Unit 4 Flashcards
The standard error of the mean (σX̅)
a function of standard deviation and sample size
Confidence Intervals
Range of sample data to determine a range of population means that are NOT significantly different from the sample mean
Point Estimate
a single value obtained directly from the sample data (i.e., the mean)
Interval Estimate
Range of sample data that we suspect (with a degree of confidence) within which the population mean resides
Confidence Interval (CI)
Range of scores with specific boundaries (confidence limits), that should contain the population mean
How to develop a confidence interval (CI)?
Capture 95% of the area under the “normal” distribution
This occurs at a standard deviation (z) of + 1.96.
Find: sample mean, and standard error of the mean
What happens if a sample size is small?
- Usually performed with samples smaller than 30
- Use the t-distribution, not z-distribution
Variability _______ with _______ sample sizes
Decreases, Larger
What does t-distribution depend on?
Sample size
Hypothesis
“a declarative statement that predicts that relationship between the independent and dependent variables, specifying the population will be studied.”
Research Hypothesis
states the researcher’s true expectation of results guiding the interpretation of outcomes and conclusions
Null (statistical) hypothesis - H0
proposes that there is no difference between certain characteristics of a population or data-generating process.
Alternative Hypothesis - H1 (HA)
one that states there is a statistically significant relationship between two variables
What is a research hypothesis?
Based on the physiology of aging, intelligence measurements, etc.
Sample size affecting a statistical power:
tighter distribution….. decrease β, increase power
Smaller sample size affecting a statistical power:
wider distribution……. Increase β, increase power
Variability : large σ affecting statistical power
wider distribution of scores… will decrease statistical power
Variability : Small σ affecting statistical power
tighter distribution of scores… will increase statistical power
What is “statistical testing”?
- Using established statistical “tests” (i.e., equations), to determine the acceptance or rejection of a null hypothesis
What is a “
the difference between a raw score and the group (sample) mean divided by the standard deviation
What is the one-sample z-test?
Determining if a sample is representative of the population
Example of one-sample z-test:
Example: is the average salary of Toronto Raptors players different from all NBA players?