AP Statistics Interpretations

What does a dot represent?

The 8 dots at 10 correct represent that 8 of 105 students guessed exactly 10 correct out of 12.

SOCS

CONTEXT

shape - roughly symmetric or skewed with clusters, peaks, and gaps

outlier - yes or no, high/low

center - median (skew) mean (symmetric) and describe with ER’s w/ about or around

spread - varies not ranges, lowest to highest, going from…to

Comparing two distributions

use ER words such as higher, lower, or less and always context

Mean

If we take many many samples of sample size n, we’d expect an average of about ____.

Is A or B better?

must describe both, why one is better and one is worse

Standard deviation

The context typically varies by SD from the mean of mean.

Percentile

Percentile% of context are less than or equal to value.

z-score

Specific value with context is z-score standard deviations above/below the mean of ____.

DUFS

CONTEXT

strength - strong or weak

direction - positive or negative

form - linear or nonlinear

unusual features - outliers

Correlation coefficient

State what r is

The linear association between x-context and y-context is weak/strong and pos/neg. (No outlier mention)

y-intercept

At x-context = 0, y-context is predicted to be ____.

slope

For each additional x-context, y-context is predicted to increase by ____ y units.

coefficient of determination (r²)

r²% of the variation in y-context can be explained by the LSRL on x =____.

residual

(actual-predicted)

The difference between the actual and the predicted is ____ at x =____.

Linear model a good fit?

Answer yes or no

The scatterplot of ___ vs ___ is roughly linear.

The residual plot shows no leftover curved pattern.

Observational study or experiment

Say which it is

No treatments of CONTEXT were applied to the subjects

Reasons for stratifying

Reduces variability in the sampling distribution to make a more precise estimate of the response variable (in context) and guarantees representation in the sample proportionally to the population. Must also explain how the stratums could be affected differently by response variable.

How could bias occur

Show how the sample could be different from the population and that it could over/under estimate the population parameter.

PROPORTION

How to create random sample/assignment

Label each number a unique integer from 1 to n

Using RNG select ___ unique numbers from 1 to n and assign the ___ corresponding to the selected numbers

Variable is confounding

Show how the variable could affect the explanatory variable (if possible) and the response variable

random selection

We randomly select from the population in context so the results can be applied to the entire population.

random assignment

We randomly assign treatments to be able to show causation to those similar to those in the experiment.