A journalist for the foodie magazine Bon Appetit reported on an empirical article that found a correlation between type of diet and level of depression:
A new study found that people who excluded meat from their diets had twice as many depressive episodes as omnivores.
The journalist summarized the methodology this way:
To investigate the relationship between a meatless diet and depression among adults, a team of Brazil-based researchers gathered data from more than 14,000 people between the ages of 35 and 74 over a six-month period. They evaluated participants using the Clinical Interview Schedule-Revised instrument, a questionnaire that helps diagnose common mental disorders like depression and anxiety. “Depressive episodes are more prevalent in individuals who do not eat meat,” the authors concluded.
Here are some questions to answer about this study.
a) First, use this table to list and classify the study's variables, based on what you've read so far in the journalist's summary.
| Conceptual variable name | How was this variable operationalized? | What seem to be this variable's levels? | Is the variable measured or manipulated? |
b) Now sketch a small graph of the study's outcomes. Before you start, consider--would a bar graph or a scatterplot make more sense?
c) Is this study correlational or experimental? How do you know?
Let's talk about whether this study can support a causal argument. Specifically, a reader might wonder, "Does being vegetarian cause people to become depressed?"
The first criterion, covariance, is met by the results of the study. They reported that "people who excluded meat from their diets had twice as many depressive episodes as omnivores."
The second criterion, temporal precedence, is not met. It seems that this was a single survey, meaning that both variables (type of diet and depression level) were measured at the same time, so we can't be sure which comes first. Indeed, the journalist included a comment about this: "[a nutrition expert] also hypothesized that plant-based eating and mental health outcomes could be associated in reverse: People who are already depressed might try and change their diets to feel better."
The third criterion, internal validity, is also not met by a correlational study. There might be a third, "C" variable that goes with both A (vegetarian diet) and B (depression). The journalist addressed this issue too: "One user implied that vegans and vegetarians are probably more tuned in to the realities of our macabre meat industry and the impending environmental collapse, which could be more of a bummer than a lack of chicken nuggets." In other words, being tuned into the "macabre meat industry" could be separately linked to both vegetarianism and depression.
d) Next let's evaluate the four big validities in this study. I'll put a quote on the left, and you can decide which of the four big validities it is addressing (hint: Some of the big validities appear more than once):
| Quote from the article: | Which big validity is this quote about? (Construct, statistical, internal, or external?) |
| Mary Mosquera-Cochran, a registered dietitian at The Ohio State University, told Healthline that the 82 vegetarians (out of 14,000 total people surveyed) wasn’t a large enough sample group. | |
| This isn’t the first study to explore a correlation between diet and mental health, though findings over the past couple of decades have been mixed. Australian research published in 2007 found that vegetarians had poorer mental health, with 21–22% reporting depression compared with 15% of meat eaters. A 2012 German study found that mental disorders tended to follow the onset of a vegetarian diet. But a French study published earlier this year found no relationship between a plant-based diet and mental health outcomes. | |
| The researchers don’t yet know what’s causing the surprising relationship, but say they’ve ruled out some obvious suspects: nutritional deficiencies, socioeconomic conditions, and lifestyle factors like drinking, smoking, and differing levels of physical activity.(This quote is referring to regression, which controlled for several possible third variables) | |
| They evaluated participants using the Clinical Interview Schedule-Revised instrument, a questionnaire that helps diagnose common mental disorders like depression and anxiety. |
