Stats Interpretations Flashcards
pass ap exam (25 cards)
Describe the distribution
SOCV
Shape: left skew, right skew, symmetric
Outliers: IQR1.5 + Q3 IQR1.5-Q1
Center: mean (non resistant to outliers) or median (resistant to outliers)
Variability: Range or Standard Deviation
Interpretation of Percentile
The % of values that are less than or equal to a given value.
Standard Deviation
The context typically varies by ____ from the mean of ____
Mean
If many many context are randomly selected, their average context will be about ___
Z: Score
context is z score standard deviations above/below the mean context
What is the association (linear graph)
DUFS
Direction: positive or negative
Unusual features: outliers
Form: linear or nonlinear
Strength: moderate, strong, or weak
Y-intercept
when x=0 the predicted y context is y intercept
Slope
With each additional X context the predicted Y context increases/decreases by slope
Residual
AP: actual-predicted
The actual value of y context is residual above/below the predicted value for x=__
Standard Deviation of residuals
The actual test scores typically vary by # and context from the predicted values based on LSRL on x=context
Coefficient of determination r^2
__% of the variation in y context can be explained by the linear relationship with x context.
Coefficient of Correlation
Use Dufs
Power
If the true proportion of context, is parameter that satisfies Ha, there is a power probability of finding convincing evidence to reject Ho in context.
Statistically Significant
Since the p-value of context is ___, the likelihood of context was large/small enough to have happened by chance alone and is/isn’t statistically significant
Conclusion of Significance Test
Because the p-value of __ is greater than/ less than alpha, we reject/fail to reject the ho. There is/ is not convincing evidence for the Ha in context.
P-value
Assuming the parameter of context the is a p probability of getting a sample of x or bigger/smaller by chance in random samples of size n
P-Value for difference of means
Assuming the true mean difference of ___ context there is a p value probability of getting a mean diff of ___ or greater than/ less than by chance alone in a sample of N
confidence level
If we constructed many, many confidence internal about from random samples of size n ___%of the intervals would capture the true parameter
Confidence Interval
We are __% confident that the interval from __ to ___ captures the true value of the population parameter context
Standard error
use when the standard deviation is not known and given sample size
One proportion z test
define
p, Ho, and Ha
Two proportion Z test
Define
both ps, Ho and Ha
One sample T test for mean
Define
mew, Ho and Ha
Two sample T test
define
both mew, Ho and Ha
