would still give an incorrect result
What are the examples of systematic error/bias?
- Hawthorne effect
- Attrition bias
- Reporting bias
- Researcher bias (including confirmation bias, publication bias, allegiance bias)
What type of bias is this?
The common action of study participants changing their actions or responses to questions when they know they are being studied.
Hawthorne Effect
What type of bias is this?
When there are systematic differences between groups in the number of withdrawals from the study.
Attrition bias
Attrition = the loss of study units from a sample
What type of bias is this?
Bias of researchers in how they handle their data or how they interpret their data
Researcher bias
What type of bias is this?
When a researcher selectively interprets the information to confirm what they believe, or hope to find.
Confirmation bias
What type of bias is this?
When a researcher seeks to publish only the results that are statistically significant, OR that support their hypothesis and they fail to publish results that aren’t statistically significant.
Publication bias
What type of bias is this?
Describes the effect of a funding agency or employer on publishing their results.
Allegiance bias