Assignment 2
Ray Park
Learning Objectives: Learn what are the general properties of a good estimate, What
are the estimators for mean, total and variance of using a simple random sample from a
finite population and what are their properties? How to determine an appropriate sample
size for simple random sampling. How to compute estimates of mean, total, proportions
using stratified sampling. How to best select sample sizes when using stratified
sampling.
Activities: Complete Assignment 2
1. An investigator is interested in estimating the total number of “count trees”
(trees larger than a specified size) on a farm of 1000 acres. This information
is used to estimate the total volume of lumber for trees on the farm and
may be used to determine the amount to be paid to the owner for
harvesting trees on this farm. A simple random sample of 100 1-acre plots
was selected and each plot was examined for the number of count trees.
The sample average for the 100 1-acre plots was 25.2 and the sample
variance was 136
a. Estimate the total number of count trees on the farm
b. Place a bound on the error of estimation
c. Use the results of this survey to determine the sample size required
to estimate the total with a bound on the error of estimation of 1500
a)
b)
c)
2. An opinion poll on education questioned 1684 adults across the United
States. For results dealing with proportions of respondents favoring certain
issues the poll is reported to have a 6% margin of error. Do you agree?
Discuss why or why not.
3. A study in a small town calls for estimating the proportion of households
that contain at least one member over the age of 65. The town has 621
households . A simple random sample of 60 households was selected. 11 of
the households in the sample contained at least one member over the age
of 65. Estimate the true population proportion and place a bound on the
error of estimation.
4. In exercise 4, how large of a sample should be taken to estimate the
population proportion with a bound of .05 on the error of estimation in two
situations:
a. Find a conservative estimate assuming no knowledge of the true
proportion p
b. Find a more realistic estimate assuming that p is relatively small
(around p=.2)
c. Which of the above assumptions do you think is more reasonable?
Give a reason for your answer
a)
b)
c)
With a bound of error of 0.05, I think taking the second, n=182, could be more reasonable.
Because, if we are too conservative using p=0.5, we are essentially estimating that almost half of
households have at least one member over age of 65. But I think that might be too high of an
estimated proportion, so using p=0.2 and having n=182 sample size might be more realistic.
5. A forester wants to estimate the total number of farm acres planted with trees
for a state. Because the number of acres of trees varies considerably with the size
of the farm he decides to stratify on farm size. The 240 farms in the state are
classified into one of 4 categories according to size. A stratified random sample of
40 farms, selected by using proportional allocation yields the results shown in the
table below on the number of acres planted in trees.
a) Estimate the total number of acres of trees on farms in the state and place a
bound on the error of estimation.
b) Graph the data on an appropriate plot and comment on the variation as we
move from Stratum 1 to Stratum 4
StratI StratII StratIII StratIV
N1=86 N2=72 N3=52 N4=30
97 125 142 167
67 155 256 655
42 67 310 220
125 96 440 540
25 256 495 780
92 47 510
105 310 320
86 236 396
27 220 196
43 352
45 142
59 190
53
21
c) The study above is to be repeated yearly and a bound on the error of
estimation of 5000 would be useful. Find an approximate sample size to achieve
this bound if Neyman Allocation is used. Allocate this sample to the 4 strata.
a)
b)
c)
6. Government is interested in expanding the facilities of a day-care center for
mentally retarded children. A sample survey will be conducted to estimate the
proportion of families with retarded children that will make use of the expanded
facilities. The families are divided into those that use the existing facilities and
those that do not. Some families live in the city in which the center is located and
some live in rural areas. This, stratified sampling is used with 4 strata: (1)Users in
the city, (2) users in rural areas, (3)nonusers in city and (4)nonusers in rural areas.
It is estimated (guessed) that 90% of present users and 50% of present nonusers
will use the expanded facilities. The cost of obtaining an observation from a user
is $4 and from a nonuser is $8. Existing records give strata sizes N1=97, N2=43,
N3=145, and N4=68
A) Find the approximate sample size and allocation necessary to estimate
the population proportion with a bound of .05 on the error of estimation
B) Suppose the survey is conducted with the sample sizes in A) and yields
the following proportions p1 = .87, p2=.93, p3=.6 and p4=.53 Estimate the
population proportion and place a bound on the error of estimation. Was
the desired bound achieved?
C) Suppose the total cost of sampling is fixed at $400. Choose the sample
size and allocation that minimizes the variance of the estimator at this fixed
cost.
a)
b)
c) Used values from above (way right side of part a)