Statistical Relationships
In the previous chapters, you were introduced to methods that a researcher can use to examine differences and some effect sizes to quantify the magnitude of those differences. For example, we could use those methods to answer research questions such as:
- Are teens getting a different amount of sleep that 9 hours?
- Is the proportion of all Minnesota children under 6 years of age who are above the MDH reference level in 2018 different than the proportion in 2012?
- Are the effects of marijuana on IQ different for users who started using marijuana as a teen compared to users who start using marijuana as adults?
- Is the proportion of Bachelor contestants that are still together with the suitor they chose different from the proportion of Bachelor contestants that are still together with the suitor they chose?
Another type of question that is common in the social and educational sciences is about the relationship between two attributes. These types of questions lay the groundwork for looking at predictors of our outcomes of interest. For example,
- What is the relationship between socioeconomic status and happiness?
- Is education level a predictor of income?
- How well do SAT scores predict how well first year students’ GPAs in college?
In the following chapters, you will learn about some of the visualizations, numeric summaries, and inferential methods that are used to examine “relationship” type research questions.