Home
Products and Services
How to Do Rate Calc...
Pesticide Recordkeeping...
How to Layout an Expt...
Placing an Order
Contact Info
Useful Links
FAQ
Privacy Policy
























Home
Products and Services
How to Do Rate Calc...
Pesticide Recordkeeping...
How to Layout an Expt...
Placing an Order
Contact Info
Useful Links
FAQ
Privacy Policy


































Home
Products and Services
How to Do Rate Calc...
Pesticide Recordkeeping...
How to Layout an Expt...
Placing an Order
Contact Info
Useful Links
FAQ
Privacy Policy


Agrosoft Systems
Bringing Technology to Agriculture the Easy Way

How To Layout Field Experiments Without Sweating it Out
... All About AgroPlotter for Windows© ...

If you are a researcher like me you already know that two of the most time-consuming but critical phases in setting up an experiment are randomization of treatments and laying out the field plots. With the advent of computers and the development of statistical software, such a task is expected to become routine... that is if you are computer savvy. For ordinary researcher, learning to use a statistical software is not that hard to do especially if you licensed one of those statistical software with point-and-click interface such as SAS®, SPSS®, Statgraphics®, and several others. However, those user-friendly routines apply to data processing, when you already have the data on hand, and admittedly, not very helpful in laying out field plots on which you would grow your experimental materials. It does not matter what software you use to select the statistical design to address your experimental objectives as long as the design you have chosen to use is appropriate. However, if your chosen design does not match the field plot layout or your field plot layout does not match your experimental design, your experiment is invalid and your conclusions would be wrong.

Not many would admit to it but some of us are still using any one of the following methods to do treatment assignment and randomization:

  1. shuffling a stack of cards
  2. use of book page numbers
  3. use of table of random numbers
  4. use of random number generator (computer)
  5. "lottery" drawing method
  6. others (dice, anyone?)

Most of the above are time-tested, reliable methods and are hard to abandon especially if you have been using them for years. Two things common about them are: (1) they are time-consuming and (2) are prone to researcher bias (whether you like to admit it or not!). One case of bias is doing an extra set of randomization to ensure that one replication is not a duplicate of previously assembled replication. I still don't know the reason, perhaps a large part of it is psychological in nature, but many researchers still feel awkward when confronted with mirror image replications (does this sound redundant yet?). Statistically, duplicate(?) replications are valid and appropriate as long as the treatments within each replications were randomly drawn and independently assembled of the other replication(s).

However you do it, randomization is very easy when dealing with the simple designs (one-factor designs) such as Completely Randomized Design (CRD), Randomized Complete Block Design (RCBD), and Latin Square Design (LS). It is completely different matter when you move up to factorial designs dealing with two or more factors in one experiment. While AgroPlotter for Windows would still be a considerable help for the simple designs, it is on the factorial designs that you would appreciate the power of AgroPlotter for Windows the most. The complexity of randomization in factorial experiments comes from the requirement that a separate randomization be performed for the various levels of factor "A" from that of factor "B" (or "C" or "D"). Obviously, depending upon the chosen design and as more factors are added in the study, such complexity grows at which time the sequence and hierarchy, on when and how randomization is performed for the levels of each factor, become very critical.

Enter AgroPlotter for Windows(c)...
AgroPlotter for Windows is a tool to help you randomize treatments and make a plot layout of your field experiment. This is NOT a statistical package that can do statistical analysis..

The following features would ensure that your use of the software would be a pleasant experience:
  1. A section to enter details to completely document your experiment including research title, start/end of study, experimental design used, researcher in charge, a memo area for entering other important information and activities (i.e., spraying, crop variety, seeding/harvest dates).
  2. AgroPlotter allows you to select from one the following statistical designs:
    • Single Factor Designs
      • Completely Randomized Design (CRD)
      • Randomized Complete Block Design (RCBD)
      • Latin Square (LS)
    • Two-factor Designs
      1. Split Plot
      2. Strip Plot
    • Three-factor Designs
      1. Split-split Plot
      2. Strip-split Plot

  3. AgroPlotter allows you to save, view, print, and delete randomization session results including the experiment details. This feature is handy for perennially conducted experiments so you can keep track of the research information and compare them from year to year. Example: Some crops (i.e., brocolli), depending upon the succeeding crop in the rotation, could have an adverse effect on seed establishment of the next crop. Repeated planting of some crops (i.e., tomatoes) on the same area could lead to a severe outbreak of some pests/diseases (i.e., bacterial wilt, nematodes). What does AgroPlotter has to do with those examples I cited? Well, if you had filled out AgroPlotter's experiment detail form, you could have documented the crop, location, dates, perhaps fertilizer and herbicide use of each of your experiments. These are all the information you need to map out a location's cropping history. Thus you can capitalize on this wealth of information to exclude potentially confounding factor(s) that may influence your research results if you are to use the same location. Alternatively, one may choose to explore the opportunity to utilize those information for subsequent research such as when doing allelophatic studies in which the identity and number of croppings of previous crop species would be critical variables.
  4. There are times that you do not want to enter detailed information of an experiment since only the start and end dates have changed from that of the previous year or season. No problem! AgroPlotter has a re-plot mode that would re-use all previously entered data except that it would also allow you to re-generate entirely different field plot layouts using the same or different number of replications, treatments, and/or plots, subplots, and sub-subplots, as the case maybe, and however you like it. You may save the result of a re-plot session using the same filename if you want to.

If you are wondering how this innovative and really helpful program could perform both randomization and assignment of treatment without you breaking a sweat (or wasting time!), allow us to give you a usage demonstration.

Click here for samples of saved and retrieved AgroPlotter sessions using the following statistical designs:

  • Completely Randomized Design
  • Randomized Complete Block Design
  • Latin Square
  • Strip Plot
  • Strip-Split Plot

Here are the new/enhanced features of the newly released version:

  • Entirely new and better graphical user interface
  • .
  • Experiment ID, title, and treatment names are all customizable which can be viewed, edited , sorted, and printed after being saved.
  • Show me. (IE user: Be sure to enlarge graphics by clicking on the zoom in button located at the right hand bottom corner of the screen).
  • Entirely new randomization of previously generated plot can be performed with just a push of a button under the same experimental design and without the need to exit the current plotting session.
  • No need to lift your fingers from the mouse when panning over your generated plots. The mouse can grab the plot and move it however you like to view any part of the layout.
  • All designs except for the Lattice and Latin Square designs can easily handle more than 50 treatments. RCBD can breeze through more than 100 treatments subject to computer memory limitations with a blink of an eye.Show me. (IE user: Be sure to enlarge graphics by clicking on the zoom in button located at the right hand bottom corner of the screen).
  • All plotting sessions can be saved, deleted, and printed from a secured database.
  • AgroPlotter now has a companion Notepad style memo area which can accomodate a lot of information that you care to note down about any particular plotting session. Entries on this notepad can then be copied to a word processing program for filing, editing, and printing as necessary.
  • Other screen capture:Capture 1|Capture 2 (IE user: Be sure to enlarge graphics by clicking on the zoom in button located at the right hand bottom corner of the screen).

System Requirements:

  • Operating system: Windows XP®, Windows Vista® or Windows 7®; runs on both 32- and 64-bit systems
  • 1 GB of RAM; more is better
  • At least 2 MB of free hard disk space
  • Optimized to run under 800x600 pixel resolution; runs better with higher resolution

Placing Your Order:


Google
 
Copyright 2013, Agrosoft Systems©

Trademarks mentioned in any of the web pages are either registered, owned, or copyrighted by their respective companies/owners/copyright holders.