Test the hypothesis that the average price of diamonds in the available data is larger than $3000
Data : diamonds
Variable : price
Null hyp.: the mean of price = 3000
Alt. hyp.: the mean of price is > 3000
mean t.value p.value df 2.5% 97.5% n sd
3907.186 12.557 < .001 2999 3788.32 Inf 3000 3956.915
Compare diamond prices by the quality of the 'cut'. It seems that diamonds with an ideal cut cost less than than diamonds with a premium cut. Seems strange. Perhaps we should use regression to control for the carats of the diamond. Try it!
Pairwise comparisons (bonferroni adjustment)
Data : diamonds
Variables: cut, price
Samples : independent
mean n sd se ci
Fair 4505.238 101 3749.540 373.093 740.206
Good 4130.433 275 3730.354 224.949 442.848
Very Good 3959.916 677 3895.899 149.732 293.995
Premium 4369.409 771 4236.977 152.591 299.544
Ideal 3470.224 1176 3827.423 111.610 218.977
Alt. hyp. Null hyp. p.value
Good < Fair Good = Fair 1
Very Good < Fair Very Good = Fair 0.886
Premium < Fair Premium = Fair 1
Ideal < Fair Ideal = Fair 0.045 *
Very Good < Good Very Good = Good 1
Premium < Good Premium = Good 1
Ideal < Good Ideal = Good 0.044 *
Premium < Very Good Premium = Very Good 1
Ideal < Very Good Ideal = Very Good 0.044 *
Ideal < Premium Ideal = Premium < .001 ***
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
What are the parameters for the compare_means function? The help text will open up in the Rstudio help file browser.
We can also try to visualize the relationship between diamond prices, carats, and the cut of the diamond.
