Formulating a Research Question
Good Research Questions
Formulating a useful research question can be somewhat more intricate than it might seem. Review the following examples for information on what inhibits and enhances the merit of a research question.
Sample Research Questions
Does being molested as a child cause a person to become a sex offender as an adult?
Problem: The question is defined in terms of a causal relationship.
Solution: In order to find any kind of causal relationship, a researcher must use an experimental design, which requires him or her to manipulate the independent variable. In this case, child molestation would be the independent variable and, thus, cannot be manipulated. So finding a relationship through a correlational study is the next best strategy. Therefore, a better question would be, "Is there a relationship between child molestation and sex offending in adulthood?" This gets tricky too, as this would mean that the researcher would have to include individuals who were molested as children and sexually offended as adults and individuals who were molested as children but did not sexually offend as adults.
Is a person's loyalty and dedication to a company correlated with his or her happiness?
Problem: The question is not stated in measurable terms.
Solution: A research question must state how the terms in it will be measured. One person's definition of company loyalty and dedication might be different from another's. The same goes for happiness. Loyalty, dedication, and happiness must be defined and stated in measurable terms. A better question would be, "What is the relationship between the number of years that a person has worked at a company and his or her scores on a measure of life satisfaction?"
Is truancy correlated with failing grades?—or Do offenders who are substance users have higher rates of recidivism?
Problem: The question is obvious. Also questions that lead to a simple "yes" or "no" do not provide a high level of insight.
Solution: This relationship between truancy and failing grades is obvious because very often truant students will fail due to attendance policies alone. Even if this is not the case, it stands to reason that truant children will not be exposed to the material as much as children who attend class, which will, therefore, be reflected in the grades for each group. This relationship between recidivism and substance use is also obvious because the use of illegal substances is a crime and is, therefore, inherently recidivism. It would be better to explore programs that are geared toward reducing truancy or programs that are likely to reduce substance users recidivating.
What factors are related to healthy emotional development in children?
Problem: The question is vague and too broad for a single study.
Solution: Such a study would be a massive undertaking. Instead, choose one aspect of emotional development in children and focus on that, such as whether authoritative, authoritarian, or permissive parenting styles are correlated with likeability in children as measured by the number of peers in a classroom who report liking them.
Is there a correlation between having an alcoholic parent and alcoholism in adulthood?
Problem: The question has already been researched.
Solution: This topic has been well researched. Instead, consider studying aspects of resiliency, such as factors that occur which appear to mitigate the likelihood that a child of an alcoholic parent will be an alcoholic as an adult.
Does the amount of sleep a person gets have an inverse relationship with earning high grades?
Problem: The question backs the researcher into a corner.
Solution: Research questions that attempt to predict a very specific relationship among the data leave the researcher with few options if the results are nonsignificant. It is much better to study simply whether the amount of sleep a person gets is correlated with his or her grades. That way, whether higher or lower grades are correlated with sleep, the researcher might still have some significant findings. Therefore, a better question would be, "What is the relationship between the amount of sleep a person gets and the grades?"
Is bullying prevention training in schools correlated with lower rates of utilization of mental health services?
Problem: The question is based on a premise that might not be correct.
Solution: This research question is predicated on the notion that children who are bullied actually utilize mental health services, which may not be the case. Children who are bullied can feel ashamed, and children who feel ashamed are not likely to seek opportunities to discuss what aroused such feelings. Instead, a better question would be, "What is the relationship between bullying prevention training in elementary schools and rates of anxiety and depression among students as measured by the Beck Anxiety Inventory for children and the Beck Depression Inventory for children?"
