{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

The differences could also be explained if

Info iconThis preview shows pages 8–10. Sign up to view the full content.

View Full Document Right Arrow Icon
nonexperimental nature of the study. The differences could also be explained if nearsighted parents were more apt to leave a light on when their children slept. 5. This is a controlled cause-to-effect experiment that attempts to show that coffee grounds suspended as sediment in coffee cause headaches. We suspect there’s a problem, since the IM – 11 | 8
Background image of page 8

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
members of the experimental group knew they were mixing coffee grounds into their pudding. This is enough to give us headaches. The argument gives only weak support to the conclusion, in our opinion. 6. This controlled cause-to-effect experiment is designed to test whether curcumin suppresses the development of colon cancer. Since the immediate target population is rats; a separate analogical argument is required to extend the conclusion to humans. There are six more members of the control group; otherwise, the only difference noted is the addition of curcumin to the diets of the experimental rats. The frequency of effect is 81 percent in the control group and 47 percent in the experimental group. The report seems to imply that the results are significant. The difference (d) here is 34 percentage points, which is large enough to be significant at the .05 level. (A “d” of 27 points is required with experimental and control groups of twenty-five.) This looks like a good study; the smallness of the numbers should encourage further study before a great investment of social resources is made, however. ▲7. Causal claim: Exercise prevents colds. The study is a controlled cause-to-effect experiment, with one experimental group and two control groups. The first control group consists of ten nonexercising volunteers; the second consists of the experimental group prior to the jogging program. The experimental group had 25 percent fewer colds than the first control group and some non-indicated percent fewer than the second control group. We don’t know enough about the groups and how they were chosen to tell if there are significant differences. Given the small size of the groups, a d of 40 percent is necessary to have statistical significance. The 25 percent figure is substantial and may indicate a causal connection, but it isn’t enough to convince us to take up jogging. 8. This huge cause-to-effect experiment was designed to test whether taking aspirin every other day reduces the risk of heart attack. Because of the large numbers, the percentages can be a little misleading. 104 of the experimental group (0.0094 percent) and 189 (0.0171 percent) of the control group had heart attacks, a difference of 0.0077 percent. While this is a tiny percentage difference, it is balanced by the enormous size of the sample: over 22,000 people. (It sounds more impressive when reported, as in the article, as a 47 percent reduction in the risk of heart attack. Results can be misleading when the occurrence of E in one group is reported as a percentage of its occurrence in the other. If there had been two heart attacks in the control group and only one in the experimental group, we’d have a case of a “whopping”
Background image of page 9
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}