MATH 140 Lecture Notes - Lecture 25: Dependent And Independent Variables, Scatter Plot, Confounding
Document Summary
On a scatterplot each dot represents a unit. (cid:272)atte(cid:396) plots shapes (cid:272)a(cid:374) (cid:271)e (cid:862)li(cid:374)ea(cid:396)(cid:863) o(cid:396) (cid:862)cu(cid:396)(cid:448)ed(cid:863), if the(cid:396)e is (cid:374)o (cid:396)elatio(cid:374)ship tha(cid:374) the shape (cid:449)ould (cid:271)e (cid:862)(cid:374)eithe(cid:396)(cid:863) In stats the word linear means that the points on the scatter plot cluster around a summary line: different than algebra where linear means that the points line up on a line. There can be four different linear relationships: strong- has a (cid:862)flat appea(cid:396)a(cid:374)(cid:272)e(cid:863), dots a(cid:396)e (cid:373)o(cid:396)e (cid:272)luste(cid:396)ed, moderate, weak- has a (cid:272)i(cid:396)(cid:272)ula(cid:396) appea(cid:396)a(cid:374)(cid:272)e(cid:863) dots a(cid:396)e sp(cid:396)ead out, none. There are also two trends: positive/increasing: both x and y are directly connected, upward trend. Rough example : negative/decreasing: x and y axis have an inverse relationship, downward trend. Outlier- points that do(cid:374)"t fit the ge(cid:374)e(cid:396)al t(cid:396)e(cid:374)d of the g(cid:396)aph: most of the time the programs can avoid letting the outlier influence the summary line.
