Experimental Designs II - Statistics Project Example

Blocking however has a number of disadvantages to a research initiative. It is not appropriate for large number of participants as large sizes reduces the design’s efficiency. Another disadvantage of blocking is in its rigidity with respect to homogeneity as it is not appropriate for populations or samples with extremely diverse features. “Interactions between block and treatment effects” also contribute to experimental error, discrediting reliability of the design (Hawai, n.d., p. 1). A factor refers to an independent variable that is manipulated in the research process and can be subjected to different treatments to generate a variety of factor levels.

Using factors benefits an experiment by providing a basis for comparison of different treatments as each factor or factor level is subjected to a different treatment (Stat, n.d.). Mains effect refers to the variation of means across factor levels within a given factor. It considers a factor level mean and compares it with means of other factor levels to determine existence of significant difference. Interaction effect however considers variation of a factor level mean across levels of another factor in a ‘multi-factored’ design to determine existence of significant combined effects among factors in a design (Stockburger, n.d.).

Covariate refers to a control setup in an experiment and reduces noise by predicting outcomes of the dependent variables. This helps in identifying outliers for appropriate measures (Druckman, Greeen, Kuklinski and Lupia, 2011). Tradeoffs define compromises that must be made in an experiment. Experimental tradeoffs are for example realized in validity. Ensuring both internal and external validity is a tradeoff that must be made with respect to sample size and sampling technique because strict focus on achieving one type of validity undermines the other.

Practicality is another experimental trade off that considers research methods that call for adjustments to ensure