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;
16 Cards in this Set
- Front
- Back
Correlation |
quantifies/describes the degree of a relationship between two variables. i.e Ice cream sales and temprature have a strong correlation |
|
Linear regression |
is the graph or model of a linear relationship between two variables. |
|
Pearson Correlation Coefficient |
(r)-basically the slope, tells how the rate of change occurs with respect to to variables
calculated by means and standard deviation r=(SUM of all)((Z score for X)×(Z score for Y)) |
|
Spearman’s rank correlation |
strength of a relationship between two ranks of variables. |
|
What makes the correlation between X and Y |
X causes Y Y causes X its a coincidence final: C causes x and y relations. |
|
experimental study |
variables/induviduals have a treatment and we observe response we can fix values of variables we want to control for example you give a football player a oxygen tank during breaks and compare it to fatigue of non -oxygenated |
|
observational study |
Natural observation, no external influence imposed |
|
Y = intercept + slopeX + error: this equation represents what? |
the linear regression equation |
|
the error in the linear regression equation assums |
Error is the random influence on Y Normal distrubution Equal variance for all data points Independent
|
|
what are Residuals and how do we use them? equations ?what do they mean? |
Residuals are basically, how far a point is from the Line of best fit. It basically can depict to us how relaible our line of best fit is. SUM of all(Y-Y^=u^)^2=(Y − intercept − slope · X)^2
The higher the residual the more error and less reliable our line of best fit. |
|
What is the concept of least squares? |
When we construct our line of best fit We should fine the mean and slope that minimizes the residuals; which reduces the variability and increases certainty
=(Y − intercept − slope · X)^2 |
|
What does a slope of zero mean? |
It means that there is no visible relationship between two variables. |
|
P value assumes what? and what does changes in the P value mean |
P-value assumes there is no existing relationship between variables. The logic indicates that as P-value increases there it is more likely there is no relationship. As P-value decreases it indicates alternatively that the chances of the there being no relationship decrease meaning the relationship is strong and chances of that being caused by chance are unlikely. |
|
What is Type I error and how is it different then type II? |
Type i IS when u reject the null but it was actually true, and Type II is when u accept the null but it was false |
|
How large do samples need to be? |
Generally, Large samples are not always better then small ones. The main thing is, is how the sample is produced. |
|
what is power |
power is the probability of jecting the null, when it is false. |