Answer:
1.083
Step-by-step explanation:
"1 indicating a coupon and all other outcomes indicating no coupon"
Probability is (number of successful outcomes) / (number of possible outcomes)
Theoretical Probability of rolling a 1: 1/8
Experimental Probability of using coupons: 4/48 = 1/12
So, the experimental probability of a customer using a coupon (that is, 1/12) is smaller than the theoretical probability of rolling a 1 (that is, 1/8).
Answer:
2x
Step-by-step explanation:
u have to add it and do a long method of working out
The data given as a whole would be called ungrouped data. Now to get the variance, you will need the formula:
s^2= <u>Σ(x-mean)^2</u>
n
x = raw data
mean = average of all data
n = no. of observations
s^2 = variance
Now we do not have the mean yet, so you have to solve for it. All you need to do is add up all the data and divide it by the number of observations.
Data: <span>90, 75, 72, 88, 85 n= 5
</span>Mean=<u>Σx</u>
n
Mean = <u>90+75+72+88+85 </u> = <u>410</u> = 82
5 5
The mean is 82. Now we can make a table using this.
The firs column will be your raw data or x, the second column will be your mean and the third will be the difference between the raw data and the mean and the fourth column will be the difference raised to two.
90-82 = 8
8^2 =64
75-82 = -7
-7^2 =49
72-82 = -10
-10^2=100
88-82=6
6^2 = 12
85-82=3
3^2=9
Now you have your results, you can now tabulate the data:
x mean x-mean (x-mean)^2
90 82 8 64
75 82 -7 49
72 82 -10 100
88 82 6 36
85 82 3 9
Now that you have a table, you will need the sum of (x-mean)^2 because the sigma sign Σ in statistics, means "the sum of."
64+49+100+36+9 = 258
This will be the answer to your question. The value of the numerator of the calculation will be 258.
<u>
</u>