« Home

Home » Biosciences » Statistics & Experimental Design » Ruxton & Colegrave: Experimental Design for the Life Sciences 2e » Student resources » Extra Questions » Chapter 3

Ruxton & Colegrave: Experimental Design for the Life Sciences 2e

Chapter 3

Use these questions to check what you have learned throughout the chapter.

  1. Discuss the concept of noise in experimental design with respect to between-individual variation and independent factors.

  2. If you tossed a coin ten times and got eight heads would this indicate to you that the coin was biased? If you tossed a coin one hundred times, what number of heads would make you concerned about bias in the coin?

  3. If I wanted to describe the height of life science students, why would the mean of a sample of ten randomly selected students be a better measure than the height of one randomly selected individual? Do you feel ten is a large enough sample size? Would five have been a big enough sample? Would you want a larger or smaller sample size if you were measuring number of CDs owned rather than height?

  4. Explain pseudoreplication in your own words. List five causes of pseudoreplication that may occur in your field. For each cause, how might you avoid the problem?

  5. If I’m interested in studying university students’ views on the usefulness of careers advice at school would data collected by questionnaires filled in by 2nd year Psychology students at the University of Glasgow give you cause for concern about pseudoreplication?

  6. How about using the same sample as the last question to find out about university students’ views on the health care provided by The University of Glasgow?

  7. In the elk habitat preference study in section 3.3, why does the book recommend taking daily measurements at random times? Why not just collect the data at the same time every day?

  8. If I wanted a random sample of University of Glasgow students in order to study how students spend their summer vacation, would the 3rd year Zoology students be a suitable sample group? If I wanted to study University of Glasgow students’ preferences between alcoholic drinks would the 3rd year mathematics students be a suitable study group?

  9. How would you go about obtaining a random sample of fifty University of Glasgow students?

  10. Define the power of an experiment in your own words.

  11. I want to test whether there is a sex difference in resting heart rate in domestic cats and have samples of ten male and female cats that I can attach to cardiovascular monitors. What factors influence my ability to detect a sex difference (should one exist)? For each factor, discuss the extent to which I can manipulate the power of the experiment by changes to the experimental design.

  12. Before you fly off on holiday, your aeroplane is checked for metal fatigue. Which should you worry more about, Type I or Type II error in this metal fatigue check?

  13. Discuss the legal idea ‘better ten guilty men walk free than one innocent man ends up in jail’ in terms of Type I and Type II errors.

  14. A newspaper reports that the level of heart disease in middle managers is higher than in the workers under them or the senior managers above them. The suggested reason is that middle management face stress directed at them from both above and below, whilst the other two groups just face a single source of stress. What problems can you see with this interpretation?

  15. A major scientific journal reports that students that play a musical instrument perform better in a memory test, proving that music lessons improve intelligence. Do you agree with this statement?

  16. Your grandmother still wants to convince you of the power of herbal medicine, and after reading this book she reports, “When I have a headache now I flip a coin and if it’s heads I take ginger and if it’s tails I don’t. When I take ginger my headache goes more quickly than when I don’t.” Are you now convinced that herbal medicine works?