A.1 Answers to Teaching Week 1 tutorial
Answer to Sect. 1.2
The process is:
- Ask the question.
- Design the study.
- Collect the data.
- Describe and summarize the data.
- Analyse the data.
- Report the results.
Answers to Sect. 1.3
- Either is OK; they are just different. Note the different way a study can be addressed.
But importantly: neither RQ is actually helping answer the issue of the cafe owner!
For example, I may be able to tell the difference between diet and regular cola... but I may like them both equally (i.e., the test rating is the same).
- The cafe owner would not allocate drinkers to drink either diet or regular softdrinks, but just find those who were drinking either and ask them to rate the taste. I guess retrospective would be best.
- The cafe owner would need to allocate drinkers to either diet or regular. Ideally, the drinkers would not know which they had consumed (really, that's next week's work). Ideally, the drinks would be allocated at random to receive diet or regular (next week too). Ideally a true experiment.
- Experimental would be better;
fewer things out of the researchers control.
Besides, people buy what they like...
Answers to Sect. 1.4
You can use the Glossary in the textbook!
Answers to Sect. 1.5
Answers implied by H5P.
- P: People at an American college.
- O: The average time taken to walk 50m.
- C: Between the way subjects were using their phone.
- I: None: an observational study.
- Explanatory variable: The way that the phone is being used.
- Response variable: The time taken to walk 50 m.
- Unit of observation: Individual subjects.
- Unit of analysis: Individual subjects.
Answers to Sect. 1.6
Many populations are possible. The "best" population: the easiest, or cheapest to get? All of the options are possibilities.
"Relaxing": Many options possible for quantifying this (heart rate? breathing rate? survey? change in heart rate before and after drinking the tea?).
Again, many options are possible, Some options for comparing to Earl Grey tea: coffee; hot water; black tea; nothing at all; anything else but Early Grey; etc.
"Between before and after drinking Earl Grey tea" is not a comparison because everyone in the population (sample) is treated the same way: There are not two groups in the population being compared.
An example RQ might be:
"Among USC students (P), the average heart rate 30 minutes after drinking (O) is different between those who drink a cup of Earl Grey tea and those who drinking a cup of hot coffee (C) when the drink is provided to them (I)".
I'm not saying this is the best; it is just an example.
Get volunteers; place into one of the groups (Earl Grey and coffee); give the beverage; measure heart rate after 30 mins. Note that many design issues are not covered until the next lecture, but you can introduce, and encourage, discussion about some of the ideas if you wish (e.g. blinding).
An example RQ might be:
"Among USC students (P), the average heart rate 30 minutes after drinking (O) is different between those who drink a cup of Earl Grey tea and those who drinking a cup of hot coffee (C)". I'm not saying this is the best; it is just an example.
Based on the above RQ: Find USC students who drink Earl Grey tea, and others who drink coffee, and measure their heart rate 30 minutes after they drink.
It would have a lot of confounding effects to account for.
For my RQ above: The volume of tea or coffee; brand of tea or coffee; how made; when beverage taken; how long steeped for; how heart rate is measured; student (full-time only?); etc. You don't need to define these, but if you have time you can have a go at one or two definitions.
Many options, but for the RQ above: Type of beverage; heart rate after 30 mins (or better, the change in heart rate).
Answers to Sect. 1.7
- The unit of analysis is the floorboard: because one board is compared to another; because the hardness is a feature of each board, not each test; because the two measurements on each board are not independent of each other (from the same board).
- There are 5 units of analysis.
- There are 10 units of observation.
- Now, only one unit of analysis.
- Within boards, the variation is small, apart from Board 1.
Between boards it is larger.
Answers to Sect. 1.8
Answers implied by H5P.
- Quasi-experimental. The classes are determined by the students, but the lecturer decides which class gets which treatment.
Answers to Sect. 1.9
Answers implied by the crossword:
Answers to Sect. 1.10
- Should be in terms of averages or means. So perhaps "Among watertanks used in southeast Queensland, are the average lead levels in tank water in concrete tanks higher than in poly tanks?"
- Everyone dies! So a time-frame is probably meant (i.e., twelve months after amputation). And compared to what? Perhaps something like: "Are lower-limb amputees more likely to die 12 months after amputation compared to upper-limb amputees?" or something.
- No defined population (frozen beans? fresh butter beans? tinned kidney beans? jelly beans? coffee beans? can sizes?); "amount" of salt should be, for example, the "average" amount (or concentration) of salt, and per 100 g or similar. "Is the mean amount of salt in tinned butter beans the same for Woolworth's Select brand as for Edgells brand?" Anyway: You can look at the label (unless, of course, you wish to query those values).
- Silly: elephants are huge, and joeys are tiny. No-one needs to test this. Also,the RQ talks about "weight" rather than "mean weight" too.
- Needs to be more specific! Perhaps: "Is the mean reaction time different for males and females?" or similar.