What should you consider when designing and conducting a statistical study?
See solution.
Practice makes perfect
When we want to design and conduct a statistical study, we have 3 major areas of a study we should be extremely careful with.
Let's describe each area one at a time.
Sampling Techniques
Recall that in most cases we do not analyze the whole population because it is impractical. Instead we choose a representative sample. Therefore, we need to consider which of the sampling techniques we should use.
In other words, we need to think of a way of choosing members of our sample so that each possible subset of the population is represented.
Survey Questions
When undertaking a statistical study we should always consider if our survey questions are correct. Let's take a look at the example survey questions. We will divide them into two groups — questions that are biased and questions that are not biased.
Biased Question
Unbiased Question
Do you agree that we should have more very useful math classes?
What do you think about having more math classes?
How many hours do you waste on addictive social media each day?
How many hours do you spend on social media each day?
Here, the biased questions contain verbs and adjectives that can suggest a response. This means that the results of a study will depend on the type of question we ask.
Randomness
We want our sample to be random because it allows us to make true conclusions about the population. However, sometimes it is very difficult to achieve full randomness. Therefore, we should consider how to make our study results as random as possible.
