a clear strategy for exploring data from a single quantitative variable.
1.
Plot the data: make a graph
2. Look for overall pattern (shape, center, spread and unusual values)
3. Calculate a numerical summary to describe center and spread.
4. The new step:
Properties of a discrete random variable X:
1. 0 P(X) 1
2.
Let X = the number of shots I make in 3 attempts.
X
0
1
2
P(X)
0.1
0.25
0.5
3
Calculate the probability that
I make all 3 shots:
I make less than half of my shots
I make at least one shot
P(X = 3)
Entering data in a list: Stat: Edit
cfw_4, 9, 11, 13, 13, 15, 15, 16, 16, 16, 17, 17, 18, 19, 20, 20
Making histograms: Stat Plot: Plot 1 graph #3
-zoom: 9 zoomstat makes a nice window
-trace will show class boundaries and frequencies
-window: Xscl chang
1. How to confirm that the transformation fit the data to a straight
line:
a. Exponential models:
Residual Plots: after the LSR equation (L1, L3) has
been pasted into Y1, L4 = Y1(L1) and L5 = L3 L4.
Look at the scatterplot for L1 vs. L5(RESID).
Remember
Graphs
The horizontal axis should include the variable name and the
possible categories. The bars should have some space between
them to indicate they are freestanding and can be arranged in any
order.
The vertical axis can be frequency or relative freque
The z-score for a weight of 8 lb is: = .31.
This z-score is between 0 and 1 so we know at least 50% weigh less than 8
lb. We also know that at most 84% weigh less than 8 lb. Thus, between
50%-84% of babies weigh less than 8 lb.
*recall the information alr
Stemplots
The numbers to the left of the line are the stems (hundreds and tens
digits) and the numbers to the right of the line are the leafs (units
digits).
You must include a key (with units) and a label/title
Leaves should be single digits (no commas)
When should we stratify?
If you think there are groups within the population who may be
similar with regard to the question of interest, you should take an
appropriately sized simple random sample from each group.
In our example, we should anticipate that
LSR equations:
a. Exponential models:
Written log y = slope (x) + intercept
After an inverse transformation (see #5 part f or example
4.7 in the book when dealing with ln transformation) you
should end up with a y = abx
b. Power models:
Written log y =