In: Statistics and Probability
Attention deficit hyperactivity disorder (ADHD) in young
children presents significant challenges to the children, their
parents, and their classroom teachers. Stimulant medication can
effectively treat 60% to 70% of youth with ADHD. Yet many parents
seek alternative therapies, and Hypericum perforatum (St John's
wort) is 1 of the top 3 alternative therapies used.
In a recent study in the Journal of the American Medical
Association, Hypericum perforatum (St John's Wort) for
Attention-Deficit/Hyperactivity Disorder in Children and
Adolescents, 2008;299(22):2633-2641, the authors conducted a
randomized, double-blind, placebo-controlled trial to determine the
efficacy and safety of St. John's wort for the treatment of ADHD in
children.
After careful pre-screening, 54 children were randomly assigned to
two treatment groups; 27 children were assigned to receive a
placebo treatment and 27 children were assigned to receive a
treatment with St. John's wort.
A measure of "inattentiveness" was used to evaluate the effect of
the treatments. At the beginning of the experiment, the placebo
group and treatment group had the same inattentiveness score. After
4 weeks of the experiment, the data in the table below were
obtained (a lower inattentiveness score indicates a longer
attention span):
Placebo Group | St. John's Wort Group |
x1 = 14.0 | x2 = 13.4 |
s1 = 3.87 | s2 = 3.22 |
n1 = 27 | n2 = 27 |
Question 1. Calculate a 99% confidence interval for the difference μ1 - μ2, where μ1 is the mean inattentiveness score after 4 weeks for all ADHD children who receive a placebo, and μ2 is the mean inattentiveness score after 4 weeks for all ADHD children who receive St. John's wort.
lower bound of interval
upper bound of interval
Question 2.Select the choice below that correctly interprets the interval in question 1.
I am 99% confident that the average change of the inattentiveness score for children in this study is in the interval.
I am 99% confident that the interval in question 1 captures the true mean difference μ1 - μ2; based on this interval it does not appear that St. John's wort is an effective treatment for ADHD.
I am 99% confident that if a child with ADHD is given St John's wort, the child's inattentiveness score will change by an amount in the interval.
If St. John's wort is tested on another sample of ADHD children, there is a 99% chance that their average inattentiveness score will change by an amount in the interval.
The inattentiveness score of 99% of the children studied changed by an amount in the interval.
The probability that μ1 - μ2 is in the interval is 0.99.
Q1:
For Placebo Group:
x̅1 = 14, s1 = 3.87, n1 = 27
For St. John's Wort Group:
x̅2 = 13.4, s2 = 3.22, n2 = 27
99% Confidence interval :
At α = 0.01 and df = n1+n2-2 = 52, two tailed critical value, t-crit = T.INV.2T(0.01, 52) = 2.674
S²p = ((n1-1)*s1² + (n2-1)*s2² )/(n1+n2-2) = ((27-1)*3.87² + (27-1)*3.22²) / (27+27-2) = 12.6727
Lower Bound = (x̅1 - x̅2) - t-crit*√(S²p*(1/n1 +1/n2)) = (14 - 13.4) - 2.674*√(12.6727*(1/27 + 1/27)) = -1.9905
Upper Bound = (x̅1 - x̅2) + t-crit*√(S²p*(1/n1 +1/n2)) = (14 - 13.4) + 2.674*√(12.6727*(1/27 + 1/27)) = 3.1905
-1.9905 < µ1 - µ2 < 3.1905
---
Q2:
Answer:
I am 99% confident that the interval in question 1 captures the true mean difference μ1 - μ2; based on this interval it does not appear that St. John's wort is an effective treatment for ADHD.