Are the Results What You’re Hoping For?
In the Results section of a research article, you want to read to see how the research participants improved and if that improvement was significant.
Here it’s important to see if the results were significant in the areas where you want to see improvements in your child. For example, if you were hoping to see improvements in working memory and multiple-step math procedures, but the study fails to show this, then this may not be the best approach for your child.
It is also critical to determine if the “treatment” group’s improvement was significantly better than the “control” group’s improvement. If not, (or if there is no control group), the seller of the product service cannot really claim “significant” improvements.
Is the Discussion Section Expected or Unexpected?
The discussion section of a research article outlines the author’s thoughts about the importance of the results of the study. Although this section is reviewed diligently in most peer-reviewed journals, it is easy to make the stats say what the author wants them to say. If you’re still not sure if the treatment method is right for yourself, your child, or your teen by the time you get to the discussion section, here are some things to consider:
-
Does the discussion agree with what the author said in the results or has he or she made a “big leap” that makes no sense?
-
Is there information in the discussion section that was not detailed in the results section? For example, if the discussion section mentions a significant improvement on a parent rating scale of ADHD or executive functions and there is no mention of this in the results section, there’s something fishy here.
-
Does the discussion make “generalizations” about cause/effect relationships or an outcome being “linked to” an action? If so, what statistical methods are detailed in the methods section? As described above, “correlations” only tell you if there is a relationship between two (or more) measures but do not indicate cause/effect so if the author is suggesting specific “causes/effects” when using correlations, this is an unsupported “leap”.
Need help applying this concept?
Call 817.421.8780 to set up an appointment.
—————————————————
(c) 2013, Monte W. Davenport, Ph.D.
