The experiment intentionally imposes the treatment on the group of subjects and objects with interests of observing the responses. However, it differs from observation study that involves collection and analysis of the data without altering the condition. Its construction and execution directly interfere the validity of the experiment it is important to do experimental design. However, treatment in the experiment is something which the researcher administers to experiment units. For example, a corn farm is divided into four fields, and each part is treated applied with different fertilizer to see which one produce the most corns. Another example is a teacher practicing different teaching methodology on different student groups within the class and investigates which one produces the best result. Therefore there should be factors that are used in the experiment which control independent variable.
It is the process of properly organizing an experiment to ensure that it provides the right type of data and enough availability those data, which will be available to answer the question of interest effectively. Hence particular questions that the experiment intent to answer should be identified before experimenting. Also, some of the variability that is expected in the experimental unit should be identified because the primary aim of the experimental design is to decrease the impact of the sources of variability on the answer to the question of interest. Thus, experimental design is used to improve the precision of the answers.
They are constraints developed that are used the condition where experiment should do so that effecting feedback or answer should be attained. For example, a farmer like to evaluate new fertilizer. Therefore, he uses the new fertilizer to treat his crops in one field of crops (field A), while on the other field (Field B) he uses the current fertilizer. However, the irrigation system in A is repaired therefore crops receive adequate water, while in B the irrigation system will be repaired in the next season, therefore, the field still have a problem with the supply of water for crops. Finally, the farmer concluded that the new fertilizers are matched superior.
The issue with the experiment is that the firmer had failed or neglected to control the impact of the difference in the system of irrigation. Hence it led to experimental bias, by favoring of certain results over the others. To avoid experimental bias, the farmer should have tested his new fertilizer in identical conditions between the two fertilizers (control group), all should receive the same treatment. Therefore, without controlling the outside variable and constraints, he can’t conclude the effect of the new fertilizer and not an irrigation system which led to better crop production.
It is a technique used in the experimental design where the treatment is assigned in random. Each experimental unit has the same opportunity to receive a specific treatment. However, the technique is used in the laboratory experiment where the environmental factors of the experiment are relatively easy to control. But it is rarely used on the field where factors of the environment are difficult to control. Some variation of the randomized experimental designs is randomized block design, replication, and complete randomized design.
For example, a researching doctor was investigating the effectiveness and efficiency of four different creams of the skin for the treatment of specified skin illness. He had eight subjects, and then he divides them into four treatment groups of twenty subject each. By use of randomized block design, the subjects are assessed, analyzed, and put in blocks of four depending on how severe the condition of the skin is. Out of four, the most severe case condition of the skin were categories to first block, then the second block until the twentieth block. Four members from each block were randomly sampled and one member to each of the four treatment groups. The data is collected from these samples for analysis.
Fisher, Ronald. “The Arrangement of Field Experiments.” Journal of the Ministry of Agriculture of Great Britain (1926).
Hunter, W. G., and J. S Hunter. Statistics for Experimenters: Design, Innovation, and Discovery (2nd ed.). Wiley, 2005.
Montgomery, Douglas C. Design, and Analysis of Experiments (8th ed.). Wiley, 2013.