實驗設計：練習題

In an agricultural experiment investigating the impact of temperature on seed germination, a scientist uses a completely randomised design. However, they later realise that the moisture content of the soil varied significantly across the greenhouse benches used.

Which of the following would have been the primary benefit of using blocking based on soil moisture in this context?

Consider an experiment where a researcher wants to test the durability of three types of road surfacing materials. They have access to four different regions with varying traffic volumes. They decide to use a randomised block design where each region is a block.

If the researcher fails to use randomisation when assigning the materials to the sections within each block, what is the most significant risk to the validity of the statistical inference?

In an agricultural study designed to compare the yields of four different varieties of wheat, a researcher decides to use a randomised block design. The field is divided into 5 blocks based on a gradient of soil fertility, and each variety is grown once within each block.

What is the primary purpose of blocking in this experimental design compared to a completely randomised design?

In a psychological study evaluating the effect of background music on cognitive task performance, a researcher decides to use paired comparisons. They select 20 participants and have each participant complete a task twice: once in silence and once with classical music. The order of the conditions is randomised for each participant.

What is the main statistical advantage of this matched pairs approach over a completely randomised design with 40 different participants?

A researcher is conducting a study to compare the effectiveness of four different fertilisers on the yield of wheat. To control for variations in soil fertility, the field is divided into three blocks based on soil nutrient levels. Within each block, the four fertilisers are randomly assigned to four equal-sized plots.

Identify the specific experimental design being used in this study.

A botanist is testing the effects of three different light intensities on plant growth. To ensure that initial plant height does not influence the results, the botanist groups plants of similar heights together before assigning them to treatment groups.

Identify the statistical technique being used to reduce experimental error and explain how it differs from a completely randomised design.

A researcher designs an experiment to test the effectiveness of three different cooling systems in computers. To account for variations in the ambient temperature of different laboratory rooms, the researcher uses a randomised block design. However, due to a logistical error, the cooling systems were not randomly assigned to the computers within each laboratory room but were instead assigned in the same order each time.

Explain how this failure to implement randomisation within the blocks might compromise the ability of the researcher to distinguish between experimental error and the actual effect of the cooling systems.