If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
Frequency Plot - Medical Test
I am working on an assignment where we are given a bunch
of data points representing medical tests taken to determine whether a person has a given disease. We have 200 test results (measuring some number from 1-50 which, at a certain cutoff value, indicates the disease is likely present): 100 for people who actually have the disease, and 100 for those who don't. Our mission is to plot the data in such a way as to evaluate the sensitivity and specificity of the test (accuracy, best cutoff point, etc.). What is the best way to plot this using Excel? G |
#2
|
|||
|
|||
Frequency Plot - Medical Test
G wrote:
I am working on an assignment where we are given a bunch of data points representing medical tests taken to determine whether a person has a given disease. We have 200 test results (measuring some number from 1-50 which, at a certain cutoff value, indicates the disease is likely present): 100 for people who actually have the disease, and 100 for those who don't. Our mission is to plot the data in such a way as to evaluate the sensitivity and specificity of the test (accuracy, best cutoff point, etc.). What is the best way to plot this using Excel? How about two separate histograms or column charts? One histogram represents those that have the disease, the other represents those that don't have the disease. The x-axis would have the test result (1-50), while the y-axis would have the number of people with that test result. The FREQUENCY and HISTOGRAM functions in Excel would be helpful. Hopefully the two histograms are almost entirely separate, and the desired cutoff point will be obvious. The overlap of the histograms should give some indication of the "sensitivity and specificity" of the test, although you'll need a statistician to put a number on that quantity. Dave "not a statistician" dvt at psu dot edu |
Thread Tools | |
Display Modes | |
|
|