what kind of model is best to use?

[#1]

To model RSV in day care centres, NetLogo was used to create the agent based model needed to simulate interactions in a day care if children were to show up with RSV. We built upon the the basicEPIdem model in the NetLogo base to develop models that allowed us to model a day care setting where a vaccine for RSV is not present and one where there are children who become vaccinated. Both models use a classroom size with 24 students and three teachers. They both allow for the students and teachers to get sick, but when the students get sick they ”stay at home” for a period of time. In this vaccinated model, when a student becomes ill and stays at home the model will recognize this and commands the student to remain stationary for the time being. After the student is full recovered, the model will introduce the healed student back into the day care simulation. The teachers continue to come to work even if they have RSV since symptoms in adults are usually not serious and it mostly likely will go unnoticed if they have the disease and they will continue to show up to work. The vaccinated model allows for one child to become vaccinated a week for the month. The goal is to compare whether vaccinating children makes a difference when it comes to the behavior of RSV.

[#2]

I'm sorry, but I'm confused by your query.  Are you asking if there is a relationship between vaccinated and not vaccinated in the data table?  It would be helpful to attach the JMP data table instead of jpg's in a Word doc.  You should have all of the data in one table and a column for Vaccinated (yes/ no).  Whether the model you are using in the simulation software has any relationship with reality can't be assessed with this data. The model used to create your results is already in the simulation software.  Do you have access to it?  To be clear, "The goal is to compare whether vaccinating children makes a difference when it comes to the behavior of RSV." should be modified to include based on the simulation software at the end of your goal statement. You would need to validate the model in the simulation software with real data to determine if vaccination has any real affect.

[#3]

Thanks a lot for your reply.

enclosed is the data . would you please help me understand how I can approach a meaningful analysis ? 

[#4 · Accepted solution]

I have tried to combine your files in the attached file.  I left out the columns that had no variable information in them.  The Vac column is 0= no vaccination and 1= vaccinated.  From this data set you can compare the vaccinated vs. the non-vaccinated.  I'm guessing the Total infected is the Y? There are also some scripts, but I really doin't know the subject or how the data was obtained.

[#5]

Thans. One more question I am trying to see how  vac affects the length of time RSV moves through a population. So i decided to do two different linear models and compare them. Here is what I got do you think it's meaningful? 

[#6]

Sorry, but I do not see the "length of time RSV moves through a population" as a value in the JMP table?  Your plots show as the number of ticks increase, so does the number of total infections, regardless of vaccination.  It is statistically significant, but it does not explain all of the variation in the data (R-sq ~43-49%).