(a) Single blind, because customers must (ethically) be told their rate-structure, but the meter-reader was blind to the rate structure. It was not matched-pair or a block design.
(b) Single-blind, because drivers knew the condition but the evaluator was blind to it. The same drivers were tested under all conditions, so it was a block design.
(c) Double-blind, because neither the drivers of the cars nor the researchers measuring the tread knew which brand was which. It was a block design because all four tires were tested on each car.
Problem 3.
(a) The explanatory variable was the rate structure, and the response variable was the amount of electricity used during peak hours.
(b) The explanatory variable was the condition of being sober, having drunk alcohol, or having smoked marijuana. The response variable was driving accuracy.
(c) The explanatory variable was the brand of tire; the response was remaining tread.
Problem 5.
(a) The color of the cloth is a confounding variable. It could be that birds always prefer red to other colors.
(b) Put the food on different colored cloth. Randomly assign five birds to eat from each of the four colors. Then see which color each bird preferred without food.
Problem 6.
A case-control study. First if there were a suspicion that the phones caused cancer it would not be ethical to randomly assign some people to use them. Second, brain cancer is rare so it would take huge number of participants in a experiment to result in a few cases of brain cancer.
Problem 7.
Smoking is an interacting variable, because it changes the relationship between the explanatory variable and the response variable. Stroke rates for women who use oral contraceptives also depend on whether they smoke.
Problem 12.
(a) Plan 1 is a retrospective study; plan 2 is a prospective study.
(b) Plan 1 has the disadvantage that participants may not remember how much they spent. Plan 2 has the disadvantage that the participants might change their normal spending behavior, knowing that they were part of a study.
Problem 13.
(a) Explanatory
(b) Response
(c) Interacting
(d) Confounding
Problem 15.
(a) It would be unethical to randomly assign people to either drink alcohol or not.
(b) The headline implies a daily drink causes better health, but this was an observational study, so a causal conclusion is not justified (see d).
(c) prospective
(d) "Confounding variables and implications of causation" is a major problem because the results are intended to show a causal relationship. But perhaps the reverse relationship is true; those that are healthier are more likely to allow themselves to drink alcohol, or perhaps those who are more relaxed about life are both healthier (less stressed) and more likely to drink; other confounding factors are also possible (diet?). "Extending the results inappropriately" would be a problem if the results were applied to any group other than that represented, which was mostly white, middle-class, married residents of the US. "Using the past as a source of data" is not a problem.
Problem 18.
(a) "Confounding variables" should not be a problem because of randomization. "Interaction variables" may be an issue, for instance, perhaps people who already had sleep problems were more likely to benefit from exercise. "Placebo, Hawthorne, and experimenter effects" may be a problem because both groups knew whether of not they were exercising, and results were self-reported, so beneficial effects may have been exaggerated by those who exercised. "Ecological validity and generalizability" may be a problem because participants were included in a community based exercise program, thus encouraging them more that it people were self-directed in their exercise routine.
(b) This was a randomized experiment, so confounding factors should have been similar in the two groups, Therefore a causal conclusion is justified.
Problem 22.
(a) "Ecological validity and generalizability" is likely to be a problem because of the setting - participants are sleeping in a lab and not at home. Experimenter effects are also possible, because participants know whether or not they ate.
(b)The Hawthorne effect is the biggest problem. Those who received the plants were treated specially. A control group should have been given something like a small figurine or something in place of a plant.