3 Interpreting the results of research

A major challenge in any research is to make the appropriate conclusions and generalizations from a particular study. Even professionals sometimes describe study results in ways that go beyond the methods used.

When researchers use a properly executed experimental method, then cause and effect conclusions may be made. If the research is nonexperimental or correlational, then we cannot conclude anything about cause and effect without further investigation.

A major issue when scientists are sharing their findings with the public is being careful about how those results are described. The problem relevant to this section is describing nonexperimental research using cause-and-effect language implying the use of a true experiment.

Vita (n.d.) listed what she concluded were terms that might either imply causation or correlation. Try the quiz below and see how your judgments match Vita’s. As you think about these terms, you’ll also see that some of these words might be appropriate in the right context, whereas others, primarily on the causal side, are never going to be OK to describe a nonexperimental study.

An example of the interpretational challenges in news articles is a report in US News, “Pressuring Kids to Diet Can Backfire, Damaging Long-Term Health” (Hirsch, 2019). The main issue in this article is the headline, which might imply causality to a casual reader.  In the article itself, Hirsh does a good job of emphasizing that the work by Berge and colleagues (2019) is a correlational study.

The study couldn’t prove a direct cause-and-effect, but parental pressure to get and stay slim was associated with poorer health in young adulthood, the study found (Hirsch, 2019, para. 11).

The design was longitudinal,  as adolescents were then followed up in adulthood. Conducting longitudinal comparisons is one way that correlational comparisons can rule out alternative explanations, because there is the ability to measure whether one behavior comes before another over time.