Distinguishing correlational and experimental research

Requirements:

• Your reflection post needs to be at least 10 full sentences; there is no maximum limit
• Your post must be written in your own words 
• If you refer to a source such as an article, video, or a book, provide a link or other identifying information about the source. Sources are not required for this reflection post, but they might be relevant. The source could be our textbook. If you refer to sources, use APA format citations (in-text citations and a reference list; see the APA resources in Canvas)
• Respond to 2 other students’ posts. Your responses have no specific length requirements, but you need to refer to something in the student’s post, showing that you read it (something like “I agree” or “Interesting!” is not enough for the points). If you respond to a student that initially responded to you, that also counts as one of your responses (the idea is obviously to generate discussion here)

GRADING:

10 points for your own post (you get the full points if your post is at least 10 sentences and your discussion is relevant to the topic; your opinions, ideas and experiences are not graded for “accuracy”)

REFLECTION TOPIC: CORRELATION

One of the important distinctions between correlational and experimental research methods is that while experiments can tell us something about cause and effect (because we carefully manipulate some variable and try to control extraneous variables), correlational research cannot (because we only observe and measure the variables and don’t manipulate them). Just because X is related to Y does not mean that X causes Y.

One example of this is that there’s a correlation between length of marriage and hair loss in men, such that men who have been married for a longer time tend to have less hair on their head. Does this mean that marriage causes hair loss? Well, maybe, but there’s probably a third variable involved here: age (older men tend to have less hair and are more likely to have been married longer than younger men).

Another example: Women who give birth at older maternal ages live longer than women whose last child is born when they are younger. Does giving birth at an older maternal age cause women to live longer? Not necessarily; a possible third variable might be rate of aging (how fast a person ages): women who age slower are still able to have children at older ages, and they also live longer because they age slower (this means that giving birth did not cause the increased longevity).  For another famous third variable example, see my lecture for this module.

Note that the third variable must cause/explain BOTH of the correlated variables.

Question : We often believe that two things are related, even though in actuality we just notice random coincidences. For example, some people may believe that when they wash their car it will soon rain, and when couples adopt they immediately get pregnant. We easily perceive patterns even when they are not there. This phenomenon is called an illusory correlation. Can you identify any illusory correlations that you or people you know hold (or have held in the past)?