## Anterior chamber depth by different factors II

- Details
- Written by A. de la Llama

This post is a follow up of the last one. I will show now how to make the same graphics but from another point of view. This time will be only for the anterior chamber depth.

This graphic is like the first one from the last post but now I will put the age group in the *x* axis so we can see how the ACD changes with the age. The data comes from here.

This other graphic will show the distributions of values for the ACD for every group as histograms, the vertical grey dashed line is at 3 mm. Remember the data is simulated from a normal distribution with the number of observations, mean and sd reported in the first and second table for the ACD. Compare it with the boxplots from the last post, although you can get here more detail, it is harder to compare within groups. Every graphic has it's uses.

## Axial length and anterior chamber depth by different factors

- Details
- Written by A. de la Llama

In the study *Axial Length, Anterior Chamber Depth-A Study in Different Age Groups and Refractive Errors* the data is shown as tables and there are no graphics. I couldn´t resist it so I made a graphic for the three tables given. The first table shows the number of observations per group, the second table shows the mean anterior chamber depth (ACD) in mm with standard deviation (SD) per group and the third table is the same as the second table but for the axial length (AL).

In the first graphic I will show you the mean ACD with the 95 % confident interval (c.i.) for the true population mean. This c.i. is calculated from the standard error (SE) so it will be larger for larger SD or smaller sample sizes and I calculate the c.i. for a *t* distribution. The vertical dashed grey line is at 3 mm.

The second one is the same for the axial lengh:

You would agree with me that these two graphics let you see much faster some characteristics of the data than the tables. You can also show more about the distribution of the data with histograms or boxplots. I will show you this with the following graph, since I don´t have the original data I will simulate it from a normal distribution with the number of observations, mean and sd reported in the first and second table for the ACD. Here is the result:

In the boxplots I made the width (in this case the height) of the box is proportional to the square root of the number of observations, the minimum is 4 and the maximum is 11 and they are both females between 41 and 60 years old so you can see them in the panel bottom right for astigmatism and myopia respectively.

Here is the interpretation of the boxplot, from wikipedia:

And here is the code to reproduce the graphics.

## Ophthalmologists and optometrists in the U.S.A.

- Details
- Written by A. de la Llama

From the medicare website I downloaded some data (well, over 2 million records) from the Official Physician Compare Data. After filtering and just keeping the ophthalmologists and optometrists as primary specialty (now down to over 80.000 records) I wanted to see how they are distributed in the U.S. Fortunatly the dataset comes with the zipcode so I plotted how many of them are per zipcode. You can click on any of the three images for a larger view.

As expected they follow a population pattern. You can find the code to reproduce the maps here.

Page 6 of 10