This Wall Street Journal article describes how a team of U.S. and European professors traveled to North Korea to teach students there about quantitative methods for studying their population.
North Korea, the story says, has "some of the least reliable statistics in the world."
Please take a look at the pop-out map of North Korea in this story. The map shows the estimated rates of child malnutrition in each of North Korea's provinces. For example, it shows that the estimated rate of child malnutrition in the province of Jagong was 9.8%. That is a frequency claim: a claim about a single variable--the rate of malnutrition--in each province.
The caption on the map is worth attending to from a research methods perspective. The caption reads,
Researchers in North Korea often face challenges. In this 2012 study of child malnutrition, conducted with the support of three U.N. agencies, local village leaders chose the children from which researchers drew their subjects, so they could have excluded the most malnourished kids.
This first part of the caption describes a potential problem with the external validity of this claim. If officials did in fact select only the most healthy kids for the sampling frame, then the estimate of the true rate of child malnutrition would be too low.
Here's the second part of the map's caption:
And the data were collected at the end of the harvest, possibly producing a temporary uptick in nutrition levels.
You could say that this part of the caption is describing a potential problem with the construct validity of this claim. By measuring child malnutrition during a time of plenty, officials did not get a valid measure of the true health of the kids in the sample.
What lessons do you think the instructors are teaching to the students in North Korea?