1.4 Exercises

Exercise 1.4.1

Each of the following is an example of a real study or scenario. State which types of bias are present, and explain why the issue might lead to a wrong conclusion.

  • Voluntary response bias
  • Nonresponse bias
  • Undercoverage bias
  • False response bias
  1. In 1936, Literary Digest magazine conducted the most extensive (to that date) public opinion poll in history. They mailed out questionnaires to over 10 million people whose names and addresses they had obtained from telephone books and vehicle registration lists. More than 2.4 million people responded, with 57% indicating that they would vote for Republican Alf Landon in the upcoming Presidential election.

This conclusion turned out completely wrong: incumbent Democrat Franklin Roosevelt won the actual election, carrying 63% of the popular vote.

  1. In the 1870’s, a study was performed by Francis Galton to address the question of whether prayer works. The researcher posited that members of the clergy spend more time praying than others, so if prayer works, they should tend to live longer. He obtained, from public records, the “mean age attained by males of various classes who had survived their 30th year, from 1785 to 1843.” Galton concludes:

When we examine this category, the value of life among the clergy, lawyers, and medical men is as 66.42, 66.51, and 67.07 respectively, the clergy being the shortest lived of the three. Hence the prayers of the clergy for protection against the perils and dangers of the night, for protection during the day, and for recovery from sickness, appear to be futile in result.

  1. “Car Talk” is a podcast and radio show hosted by NPR. The hosts of the show regularly ask listeners to call in and give their opinions on various topics related to driving, such as, “Should we build more bike lanes on the roads?”

Exercise 1.4.2

For each of the following scenarios, identify the sampling method.

  1. A survey is sent to everyone who purchased Amazon Echo last year, asking them to rank their experience on a scale of 1-10.

  2. Canadian Prime Minister Justin Trudeau sends a poll to 100 random people from each Canadian province, asking if they support his new tax reform.

  3. Garfield, the owner of Fantasy Costco, orders his employee Davenport to stand outside the door and ask each entering customer what they plan to buy.

  4. Admiral Adama wants to know about morale in his fleet. He randomly selects 3 of his 20 ships, and then asks 100 random people aboard how they are doing.

  5. My friend asks me how much money I spend on board games. I randomly choose 5 board games from my shelf of 78 games, and look up their prices. I find an average cost of $25, so I estimate that I have spent $25*78 = $1950 total on board games.

  6. The California Democratic Party wants to know who Californians support for governor in 2022. They randomly choose 50 registered Democrats, 50 registered Republicans, and 50 Independents, and they ask each person’s preferred candidate.

Exercise 1.4.3

The following descriptions of experiments are excerpts from Wikipedia articles for three famous experiments: Louis Pasteur’s work on vaccines; Roger Wolcott Sperry’s studies of the human and animal brain; and the Kansas City Police Department’s study of impacts of patrols.

For each of the three experiments, identify the subjects or experimental units, identify the treatments, and describe any blinding or blocking that may have been involved.

  1. Louis Pasteur’s first work on vaccine development was on chicken cholera. He received the bacteria samples (later called Pasteurella multocida after him) from Henry Toussaint. He started the study in 1877, and by the next year, was able to maintain a stable culture using broths. After another year of continuous culturing, he found that the bacteria were less pathogenic. Some of his culture samples could no longer induce the disease in healthy chickens. In 1879, Pasteur, planning for holiday, instructed his assistant, Charles Chamberland to inoculate the chickens with fresh bacteria culture. Chamberland forgot and went on holiday himself. On his return, he injected the month-old cultures to healthy chickens. The chickens showed some symptoms of infection, but instead of the infections being fatal, as they usually were, the chickens recovered completely. Chamberland assumed an error had been made, and wanted to discard the apparently faulty culture, but Pasteur stopped him. Pasteur injected the freshly recovered chickens with fresh bacteria that normally would kill other chickens; the chickens no longer showed any sign of infection. It was clear to him that the weakened bacteria had caused the chickens to become immune to the disease.

  2. Roger Wolcott Sperry first became interested in “split-brain” research when he was working on the topic of interocular transfer, which occurs when “one learns with one eye how to solve a problem then, with that eye covered and using the other eye, one already knows how to solve the problem”. Sperry asked the question: “how can the learning with one eye appear with the use of the other?” Sperry cut nerves in the eyes of cats so the left eye was connected to the left hemisphere and the right eye was connected to the right hemisphere; he also cut the corpus callosum. The cats were then taught to distinguish a triangle from a square with the right eye covered. Then the cats were presented the same problem with the left eye covered; the cats had no idea what they had just learned with the right eye and because of this could be taught to distinguish a square from a triangle. Depending on which eye was covered, the cats would either distinguish a square from a triangle or a triangle from a square, demonstrating that the left and right hemispheres learned and remembered two different events. This led Sperry to believe that the left and right hemispheres function separately when not connected by the corpus callosum.

  3. Read the full (very short) article here: https://en.wikipedia.org/wiki/Kansas_City_preventive_patrol_experiment