4.2 blocking
- 1. Blocking Just as in sampling, when a population consisted of groups of individuals that were “similar within but different between”, a stratified random sample gives a better estimate than a simple random sample Blocking is used to separate the experimental units into categories before an experiment based on some way that is expected to effect the response to the treatments Researchers want to determine if a new detergent for clothes that require Hand-Washing works better in warm water and cold water
- 2. Blocking Researchers want to determine if a new detergent for clothes that require Hand-Washing works better in warm water and cold water. Develop an experimental design using Blocking: Randomized Block Design – the random assignment of experimental units to treatments is carried out separately within each block.
- 3. Blocking Blocking allows us to draw separate conclusions about each block, thus providing more precise overall conclusions A wise experiment will develop blocks based on the most important unavoidable sources of variability among the experimental units - randomization will balance out effects of remaining lurking variables and allow unbiased comparison of treatments ** Control what you can, block what you can’t control, and randomize to create comparable groups.
- 4. Blocking Women and men respond differently to advertising (think about car advertisements). Researchers want to design an experiment that compares the effectiveness of three different advertisements for the same product. a. Why might a randomized block design be preferable to a completely randomized design. b. Outline a randomized block design using 300 volunteers (180 men & 120 women) as subjects. Describe how you would carry out the random assignment required by your design.
- 5. Blocking Matched Pairs – special cases of a randomized block designs using blocks of size 2 - Create blocks of matching pairs of experimental units (2 that are similar) - Randomly assign one treatment to each member of the pair Sometimes a “pair” can be just one experimental unit that gets both treatments, one after the other. The randomization is the order of the treatments as that may influence the response.
- 6. Blocking Type of design determines the method of analysis: Completely randomized design – compare measures of two groups (pulse rate stand vs. sit) Matched Pairs Design – each unit is a block, compare effects of treatments within each block by examining differences (stand vs. sit)