What is your software? Don't remove outliers they're very important in this type of work. Separate your question into parts. It appears like you have several questions that are run together. Best, D. Booth

Dear Booth,

Thank you for your response. Yes, I did the analysis with outliers. I just wanted to confirm whether it is correct or not. Thank you.

I used Stata statistical software.

As you said, I have several problems those combined. I will explain those clearly. My problem is data presenting. I feel like it is not presented clearly, so readers will not understand what I am saying. Sorry for long reading.

Suppose my 1st treatment is Sucrose concentration with 2 levels (0.5 M and 0.75 M). First, I did the Wilcoxon rank sum test for overall data to see which level shows the best survival and recovery. Because I want to select the best treatment combination from each treatment levels. With respect to Sucrose concentration, p value is 0.0245. So, there is a significant difference between 2 levels. When I did the ordinal logistic regression, it shows higher survival rate in 0.75 M than in 0.5 M sucrose concentration. I used box plots to present this data. At that point, I used data separately, as with Liquid Nitrogen application and not application (these treatments normally do to the same sample step by step, not separately, get as treatment combination). When I did the comparison of median within box plots according to Two-sample Wilcoxon rank-sum (Mann-Whitney) test at P ≤ 0.05, there is no significant difference. There are outliers in the box plots. Other cases also it is not clear in box plot what I have done in the analysis. That is the biggest problem. So how can I present this data? Is it wrong that analyzing overall data for each treatment or do I need to separate as Liquid Nitrogen application and not application and get the p values? Normally ordinal logistic regression, we interpret data with values of odd ratio. Do I need to explain that much or simply which have highest and with have lower?

Sir, thank you very much for your help and time.

Best, Iroshini

Hi again. First, I think that you are not approaching this in the best manner. I assume that you are using survival methods because that's your goal i.e. you want to know which treatment maximized survival. I think that there's a better way to do it. I just publi

Following