U nit 5 Practice Test
M ULTIPLE CH OICE. Choose the one alternative that best completes the statement or answers the question.
1) W e ha ve calculate d a 95% confidence interval an d w ould prefer for our next confidence interval to
ha ve a sm aller m arg

Chapter 13: Experiments and Observational Studies
Observational studies
observational studies - researchers don't assign choices, they observe them
retrospective study - subjects are selected and then their previous choices/
behaviors are studied
can

Class Notes
Probability
"And" means multiply
"Or" means add
"Not" means 1 minus %
Chapter 14: From Randomness to Probability
Random phenomenon - a phenomenon is random if we know what outcomes could
happen, but not which particular values will happen

Class Notes
Used when events are not disjoint => center of the venn diagram
false negative - you're okay but you're sick
ex. girl seems okay but she's crazy
false positive - wrong but you're bad
ex. turned girl down, but she would have been a good da

Chapter 16: Random Variables
random variable - a random variable assumes any of several different numeric values
as a result of some random event. random variables are denoted by a capital letter
such a X
discrete random variable - a random variable th

Chapter 18: Sampling Distribution Models
sampling distribution model - different random samples give different values for a
statistic. The sampling distribution model shows the behavior of the statistic over
all the possible samples for the same size n.

Chapter 17: Probability Models
Bernoulli trials, if.
there are two possible outcomes
the probability of success is constant
the trials are independent
Geometric probability model - is appropriate for a random variable that counts the
number of Bernou

Ema an yd-has 129+
T390. 1. = FAJ Posi'liw.
L Nu m ha M i+
Ins sum.
*Go k3 is hit; Po
*Imud'mguno 34 dwu- oL=\w.r-5pimar
*Ggahumghn bum Lu.- mm,.h.ku,.
'lyziu-m-
15?e 2. L033 OK= =.02. -72-L Fair-n3 *3 l m
L.)M as urban EMMA _ a . :
MW'EA'h '4 . Luau *

Chapter 19: Confidence Intervals for Proportions
Standard Error - when we estimate the standard deviation of a sampling distribution
using statistics found from the data, the estimate is called a standard error.
Confidence Interval - a level C confidenc

Means Testing
T-test => hypothesis test
confidence interval
no p-hat; now y-bar
no p; now u
no more z but t
no normalcdf, but tcdf
dF - degrees of freedom (n - 1)
check conditions and assumptions
must be random, independent, 10%
nearly normal n

Chapter 20: Testing Hypotheses About Proportions
Null Hypothesis - the claim being assessed in a hypothesis test. Usually, the null
hypothesis is a statement of "no change from the traditional value," "no effect," "no
difference," or "no relationship." F

Categorical Data
Chi Square tests
three types
goodness of fit => distribution is as expected (ex. dice, hockey players births)
homogeneity test => 1 variable (ex. eye color) with two groups (ex. men,
women)
independence test => 2 variables with more

Chapter 4: Displaying and Summarizing Quantitative Data
Histograms
displays the distribution at a glance
data is sliced up into bins (at least 5) => provide the building blocks for the
histogram
there are no spaces in between the bars; if there are, t

Chapter 7: Scatterplots, Associations, and Correlation
Scatterplots
allow you to see patterns, trends, relationships, and even the extraordinary
value sitting apart from the others
shows relationship between two quantitative variables
ask about associ

Chapter 5: Understanding and Comparing Distributions
Boxplots and 5-number summaries
Draw a vertical axis spanning the extent of the data. Then draw boxes based on
the lower and upper quartiles and the median (then connect them to form a box)
place the

Chapter 6: The Standard Deviation as the Ruler and the Normal Model
The standard deviation as a ruler
trick in comparing different looking values is to use standard deviations
measure the amount of standard deviations something is from the mean
Standa

Class Notes (9/10)
Types of data
Categorical (qualitative) - a quality (ex. shoe size, phone numbers)
quantitative - a number (ex. assigning numbers to men and women)
marginal distribution - ex. on page 24 and 25
Chapter 3: Displaying and Describing C

Class Notes
Re-expressing data
skewed right - usually take the log of the data
skewed left - square the y-axis
try log first, then square root, the inverses and stuff
only do this to the y-axis
once you re-express the data, re-check the residuals be

Chapter 9: Regression Wisdom
Getting the "bends": When the residuals aren't straight
always good to check the scatterplot of the residuals for bends that may have
been overlooked in the original scatterplot
Sifting residuals for groups
whenever we sus

Class notes (8/26)
Survey leads to an inference => you can't ask everyone
problems with surveys:
error - these can't be helped; normal => gun target: high, low, left, right
bias - very bad; try to eliminate => bent gun sight
response bias - feel comp

Chapter 11: Understanding Randomness
The best ways we know to generate data that give a fair and accurate picture of the
world rely on randomness, and the ways in which we draw conclusions from those
data depend on the randomness
Simulation
Building a

Class Notes
you cannot go backwards on the linear model equations (ex. you can solve for fat
with a given amount of protein, but you can't solve for protein with a given fat
amount)
r^2 = x% => x% of y-value can be explained by variations of the x-value