Loading phys211..

Even in a perfect world, measured quantities normally varies and one goal of physics is to detect real variation on top of normal "noise".

There are precise quantitative ways to do this but it requires more background in statistical methods. In this course, we will mostly use visual tools to make decisions about data.

For lab 1, we looked at small vs big push of the IOLab. We used the boxplot feature in StatKey to visualize data. The boxplot gave us a very nice visual of multiple statistics such as the mean, median, and the range. The boxplot on Statkey even allowed us to quickly determine outliers.

That way, you were able to reach conclusions regarding whether the variability of the average acceleration while sliding between a big vs small push was just normal noise or whether there are significant and persistent differences.

Note that a perfectly valid conclusion to make from data is that more data is needed or better measurement techniques/better design for the experiment. It is usually one or the other, not both. A good scientific statement from lab 1 could have been: "Unlike the max acceleration which was clearly different between big and small push, the average acceleration during sliding is much less different between big and small. The noise appears too big to clearly differentiate the cases and a better experimental setup would be needed to definitely measure a difference if there are any."

We will continue using the Boxplot tool in this course.

Type of Variables

Independent variable: This is the one changed by the scientist.
Dependent variable: This is the variable that the scientist observes to see if or how it is affected by changes made in the independent variable.
Controlled variable. This is a variable that the scientist will try to keep constant while changing the independent variable.

In your lab this week you will study how kinetic friction depends on the mass of the object.

Type	Variable
Independent	mass of IOLab+ stuff added to i
Dependent	kinetic friction
Control	surface

There are often multiple control variables. It really refers to anything that you are NOT changing.

Two quantitative variables.

We will often look at cases where we have two quantitative (numerical) variables. We will use the scatterplot to visualize.

A scatterplot is a graph with the independent variable (or function of) as the horizontal axis and the dependent variable (or function of) as the vertical axis. The paired data is represented as a dot on the graph.

When we look at a scatterplot, we are interested to see whether there is a pattern, a trend.

For example, the image below shows a scatterplot created in Statkey of the ph vs average mercury for 53 different lakes in Florida.

While the data is pretty scattered, a careful observation of this scatterplot reveals there there seems to be a trend in that high ph lakes seems to be low average mercury while the reverse is true of low ph.

On the right of the image, you see the mean and standard deviation of each variable as well as the sample size. There is also a new statistics we have not seen before called the correlation.

The correlation is a measure of the strength and direction of linear association between two variables.

The correlation often denoted r is a number between -1 and 1. Values close to +1 or -1 means strong association (positive or negative) while values close to 0 means no association.

Please use a modern browser to view our website correctly. Update my browser now