In a study of obesity the following results were obtained from samples of males and females between the ages of 20 and 75:
n No. Overweight
Male 150 21
Female 200 48
Can we conclude from these data that in the sampled populations there is a difference in the proportions who are overweight? Let α=0.05
Test For Significance of Different Of Proportion
Step1: Set Up Hypothesis
There Is No Significance between them - Under The Null Hypothesis "H_0: p_1 = p_2"
There Is Significance between them - Under The Alternate Hypothesis "H_1: p_1 \\neq p_2"
Step2: Test Statistic
Sample 1
Probability Success "( X_1 )=21"
No. of observaed "(n_1)=150"
"P_1= X_1\/n_1=0.14"
Sample 2
Probability Success "(X_2)=48"
No. of Observaed "(n_2)=200"
"P_2= X_2\/n_2=0.24"
Finding a "\\hat P" value For Proportion "\\hat P=\\dfrac{X_1+X_2}{n_1+n_2}=0.1971"
"\\hat Q" Value For Proportion= "1-\\hat P=0.8029"
if n>30 So, we use Test Statistic "(Z) = \\dfrac{(P_1-P_2)}{(\\sqrt{\\hat P\\cdot \\hat Q(\\frac{1}{n_1}+\\frac{1}{n_2}}))}=\\dfrac{0.14-0.24}{\\sqrt{0.197\\times 0.8029(\\frac{1}{150}+\\frac{1}{200})}}"
"Z_{ cal}=-2.32711\\\\\n\n| Z _{cal} | =2.3271"
Step3: Tabulated Value
The Value of "|Z_{ tab}|" at LOS 0.05% is 1.96
We got "|Z _{cal}| =2.3271" & "| Z_{ tab} | =1.96"
Step4: Make Decision
Hence Value of "| Z_{ cal} | > | Z _{tab}|" and Here we Accept Ho
There Is Significance between them, i.e there is a difference in the proportions
P-Value
Two Tailed ( double the one tail ):
"H_1 : ( P \\neq -2.3271 ) = 0.02"
Comments
Leave a comment