Question

Question 1

Between 20 and 25 violent acts are portrayed per hour on children’s TV. A study by Johnson et al (2002) looked for links between TV viewing and aggression by following the TV viewing habits of children 1-10 years old. Of these, 88 watched < 1 hr TV per day, 386 watched 1-3 hrs and 233 watched > 3 hrs. Eight years later, researchers looked to see if any of the children had records of aggressive assaults on other people. The numbers of children with aggressive assaults were 5, 87 and 67 for each TV category, respectively. The remaining children were not aggressive.

1. Create a contingency table of the results from the study. Carry out an appropriate statistical test to determine whether TV viewing is associated with future violence. What is the conclusion from your test?
2. Do the results of the study prove that TV watching behavior causes increased aggression in children? Why?

Question 2

It is thought that children who attend daycare centers build a more robust immune system because they are exposed to more germs. A study by Gilham et al (2005) examined whether there was a relationship between attending daycare and the probability of later developing the white blood cell disease Acute Lymphoblastic Leukemia (ALL). They compared 1272 children with ALL to 6238 children without ALL. Of the kids with All, 1020 went to daycare. Of the kids without ALL, 5343 went to day care. The rest of the children did not go to daycare or have regular contact with other children.

1. Is this an experimental or observational study?
2. What are the odds of kids who went to daycare developing ALL? What are the odds for the kids who did not go to daycare?
3. What is the odds ratio for ALL, comparing kids who went to day care and those that did not?

Answer 1

Solution:

2)The contingency table for the given data is as follows:

Daycare	With ALL	Without ALL	Total
Yes	1020(a)	5343(b)	6363
No	252(c)	895(d)	1147
Total	1272	6238	7510

where a=Number of exposed cases

b=Number of exposed non-cases

c=Number of unexposed cases

d=Number of unexposedcnon-cases

a) This study is an observation study. Here the researcher observes the subject to analyse the variable of interest without assigning any treatment to the subject under study.

b) The odds of kids who went to daycare developing ALL=1020/252=4.05

The odds for the kids who did not go to daycare=252/1020=0.25

c)Odd ratio=ad/bc=(1020*895) / (5343*252)=912900/1346436=0.68