(10 pts)
State vs District proportions
states in my sample ALASKA, OHIO, MICHIGAN, TEXAS, DISTRICT OF COLUMBIA
REGRESSION LINE
%40 proportion
p-value is > 0.5 (%95 CI) DO NOT TRUST
- The states in my sample are Rhode Island, Maine, Colorado, Ohio, and Tennessee.
- The equation for my regression line is:
- Proportion of seats for 40% Democratic votes?
- Yes I can trust the prediction because my p-value < 0.05
- There are 3 states in my sample size: Delaware, Illinois, and Wyoming.
-Regression line: y=2.5829x−0.6193
-p-value: 0.1653
-40% proportion: y=2.5829(0.4)-0.6193 y=0.41386
My P-value is greater than 0.05, therefore we cannot trust the prediction at a 95% confidence level
-6 states in my sample: Kentucky, Montana, Ohio, Illinois, Oregon, and Rhode island.
-Regression line equation- y=3.3648x-1.1796
-40% proportion- y= 3.3648x(.4)-1.1796 = 0.16632
-p value-0.0095
-0.0095 is less than 0.05 so we can trust the prediction
There are 6 states in my sample: Nebraska, Virginia, California, Hawaii, Vermont, and Massachusetts.
The equation of the regression line is
y= 3.2404x - 1.1155
The proportion of seats for a statewide Dem. vote of 40% is
y= 3.2404(.40) - 1.1155
y= .18066
The p- value 0.0005 which is less than 0.05, which means that we can trust the prediction
My States: Idaho, Nebraska, Indiana, Missouri, North Carolina, Washington, New York, Massachusetts
N= 7
Regression line: y = 2.8285 x − 0.9037
y=2.8285 (0.4) −0.9037
y=0.2284
p-value = 0
TRUST
I have 3 states in my sample. They are Georgia, Wisconsin, and Montana.
My regression line is y = 7.8938x - 3.4421
My proportion of 40 percent is
y = 7.8938*0.4 - 3.4421
or y = -0.2846
My p-value = 0.0091
Therefore I trust my prediction because p < 0.05
- States: New York, Virginia, Wisconsin, Iowa, Idaho, West Virginia, Utah, and Alaska.
N=8 - y=2.6998x - 0.9234
-
y=2.6998(0.4) - 0.9234
y=0.15652 - p-value= 0.0035. Since this p-value is relatively small, I think it would be safe to trust this prediction.
The states in my sample were:
Louisiana, Missouri, Nevada, and Rhode Island
My sample size is 4.
My equation of my regression line for my proportion is:
y = 2.0730x−0.3921
The proportion of seats my regression line predicts for a statewide Democratic vote proportion of 40% is:
y = 2.0730 x 0.40 - 0.3921
y = 0.4371
My pValue is 0.0443. I would trust my pValue because it is less than 0.05.
- I have five states in my sample. This includes: Louisiana, Arizona, Virginia, California, and Massachusetts.
-
y = 1.5158 x - 0.1259
-
y = 1.5158 (.40) - 0.1259 = 0.48042
-
p-value = 0.0067The p-value is less than .05, so I can trust the prediction.
- I had 3 states in my sample:, West Virginia, Mississippi and Delaware.
- The regression line equation is: y = 4.0330x - 1.3753
- For a 40% Dem vote proportion the formula predicts y = (4.0330*.40) - 1.3753 = 0.2379
- With a p-value of 0.1294 there is less than a 95% confidence interval that this prediction will be true.
In my sample, there are three states which include South Carolina, Colorado, and Washington.
An equation of the regression line for my proportions comes out to be
y = 4.5966x - 1.8377
The proportion of seats my regression predicts for a statewide democratic vote proportion of 40% is 0.00214.
My p-value of 0.0661 (which is greater than 0.05) indicates that I shouldn’t trust my prediction.
-
How many states were in your sample and and which ones were they?
5, New Hampshire, Connecticut, Wisconsin, Mississippi, West Virginia -
What is an equation of the regression line for your proportions?
y=4.3288x−1.5043 -
What proportion of seats does your regression predict for a statewide Democratic vote proportion of 40%?
0.0257 -
Does the p-value indicate that you should trust your prediction?
Yes, 0.0257<0.05
In my sample there were seven states: Wyoming, Nebraska, Tennessee, South Carolina, Colorado, Nevada, and Vermont.
An equation for my regression line is y=2.7322x−0.8145
My data predicts a .282 proportion
I don’t trust the prediction as the p-value is > .05