-
What is statistics?
Value that describes a sample
-
What is population?
Set of all inividuals of interest in a particular study
-
What is sample?
Set of individuals selected from a population, usually inteded to represent the population in a research study
-
What is parameter?
A value that describes a population
-
Independent variable vs. dependent variable
- Independent variable: the variable being manipulated by the researcher
- Dependent variable: the one that is observed to assesss the effect of the treatment
-
Symbolic notations
-
- Add all of the scores in the group.
- Compute inside parentheses first, then square it.
- Square each score, then add up squared scores.
-
-
-
-
-
-
-
-
-
Measures of Central Tendency: Mode
The score or category has the greatest frequency
-
Measures of Central Tendency: Median
The score that divides the distibution in half
-
Measures of Central Tendency: Mean
Arithmetic average
-
Which Measure of Central Tendency is most influenced by outliers?
- Mean
- Incorporate every score
-
Variance Equation
variance = mean squared deviation =
- sum of squared deviations
- number of scores
-
-
Population Standard Deviation
-
Sample Standard Deviation
-
-
-
What causes variance to go up?
-
What causes variances to go down?
-
How to determine probability of z-score
Base probability percentage with z-score of normal distribution
- Example:
- Given:
-
- What percent scored higher than you on a test?
- P(z>+1.00)=.1587 (15.87%)
-
How to find a z-score with that probability
Example: For a normal distribution, what z-score separated the top 10% from the rest of the distribution?
-
Standard error
- Standard distance between M and
- How well a sample mean estimates its population mean
- Smaller standard error, better estimation
-
Standard Error vs. Standard Deviation
-
-
Independent t-test
- A research deisng that uses a separate group of participants for each treatment condition (or for each population)
- AKA between-subjects deisgn
- Estimated Error:
- Pooled Variance:
-
-
#38 on Test: On average, what value is expected for the F-ratio if your null hypothesis is true?
-
Correlation
A statistical technique that is used to measure and describe a relationship between two variables
NO manipulation of variables, just looking if 2 variable have a relationship to each other
-
Positive correlation
2 variables that you are comparing move in the same direction
-
Negative correlation
2 variables that you are comparing move in opposite directions
-
Parametric test
- Inferences about population from statistics that you calculated
- Test hypotheses about population parameters
- Require a numerical score for each individual
- Require data measured on an intercal or ratio scale
-
Nonparametric Tests
- Do not make statements about relationships to population parameters
- Data can be measured on nominal or ordinal scales
- Require fewer assumptions
- Are less sensitive than parametic tests
|
|