Discussion:
Based on the health promotion below, how to complete the research project and what hypothesis testing would be involved.
Health Promotion
A study is planned on the effects of a new health-education program promoting smoking cessation among heavy-smoking teenagers (20 cigarettes/equal to one pack per day). A randomized study is planned whereby 50 heavy-smoking teenagers in two schools (A and B) will receive an active intervention with group meetings run by trained psychologists according to an American Cancer Society protocol; 50 other heavy-smoking teenagers in two different schools (C and D) will receive pamphlets from the American Cancer Society promoting smoking cessation but will receive no active intervention by psychologists.
Random numbers are used to select two of the four schools to receive the active intervention and the remaining two schools to receive the control intervention. The intervention is planned to last for 1 year, after which study participants in all schools will provide self-reports of the number of cigarettes smoked, which will be confirmed by biochemical tests of urinary cotinine levels.
The main outcome variable is the change in the number of cigarettes smoked per day. A participant who completely stops smoking is scored as smoking 0 cigarettes per day. It is hypothesized that the effect of the intervention will be to reduce the mean number of cigarettes smoked by 5 cigarettes per day over 1 year for the active-intervention group. It is also hypothesized that teenagers in the control group will increase their cigarette consumption by an average of 2 cigarettes per day over 1 year.
Let us assume that the distribution of the number of cigarettes smoked per day at baseline in both groups is normal, with mean = 30 cigarettes per day and standard deviation = 5 cigarettes per day. Furthermore, it is expected, based on previous intervention studies, that the standard deviation of the number of cigarettes per day will increase to 7 cigarettes per day after 1 year. Finally, past data also suggest that the correlation coefficient between number of cigarettes smoked by the same person at baseline and 1 year will be 0.80.