In a field experiment if there is possibility for both designs factorial RBD and Split plot design, and both qualifies as per degrees of freedom, then what design should be choose and why?
The main difference between split plot and factorial RBD is the way of laying the respective treatments/factors to be studied. In split plot, the experimental area is first divided into almost homogenous blocks to minimize variation within blocks. It is followed by random allocation of different levels of one treatment factor (in main plot factor) in each of blocks and then each main plot is further divided into the number of levels of another treatment/combination of treatment factors (sub plot factor) with random allocation of each level of sub plot factor/s. Whereas in factorial RBD, the first step of making blocks is same but the combinations of different levels of respective treatment factors are allocated randomly in each block. The design of layout is selected based on practical applicability of nature of treatment factors to be studied.
Split plot design is generally used when there are 2 factors one of which requires lagre plot size while other shorter sized plot so called main plot treatment (like irrigation, tillage and other operation which need to have large space to perform that treatment) and subplot treatments (require lesser sized area)..while in case of factorial RBD, this is also of 2 factors design but both of the factors may not require large or small sized, may be similar sized plots...like if we want to study interactive effects of 2 factors so called different nitrogen doses and different phosphorus doses, one should choose factorial RBD instead of split plots though both design have 2 factors to study...in factorial RBC we call 2 factors with different levels.. example: in one factor if we want to select nitrogen doses so this will be one factor with different doses as levels...in case of split plot design, only sub plot treatments and their interection with main plot is analyzed with more degree of precision while in factorial RBD interaction among different factors with levels are measured more precisely
The main difference between Randomized Block Design (RBD) and Split Plot Design is that, in the case of RBD, our purpose is to study the effect of one factor, which has different levels of equation precision for all levels. Suppose you are going to study the effect of different levels of nitrogen alone on the grain yield of one crop. In this situation, only the rate of nitrogen will vary, and the rest of the practices will remain the same for all the treatments. All treatments will be replicated an equal number of times for each treatment. In such a situation, the RBD design is recommended.
But what happens in the case of a split-plot? You want to study the effect of two factors, one with more precision and one with less precision.
The factor which is very important to you will be kept as a subplot, and the factor which is less important to you will be kept as the main plot.
For example, you want to know the effect of irrigation (irrigated and rainfall) and different levels of zinc such as 20 kg, 40 kg, 60 kg, and 80 kg/ha) on wheat yield. However, in this situation, your focus is more on zinc than irrigation, and you want to study it precisely so you will keep zinc as a subplot that will be replicated more times, so the result will be more accurate. Likewise, you will keep irrigation and rainfall as the main plot, and its effects will be studied with less precision.
Your treatments will be like this:
A: Irrigation (Main plot)
a-Irrigated
b-Rainfall
B. (Zinc) Subplot
20 kg/ha.
b. 40 kg /ha
c. 60 kg/ha
d. 80 kg/ha
I hope this has clarified things for you.
regards
nesar
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments