In: Statistics and Probability
A team of researchers wanted to assess the effectiveness of the 4 components of a weight loss program by comparing the distribution of weight status across 4 different treatment groups. The data is summarized in the table below:
Intervention Component: Normal weight / Overweight / Obese
Control Group(n=100) 50 35 15
Calorie Counting(n=100) 70 20 10
Physical Activity(n=100) 73 15 12
Combination(n=100) 80 15 5
Is there a difference in weight status by intervention component? (Hint: Are intervention component and weight status independent?) Run the appropriate test at a 5% level of significance.
Indicate the correct competing hypotheses:
H1: The null hypothesis is not false.
H1: The null hypothesis is false.
H1: The null hypothesis is not false.
H1: The null hypothesis is false.
The equation is supplied below, but indicate WHAT TYPE of Chi-square test you need to conduct:
χ2 =(sigma)O-E2 / E
Indicate the decision rule:
Compute the expected (E) number of students for each category:
Compute the test statistic and indicate the outcome:
True or false: We DO NOT have statistically significant evidence at α=0.05 to show that intervention component and weight status are not independent.
Indicate the most accurate p-value associated with the above-mentioned conclusion:
Column 1 | Column 2 | Column 3 | Total | |
Row 1 | 50 | 35 | 15 | 100 |
Row 2 | 70 | 20 | 10 | 100 |
Row 3 | 73 | 15 | 12 | 100 |
Row 4 | 80 | 15 | 5 | 100 |
Total | 273 | 85 | 42 | 400 |
Null and alternative hypothesis:
Answer B.
H0: Intervention component and weight status are
independent.
H1: The null hypothesis is false.
Type of chi square:
Answer B : Test of independence
Critical value:
At = 0.05 and df =(4-1)(3-1) =6, critical value, =CHISQ.INV.RT(0.05, 6) = 12.59
Decision rule:
Answer D : Reject H0 if > 12.59
Expected number(E) :
table with squared distance:
Test statistic:
Answer: A.
e: Answer B. False.
p-value = CHISQ.DIST.RT(24.974, 6) = 0.0003
Answer D