I genuinely enjoy teaching lessons involving statistical tests in psychology. There is no sarcasm here; I think they are fun to teach!
For those of us who have migrated from AQA A the sign test is new – as is teaching statistical tests at AS (or year one). What follows is my basic proposal for teaching this topic:
Starter: What is the point of a statistical test?
Before teaching anything about statistical tests, I like to show my students a series of fictional bar charts and ask them whether or not the difference between two groups is big enough for the research hypothesis should be accepted. I vary the gap between the bars, the sample sizes, etc. After debating a few of these, pose the question: What is wrong with the way we are deciding this? What is the problem for psychology as a science?
The idea you are trying to tease out here is that it is not objective. Usually members of the class will disagree with each other on each bar chart, so that shows the inherent issue with relying on subjective human judgements…… this is why stats tests are needed…
Introducing the sign test with mind power
Students need to understand the idea that differences between two groups can be caused by chance and that a stats test is calculating the probability that chance has caused that difference.
As such, it is good to use a context that students understand the probabilities and the involvement of chance: flipping coins.
The students should be instructed to flip a coin 5 times and to count how many times it lands tails. Then, the students should be instructed to flip a coin 5 times again but this time to use ‘mind power’ to will the coin to land tails – ensure they take this seriously! Again, they should record how many times the coin lands tails.
The results can then be collated in a table like the following:
|Participant||Condition 1: Amount of tails||Condition 2 (mind power): Amount of tails||Difference (condition 2 – condition 1)||Sign|
Following this data collection, there are lots of opportunities to consolidate research methods skills such as calculating the mean, drawing bar charts, identifying the independent and dependent variables, writing hypotheses, etc.
Once the mean amount of tails for each condition have been calculated, it is likely that you can discuss the fact that there is a difference between the two conditions. So, does that mean that mind power genuinely influenced the result? The students should get the idea that despite the fact that there is a difference between the two conditions (if there isn’t, quietly fudge the results!), it is probably the result of luck/chance… however, to be objective about it, a statistical test should be used to determine whether the probability of the difference being caused by chance is low… and therefore mind power did influence the result!
The students can use the table to conduct a sign test. Full instructions and critical value tables can be found p218–219 in the Complete Companion Student Book (4th edition).