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;
26 Cards in this Set
- Front
- Back
|
Units |
To understand a number one must know its units 8 miles/weeks/year? |
|
|
Distribution of categorical variable |
Both bar or pie chart can be used to graph the distribution of a categorical variable |
|
|
Distribution of a quantitative variable |
a histogram can be used to show the distribution of a quantitative variable |
|
|
Prevalence |
The number or share of a population that has a particular disease or condition |
|
|
Incidence |
Rate at which new cases of a disease appear in the population |
|
|
Mean (average) |
Total up all the values, then divide by n` |
|
|
Median |
Point that splits the distribution into two equal halves |
|
|
Relative risks and odds ratios |
Show relationships between categorical variables with only two possible values |
|
|
Relative risk |
Group 1's probability outcome/Group 2/s probability outcome |
|
|
Odds ratio |
Odds of outcome for Group 1/Odds of outcome for Group 2 |
|
|
Correlation |
Measures strength and direction of relationship between two quantitative variables Represented by r (correlation coefficient) Ranges from -1 to 1 Correlation = 0 means no relationship |
|
|
When correlation is positive |
Variables tend to move in same direction When one variable goes up, the other goes up |
|
|
Negative correlation |
Variables tend to move in opposite directions |
|
|
What does a given r really mean? |
If a variable (x) changes by 1 standard deviation, the correlation r is the expected standard deviation change in the other variable (y) |
|
|
Simple regression |
Best-fit straight line How the independent variable predicts the dependent variable Y=Bo+B1X1--> Constant+Regression Coefficient: slope |
|
|
Multiple Regression |
Uses multiple independent variables to predict the dependent variable Y= Bo+B1X1+B2X2 |
|
|
Practical significance |
magnitude matters- how big is the effect |
|
|
Practical significance definition |
Extent that the magnitude (if true) would be important or relevant in the real world Depends on judgement Must be able to interpret results in units and ways relevant to practice and policy |
|
|
Running a significance test |
-Calculate a test statistic -Find the p-value |
|
|
Calculate a test statistic |
Estimate a difference or relationship using sample data Compute how far the estimate is from the null Divide by the sampling variability (stand. error) Test statistic=(Estimate-Null)/SE |
|
|
Find the p-value |
Probability of the estimate, when null is true Use software or a table to find the p-value If p-value is small enough, null hypothesis can be rejected |
|
|
P Value |
Probability of observing our sample estimate (or an estimate further from the null), if null is true P values are universal! |
|
|
Low P Value |
More statistical significance--> null very unlikely, alternative possible |
|
|
How low is low enough? |
Common convention is .05 or less |
|
|
How many independent variables can one regression have? |
Depends on: Amount of date (# of observations) -- 1 indep var for every observation Multi-collinearity, how much variation in independent variable Precision desired |
|
|
Interaction variable |
Product of two other variables Describes how one variable moderates the effect of the other |
