But the accompanying note indicates that only the difference from the
highest to the lowest two is statistically significant and the
next-to-last exceeds that threshold by at best one or two points. So
in reality, the chart says it is essentially a toss-up between
everybody else and Goodman.
Of course, there's insufficient information published on the
free-access page to allow one to determine whether the normalizations
and sample sizes/relative populations, etc., are such as to make the
data have any meaning whatsover (as is usually the case w/ CR imo :( ).
I think it does demonstrate one basic fact as others have mentioned --
buy cheap, get less.
Yep, and to boot, the graph scales and axis labeling are so imprecise
you can't reliably pick out the numerical values to compute actual
differences between computed means. It would have been much more
meaningful if presented as a bar chart w/ estimate and high/low
confidence limits to show the amount of overlap (and thereby also
reflecting the relative precision of estimates for the various brands
owing to sample size). But, that would give away too much as CR is,
above all, mostly about selling CR.
It only makes sense, however, that the low-price-spread has to give up
something for that position in the market so that one conclusion as
noted previously I would venture is probably valid.
But, in all fairness, I do have to give CR credit for at least putting
in the footnote about the size of the overall confidence interval
between estimates -- while most folks who look at it will either ignore
it entirely (as several earlier posters apparently did) or fail to
understand what it means in drawing a conclusion from the data, at
least it was given...
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.