Tuesday, December 12, 2023

A sesame seed bagel is like a national survey with a random sample

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sampling is an important statistical part of conducting a survey statistically. I was looking for an analogy (simile or metaphor) to explain a national survey with a random sample. As shown above, it is like a sesame seed bagel. Each seed can represent an individual in the sample. They are spread over the entire surface of the bagel, which represents the nation. After time passes the survey gets stale.  

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Also, as shown above, there is a margin of error associated with sampling. For a margin of plus or minus three percent a sample size of 1,067 is needed.

 

How about a nonrandom sample? Statistical jargon calls this a convenience sample, for which the Wikipedia page says:    

 

“Convenience sampling (also known as grab sampling, accidental sampling, or opportunity sampling) is a type of non-probability sampling that involves the sample being drawn from that part of the population that is close to hand.”

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Often that convenient survey sample consists of university students in introductory public speaking or psychology classes at a single university. As shown above, it is like a bagel with just a single pecan. Results from such a sample do not represent a nation.

 

Back on December 20, 2016 I blogged about Bursting the overblown claim that 95% of Americans fear public speaking at some level. A comment on that post by Michelle Mazur warned:

 

“McCroskey & Richmond's research was all based on convenience sampling of undergrads at West Virginia University. To say that is generalizable to the greater population defies all logic and good research practices. Public polling actually uses randomized sampling that is representative of the population. …”

 

The image for margin of error was colored in from this one at Wikimedia Commons.

 


No comments:

Post a Comment