No. If you want a correct answer you must do things correctly. Download R it is free and I have attached a copy of of a manual. You can find details on how to do your analysis on YouTube. Follow that. If you want a point and click analysis download the Rcommader package from the R site and use that. Believe me you will be glad you did. Best wishes, David Booth

I agree with David Eugene Booth .

The best thing for analyses you care about is to use software where you can save the code, so that later on you can reproduce the analysis, either to remember exactly what you did, or to check some assumption or plot that you didn't think of until someone questioned you about it six months later.

That being said, an option for point and click software is Jamovi. It's free, and for a one-way anova with blocks (if that's what you want to do), it is easy to use and fairly rich with output options (nice tables, post-hoc analysis, partial-eta squared, descriptive statistics). The only thing I don't like is that it uses Levene's test for the homogeneity assumption, instead of using a plot.

There's really nothing that replaces using a full featured software program like R (or SAS or SPSS, and others, depending on your field and what your colleagues use). R really isn't that difficult to use if you have some good examples. One problem is that a lot of books and courses start with the programming aspects, where a lot of users in the beginning just need examples of analysis of simple experiments.

So, here I'll plug my own book using R, the chapter on a one-way anova with blocks: http://rcompanion.org/handbook/I_06.html .

Thanks for your suggestions, David Eugene Booth, Salvatore S. Mangiafico

I would be glad to have powerful statistic software. Installing new programs is not that easy for me due to corporate policy. So, I am tied to excel to do a spatial adjustment before I can run the ANOVA in piaf (my trial program). The experiment is set up with 4 replications, so, there are enough degrees of freedom. Instead, I could sort my relications by yield and not by block, but to me, this does not fit to the randomized block model.

If corporate policy is forcing use of Excel for data analysis then corporate will get what it richly deserves. I hope you are not trapped in the middle.

In this situation you can try R. It is free. See if IT will allow it. You could ask if there is a recommended statistical analysis software that is approved for you to download. You could see if they have a company statistician for data analysis. If the company will not provide a means to analyze data, will they let you move data over to your personal computer where you can download R?

Maybe the t-test in Excel is the only thing that corporate can manage. In that case, one alternative is to report the results from each field and do not try to combine fields.

Four replicates may be corporate policy, but it maybe no better than a random number generator. A perfect design to force the experiment to give the answer that you know is true. However, four replicates at any one point in that area might be the best choice for a company that is selling products over a wide area. The company is interested in the variability in their product over North America, while any one field trial will only tell them about the result from a few hectares.

Well, in defense of us old guys, in graduate school we used to do two-way anova on exams with a hand calculator only.

... Of course, *cough*, I can't say I got 100% on those exams...

There are a few options in the R vs. Corporate Policy battle.

1) You can install R, and I think RStudio, as portable files on, say, a USB drive. And you would then install the supplemental packages on the same drive. I've done this, and I can't remember if it did or didn't get tricky telling R to install packages here, and look for installed packages here. ... Of course, you will then need permissions to be able to run a program that isn't installed on work computers.

2) If you go to https://rdrr.io/snippets/ , you can run R programs online without installing anything. They have a good number of supplemental packages installed, so you can probably do whatever you need to do there. If you have a small data set, you can make it part of the R script you are running. One potential legal problem is that the site says that anything you submit is public.

3) Talk to the boss and make a case that if they want good results, they have to let you use appropriate software.

a) Usually with R, you will find yourself downloading a bunch of supplemental "packages" as you need them. So maybe you can download them all with IT ahead of time. Usually, you also update these packages regularly, but if you're not allowed, usually that's not fatal.

b) For the kind of things you are doing, Jamovi may be sufficient, and may be easier in these aspects because, I think, it's all self-contained. Likewise, there's JASP, which looks suspiciously similar to Jamovi. And there's a free version of Blue Sky Statistics. I've used each of these, and they all seemed pretty decent for what they are. In any case, you might be able to get these installed at work so you have them when you need them.