Principles of Experimental Design

A market research company wishes to find out whether the population of students at a university prefers Starbucks or Dunkin Donuts. A random sample of students is selected, and each student is asked first to try both but the order they try them is randomly decided with a coin toss (heads: Starbucks, then Dunkin. tails: vice versa). They then indicate which brand they prefer. This is an example of

• completely randomized experiment

• an observational study

• blocked experiment

• matched pairs experiment

In order to perform an experiment using 60 members of a gym, I first divide the list of members into men and women because I feel that the results will be different based on gender. I then randomly choose 30 men and 30 women. I assign half of the men to the treatment group and half to the control group. I repeat this procedure with the women. This is an example of a:

• completely randomized experiment

• an observational study

• blocked experiment

• matched pairs experiment

Researchers want to know how diet influences a child's brain development. In a sample of 100 5 year old children, 20 of the children are given a vegan diet, 20 are given a pescatarian diet, 20 are given a standard American diet, and 20 are given a paleo diet. This is an example of...

• completely randomized experiment

• an observational study

• blocked experiment

• matched pairs experiment

Dentists want to know if a fluoride mouth wash will help prevent cavities. After 6 months, they will compare the number of cavities in those who used the mouth wash versus those who did not. Which is an example of a matched pairs design for this experiment?

• 
  Dentists sort 40 patients into groups based on age, such as 10-14, 15-20, 21-26, and 26-30. Half of the patients in each group are given the mouth wash, half are not.

• Dentists sort 50 patients into pairs that are the same age. 1 person is given the mouth wash and 1 is not.

Participants are assigned to groups, based on gender. Then, within each group, participants are randomly assigned to treatments. For this design, 250 men get the placebo, 250 men get the vaccine, 250 women get the placebo, and 250 women get the vaccine. What type of experiment design is this?

• Completely Randomized Experiment

• Blocked Experiment

• Matched Pairs Experiment

• This is not an experiment

Can pleasant aromas help a student learn better? Two researchers believed that the presence of a floral scent could improve a person's learning ability in certain situations. They had 22 people work through a pencil-and-paper maze 6 times. Three of the times they wore a floral-scented mask and three times they wore an unscented mask. The three trials for each mask followed one another. Testers measures the length of time it took subjects to complete each of the six trials. They reported that, on average, subjects wearing the floral-scented mask completed the maze more quickly that those wearing the unscented mask, although, the difference was not statistically significant. This study is:

• A convenience sample

• An experiment, but not double-blind

• A double-blind experiment

• An observational study

Identify the design for the experiment.

• Single blind, blocked by pesticide

• Completely randomized, the treatment is pesticide

• Randomized block, blocked on tree type, the treatment is pesticide

• Double blind, the factors are tree type and pesticide

They are attempts to establish a cause-and-effect relationship between two variables by manipulating one variable.

• Observational Studies

• Sample Surveys

• Experiments

• All of the above

Which of the following is an example of a matched pairs design?

• Randomly assigning subjects to two different treatment groups.

• Dividing subjects into blocks based on gender and then randomly assigning treatments.

• 
  Pairing subjects based on age and then randomly assigning treatments within each pair.

• Assigning treatments based on the order of arrival.

Which of the following is a key advantage of using a randomized block design?

• Allows for the elimination of all bias

• Controls for known sources of variability among subjects

• Increases the likelihood of finding a significant result

• Requires fewer sources than other designs