Question:
Statistics question?
Anonymus
2007-01-06 14:21:27 UTC
Hi
I have a stats question I'm stuck on, can somebody help?
I can't copy the histogram so I've scanned it:
(one of these two links)

[IMG]http://ul.hamzahkhan.com/files/3/Stats/scan0004.jpg[/IMG]

http://ul.hamzahkhan.com/files/3/Stats/scan0004.jpg

Its Part B that I need help on. Sorry if its too big or you can't read my handwriting. Thanks
Seven answers:
saljegi
2007-01-06 16:02:37 UTC
The histogram shows a positive skew which means that the data doesn't follow a normal distribution (symmetrical or evenly spread about the mean). This is because there are a few values at the higher end, possibly outliers, that will distort the mean. For the same reason, the standard deviation would be higher than perhaps it should be and not give a fair representation. A better measure of spread would be the interquartile range as this would show you where the middle 50% of the data is. This could then be displayed in the form of a box & whisker plot with the median shown and the whiskers drawn to exclude the outliers.
modulo_function
2007-01-06 14:37:10 UTC
Mean and standard deviation are most appropriate for a symmetrical distribution. This clearly is not.



For non-symmetrical distributions, median might be more appropriate than mean.



For spread.....I don't rightly know. You essentially want a measure of the 'skewness', ie., how far from symmetrical it is.



The histogram resemble a gamma distribution. I don't think that that is what is asked for.
JasonM
2007-01-06 14:32:12 UTC
I would say because the data is so scewed. Most of the data is found at one end of the histogram, so the mean would be a bit biased because of this. I'd say the median might be a bit better, but not by much.
?
2016-12-15 22:23:27 UTC
Having taken numerous college point information courses, i might tend to question any statistical outcomes published everywhere. there are certainly one of those super variety of vairables that for the main area, information are irrelevant in serving as a correct predictor of a lot of something. BB, Raji the fairway Witch
me
2007-01-06 14:39:01 UTC
The mean: all of the data is shifted towards one side.



Standard deviation: The Frequency Density ranges so much that the average would not be close to any of the numbers.



Don't know any alternatives though, sorry.



Hope that helped.
t11omo
2007-01-06 14:29:33 UTC
Yeah, erm, right! Good luck with that! x
Coach
2007-01-06 14:24:00 UTC
78.4 % of all statistics are just made up on the spot!


This content was originally posted on Y! Answers, a Q&A website that shut down in 2021.
Loading...