In: Math
Researchers selected a simple random sample of 4048 medical
records of adults diagnosed with gum disease. In all, 2226 were
current smokers, 891 were former smokers, and 931 never smoked
regularly. Their research question is:
Do these data indicate that gum disease is equally likely
regardless of smoking status?
Using a significance level of 0.05, what is the appropriate conclusion for this test?
The data are consistent with an equal representation of current, former, and never smokers among adults diagnosed with gum disease.
Current smokers make up a significantly greater proportion of adults diagnosed with gum disease than former or never smokers.
Current smokers are most likely to have gum disease.
There is significant evidence that current, former, and never smokers are not equally represented among adults diagnosed with gum disease.
Chi square test for Goodness of fit
expected frequncy,E = expected proportions*total
frequency
total frequency= 4048
observed frequencey, O | expected proportion | expected frequency,E | (fi - ei) | (O-E)²/E | |
2226 | 0.333 | 1349.333 | 876.667 | 569.573 | |
891 | 0.333 | 1349.333 | -458.333 | 155.684 | |
931 | 0.333 | 1349.333 | -418.333 | 129.696 |
chi square test statistic,X² = Σ(O-E)²/E =
854.9531
level of significance, α= 0.05
Degree of freedom=k-1= 3 -
1 = 2
P value = 0.0000 [ excel function:
=chisq.dist.rt(test-stat,df) ]
Decision: P value < α, Reject Ho
There is significant evidence that current, former, and never smokers are not equally represented among adults diagnosed with gum disease.