As with measurement, one of the most important things about study design is that it is appropriate to the research question. Descriptive and correlational designs are very flexible, but lack internal validity, requiring that researchers use these designs for specific types of questions only. In the following, we discuss the appropriate application of these designs in different circumstances.
Answer the following questions in paragraph form. Your answers may be brief, but should include a sentence of justification for your answer. When you have prepared answers, compose them into a discussion post and post them. When you are finished, take a look at your classmates posts do you agree with their answers? Why or why not?
A researcher interested in public education investigates the performance gap often seen between inner-city schools and suburban schools. He measures multiple indicators of teacher performance, school atmosphere, student intelligence, and community atmosphere. He also collects demographic information on the students and their families. He finds that there are no differences between the schools, but that there is a significant difference in the socio-economic status (SES) of the students and their families. He makes the claim that having a low SES causes poor performance in school. Is this a valid claim? Why or why not?
A researcher interested in the idea that people who engage in conscious relaxation (like meditation or relaxation exercises) experience fewer age-related aches and pains (i.e. not chronic pain) than people who do not consciously relax. Can you apply a correlational design to test this idea? How so?
A researcher finds that participants who live in urban areas generally vote democratic and those who live in rural areas tend to vote republican. He claims that living in crowded conditions causes people to favor democratic attitudes and vice versa for republican attitudes. Is this a valid conclusion? Why or why not?
A group of researchers runs a study in which they look at the relationship between music preferences and reading preferences. What statistical test should the researchers use on their data? Explain your answer.