Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
13 Cards in this Set
- Front
- Back
Descriptive vs Inferential Statistics
|
Descriptive - Procedures that simplify and summarizes data
- Mean, Median, Mode etc Inferential - Hypothesis Testing to determine relationships between variables and making predictions. |
|
Internal vs External Validity
|
Internal
- Estimate of the degree to which causal inferences can be made. Experiments in highly controlled laboratory settings have better internal validity. External - The extent to which the results can be applied to different people, places, and times. |
|
Continuous Variables
|
Interval - Arranged in equally spaced units and difference between points can be measured
- No Absolute zero - Temperature Ratio - There is an absolute zero - Magnitude of differences between categories is the same - Age |
|
Categorical Variables
|
Nominal
- No Natural Order - Gender: Binomial, Race: Polynomial Ordinal - Categories with natural rank or order - Income levels, grades |
|
Boxplot
|
Max & Min
75th and 25th percentile Median |
|
Quantiles
|
- Percent of sample points that are less than or equal to it
- Median is 50th percentile - Less sensitive to outliers |
|
Calculating Outliers
|
Upper outlier: 75th Percentile + 1.5 x IQR
Lower outlier: 25th percentile - 1.5 x IQR Extreme outlier is 3 x IQR |
|
Measures of Spread
|
Variability
Range, Variance, Standard deviation |
|
Normal Distribution Ranges
|
1 STD - 68.2%
2 STD - 95.4% 3 STD - 99.7% |
|
Confidence Interval
|
CI = Mean +/- (1.96 * SE)
SE = SD / Sqrt(n) |
|
Error Types
|
Type 1 - False Positive
- Reject the null when null is true - alpha is Type 1 error Type 2 - False Negative - Fail to reject Ho when alternate is true - Beta |
|
Power
|
Probability of rejecting the null given that alternative is true
Power of test = 1 - Beta |
|
Using CI to test Null
|
If null does not fall between CI - Reject null
If null falls between CI - Fail to reject null |