A utility company's line clearance crew must estimate
personnel requirements for the strategic plan of the company. Consequently,
they need to estimate the mean number of labor-hours required to perform
each of the following three clearance tasks: trim a tree, remove a tree,
or cut a square foot of brush. Division standards require them to be 95%
confident they err by no more than 0.04 labor-hour from the true mean for
each category. They divide the total space under all of their lines into
square mile segments, so they can estimate the average hours needed per
mile. One clerk selected a convenient nonrandom sample of values
for each of the three categories, just to get an idea of reasonable magnitudes
before the more formal technical estimates. The three samples follow. The
following questions concern the number of one-mile segments the crew should
randomly select and gather data from in order to meet their accuracy standards.
Labor-hours per
Labor-hours per
Labor-hours per Thousand
Trimmed Tree
Removed Tree
Square Feet of Cut Brush
0.86
0.39
0.80
0.74
0.42
0.57
0.72
0.37
1.50
0.78
0.53
0.21
1.29
0.94
0.32
0.36
.
0.71
0.52
0.96
S 0.25
0.34
0.48
Schuler, Mark D., "A Successful Distribution Line-Clearance Program," Transmission & Distribution, April, 1993.
1. Judging from the convenience
samples' information, which variable will require
the largest
sample? Explain.
2. Estimate the required sample size for each category.
3. Suppose the line crew randomly
selects the number of one-mile segments
suggested in
Question 2 and collects the appropriate set of n values to estimate
each of the
three labor requirements. The standard deviations computed from
these three
samples all fall short of the sample standard deviations computed
from the convenience
samples (0.25, 0.34, and 0.48). Does the line crew need to
increase any
of their sample sizes to maintain 95% confidence of an error of at
most 0.04 labor-hour?
Explain.
Return
to Worksheet Menu
Return
to Statistics Home Page
Return
to Jarrell Home Page