Discussing Scatterplot
General Pattern
Deviations
Clusters? (not in Activstats)
Outliers? (label if can)
Shape (linear, curved (bent) ,
...?)
Strength of relationship (how unfuzzy) "Weak,
moderate, strong"
Direction
Positively associated: y increases
as x increases (generally).
Negatively associated: y decreases as
x increases.
Mark subgroups differently to do comparisons. (Subgroups defined
by categorical variable, like Sex, Region of country)
Example: Activstats (7) Scatterplots HW RMW-3, Governer's salaries
vs. Avg. pay (1993)
| Activstats: Ch.8, Correlation, pp. 1-2, + 1st activity p. 3.
The Teacher note at the bottom of p. 2 --"Correlation with the Scatterplot
tool"--belongs with the 1st activity on p. 3, which introduces the tool.
I like a little better the similar tool that WHFreeman provide on the text
website. More on that next time.
I strongly advise doing the activities with aqua or yellow blackboards--they should cement the properties of r. Must know: Correlation with SPSS. Properties of r (first activity, 8-1, or Moore, pp. 99-100) |
| Hand in:
Relationships & Scatterplots: Moore sec. 2.1 Moore p.81, 2.1 (rel. or explan/resp) DoMoore 2.17a, p.98, graphing by hand! Moorep. 91ff, 2.8 gpa vs. IQ 2.10 states'mean/median income -From now on, make all scatterplots on SPSS! The rest are from Activstats Ch 7 (Scatterplots) HW unless otherwise noted. MRA-2 Speed vs. fuel consumption.(SPSS) This is also Moore p.87, 2.6 Save the data file and output file to your own disk, for later use. MRA-4 Metabolic rates. (SPSS)This is also (almost) Moore, p.89, 2.7 Graph the males and females simultaneously. (Use the handout.) In the Interactive graphs, there isn't an easy way to do just the females. (In the Noninteractive graphs, if you want to print just the females, change the male points' color to white.) Save your data file and output file for Moore's problem 2.22, later. MRB-1 Is wine good for your heart? (SPSS) This is also Moore p. 93 2.11. Note that this is wine consumption. Label France, where almost all alcohol drunk is wine, and Ireland and Great Britain, which have substantial alcohol consumption but not as wine. MRB-2. This is also Moore p. 95, 2.13corn plant density. (SPSS) (Notice how the data is entered for SPSS--not as displayed in the text! but with the first column giving planting rates and the second giving yield. To get means for the separate groups, do Analyze>Desc.Stats>Explore, Yield =>Dependent, Rate=> Factor. By hand, mark them on your SPSS graph and draw the lines. Notice the boxplots that come free with that analysis show the medians for each group, a similar result.) (This problem looks forward to Sec. 2.3, sort of) |
Read, to discuss
Moore p. 82, 2.3 (Breast CA vbles) |
Optional (more practice)
Moore p. 92, 2.9 hotdogs&sodium
|
| Sievers home | Math151-Sp02/Day11.htm | 1:30pm | 2/20/02 |