Hello, thanks for the video. I was just wondering if replication would increase accuracy, precision or both. In control charts it's possible to distinguish systematic error from random error. I was trying to understand if replication helps to solve for systematic error in design of experiments.
There are many possible designs for 2 independent variables. The question of what is best to use depends on your purpose and what is feasible. As a starting point, you may want to look up a Factorial Design.
The analysis is very similar for unbalanced data, but becomes quite a bit different for incomplete data. ua-cam.com/video/pDmALCp7Mb4/v-deo.html&ab_channel=JaradNiemi
The CRD refers to how you assign levels of treatments to experimental units. The CRD allows you to make causal claims about the treatment. When I use the term sampling, I am referring to how you obtain your experimental units from some larger population. This is typically very specific to the field of study. For example, in agriculture, you typically just have the fields you have and therefore no sampling is possible.
THANK YOU!! I now understand it more
Hello, thanks for the video. I was just wondering if replication would increase accuracy, precision or both. In control charts it's possible to distinguish systematic error from random error. I was trying to understand if replication helps to solve for systematic error in design of experiments.
Thank you!! But how we get the state that we have 85% of the variability in our analysis?
The R-squared value at 7:47 is 0.85 which says that this model describes 85% of the variability in the data.
Can CRD be used for experiments with a control group with 2 independent variables? If no, what is best to use?
There are many possible designs for 2 independent variables. The question of what is best to use depends on your purpose and what is feasible. As a starting point, you may want to look up a Factorial Design.
Thank you very much for your video. Do you know how I can analyze a CRD with unequal repeats in R?
The analysis is very similar for unbalanced data, but becomes quite a bit different for incomplete data. ua-cam.com/video/pDmALCp7Mb4/v-deo.html&ab_channel=JaradNiemi
What is the sampling used? is CRD the research design?
The CRD refers to how you assign levels of treatments to experimental units. The CRD allows you to make causal claims about the treatment. When I use the term sampling, I am referring to how you obtain your experimental units from some larger population. This is typically very specific to the field of study. For example, in agriculture, you typically just have the fields you have and therefore no sampling is possible.
Hi sir, may I ask how to know the number of treatments?
In this video, I show the two treatments at 03:14. With that being said, the actual experiment that these data are based on had 5 treatments IIRC.
😂