mean and standard deviation.

cmml100

New member
Joined
Jul 18, 2012
Messages
23
Just a really dumb question here. If I surveyed a group of 30 people and a get a mean age of 20 and a standard deviation of 15. What does that tell me?

I am guessing that it implies most of my participants range from age 5 to 35? And given that there are only 30 people, and the wide range, it is a bit pointless to look at the mean and standard deviation after all.

Am I correct in saying that?

Mandy
 
Just a really dumb question here. Who told you that?!


[standard deviation of 15] implies most of my participants range from age 5 to 35? No, that's not my interpretation, based on what you've provided

Okay -- it seems that you have stated a point, above.

it is a bit pointless to look at the mean and standard deviation after all.

Am I correct in saying that?

You tell me. :???: If you were to not have looked at the mean and standard deviation, then you would not have been able to state the first point.

More seriously, I would say that the higher a standard deviation, the farther away from the mean are most of the data.

Standard deviation is what you get when you look at how far each piece of the data is from the mean and then average those numbers.

By the way, do the 30 people surveyed comprise some type of population? Are they a sample from a particular population? Were they selected randomly, as defined by statistics definition? These are considerations that one must understand, before using means or standard deviations to draw specific conclusions.

You did not state any exercise, but, if your question pertains to an exercise that's testing your view of standard deviation, then you should go back to the textbook examples or look for more examples online AND skip over anything that does not click for you in the rest of this post.

Have you yet been introduced to the concept of "normally distributed data"? If not, it's coming. It's relevant.

Again, I do not know the content/context of your survey. Here's part of an example explaining why standard deviation is important (taken from another website). Food for thought, before you ponder the meaning of your standard deviation result again.




If you are comparing test scores for different schools, the standard deviation will tell you how diverse the test scores are for each school.


Let's say Springfield Elementary has a higher mean test score than Shelbyville Elementary. Your first reaction might be to say that the kids at Springfield are smarter.


But a bigger standard deviation for one school tells you that there are relatively more kids at that school scoring toward one extreme or the other. By asking a few follow-up questions you might find that, say, Springfield's mean was skewed up because the school district sends all of the gifted education kids to Springfield. Or that Shelbyville's scores were dragged down because students who recently have been "mainstreamed" from special education classes have all been sent to Shelbyville.


In this way, looking at the standard deviation can help point you in the right direction when asking why information is the way it is.





So, we see that standard deviation gives us information about "how spread apart" the data are, how far "away from the mean" a majority of the data are. That's why the referenced author describes standard deviation as a kind of "average of averages". It really is, too, if you study the formula for it.

Standard deviation becomes clearer, the more examples you look at. Study a little more, and then see what you have to say about the value 15.

Cheers :cool:
 
Last edited:
ok, that makes sense now. But just one more thing. If I cannot guarantee normal distribution, does that mean it is again pointless to look at std dev?

The survey context is this: a poll table was set up at a university during lunch hour. People just came by to fill in a questionnaire regarding certain university policy. It just suddenly occurred to me that the background of the participants could be somehow correlated with the results. May be depending on how long they have been with the university would be make a difference. (Sorry, I used age in the first post. probably the wrong word to describe this). So then I tried to look at the age distribution.

I am now thinking: A) if there is a wide spread of "age", then different age groups are represented. B) if regardless of participant background, the poll as a strong convergence (ie a small std dev), then C) I can conclude that the small scale pilot survey does not indicate any correlation between age and poll opinion.

But then given the context, I am not sure that I have the normal distribution here. I suspect even if I have a much large sample, the distribution would still somehow be a skewed one. Probably skewed towards the lower end I think.
 
By the way, I like to calculate the ratio between the standard deviation and the mean to get a measure of relative dispersion.

Ah, that is lovely. I kind of do that in my head when I look at the numbers. I did not realise we are actually allowed to do that too:D.
 
I would suggest - take a few proper courses in statistics or consult a statistician to analyze your data.

A little statistics is a very dangerous thing.
 
I would suggest - take a few proper courses in statistics or consult a statistician to analyze your data.
A little statistics is a very dangerous thing.

yes, I really need to request that. When I did statistics at school, I simply regurgitated the stuff and got my A's for university. But my strength really lies in Pure and Applied Maths. hmmm... it's time I went back to school for some proper stats. ^_^
 
Top