AP Statistics help

Vin{MBL}

5185 Posts

April 19 2008 19:15 GMT

I'm doing a past paper - I just corrected it and I have a question.

One of the questions asked to compute a margin of error for a least squares regression of 7 points. (on the graph).I used a t distribution with df = 6 (because there were 7 points). However, the answer key used df = 5 and specifically stated that using df = 6 is partially correct. I don't understand why it is correct to use df = 5.

The computer output included this:

Source DF
Regression 1
Residual 5
Total 6

Thanks in advance.

tiffany

3664 Posts

April 19 2008 20:43 GMT

it's been four years since i did any statistics, but sometimes you have df = n - 2. what is the context of the problem?

Vin{MBL}

5185 Posts

April 19 2008 21:30 GMT

John believes that as he increases his walking speed, his pulse rate will increase. He wants to model this relationship. John records his pulse rate, in bears per min, while walking at each of seven different speeds, in miles per hour (mph). A scatterplot and regression output are shown below.

shows a graph with correlation = .985
There's a table with computer output. of std. deviation etc;

(c) John wants to provide a 98% confidence interval for the slope parameter in his final report. Compute the margin of error that John should use. Assume that conditions for inference are satisfied.

Specialist

United States803 Posts

April 19 2008 22:42 GMT

I think for least square regression stuff, the degrees of freedom is n - 2. Because I remember talking about "why is this n-2 thingy here" with my girlfriend.

I THINK its n - 2. I have not been paying much attention in my AP Stat class for the past month >.<

Vin{MBL}

5185 Posts

April 20 2008 00:01 GMT

i just checked it you're right it is just n - 2 for regression. thanks

Please or register to reply.

AP Statistics help

Completed

Ongoing

Upcoming