7.36 Beer and blood alcohol content: Many people believe...

7.36 Beer and blood alcohol content: Many people believe that gender, weight, drinking habits, and many other factors are much more important in predicting blood alcohol content (BAC) than simply considering the number of drinks a person consumed. Here we examine data from sixteen student volunteers at Ohio State University who each drank a randomly assigned number of cans of beer. These students were evenly divided between men and women, and they differed in weight and drinking habits. Thirty minutes later, a police officer measured their blood alcohol content (BAC) in grams of alcohol per deciliter of blood (Malkevitc and Lesser, 2008). The scatterplot and regression table summarize the findings. BAC (grams per deciliter) 0.15- 0.10- 0.05- 2 4 6 8 Cans of beer Estimate Std. Error t value P(>|t|) (Intercept) -0.0127 0.0126 -1.00 0.3320 beers 0.0180 0.0024 7.48 0.0000

Transcript

00:01 All right.

00:02 So in this problem, we're given a lot of information about a linear regression model.

00:09 So you can see in the scatter plot, i'm just going to sort of sketch it out real quick.

00:13 We're looking at how the number of cans of beer increases the blood alcohol level for people, and then it looks something like that, that one point way out there.

00:24 And then you have a computer output, which is really important as well.

00:30 And i don't see any questions, but we'll go through some kind of common questions that you might see.

00:35 The first thing is, what type of relationship is there? all right, well, there is a pretty strong relationship.

00:48 It does look like a line, and as one increases, the other increases, so it is positive.

00:55 And you want to make sure whenever you're detailing that relationship that you include context, that tells the reader more about what you're talking about.

01:01 So there's a strong linear positive relationship between cans of beer and the blood alcohol content level for people consuming that.

01:16 So that's really the most you probably could draw out of the scatter plot.

01:23 You can see there's a positive relationship.

01:25 It's pretty strong.

01:26 The points are going to be pretty close to a line if we were to draw those through there.

01:31 And it is a line.

01:31 It's not really a curve.

01:33 If you include that last point up here, you might could say it was curve, but we're going to focus on linear because that's what the computer output gives us below.

01:41 So then looking at the computer output, it tells us that our y intercept is negative 0 .127, which means that when zero cans of beer are consumed, there is a predicted blood alcohol content of negative 0 .127, which kind of makes no sense, because we know it would actually be zero.

02:17 All right.

02:18 So that's kind of the y intercept in this case is a little bit, you know, not meaningful.

02:26 And then we have our slope, which is the number right below that.

02:29 Those first two numbers tell you your y intercept in your slope always, which is 0180.

02:34 And that means that as the number of cans of beer increases by one, the blood alcohol content is expected to, or predicted to, increase by about 0 .0180.

03:00 All right, which makes sense...

Question

Please give Ace some feedback