INFERENTIAL STATISTICS

Sampling Theory
Inferential statistics: Refers to the process of selecting and using a sample statistic to draw
conclusions about the population parameter. It involves estimation and testing of hypothesis.
Population: Is a complete set. E.g. the university.
Sample: Is a part or subset of a population under study.
Sampling: Is a process of selecting a sample from a population.
Sampling theory: Is a study of relationship existing between population and samples drawn
from a population.
Random sampling: Is a process of selecting a random sample from a finite population in which
each member has an equal chance of be selected.
Systematic sampling: This relies on arranging the study population to some ordering scheme
and then selecting elements at regular intervals through that ordered list.

Estimation Theory
An estimator is a statistic that estimates some facts about population. E.g. sample mean x is an
estimator of population mean µ, sample variance s2 is an estimator of population varianceδ 2.
A good estimator must be efficient, consistent, sufficient and unbiased.
 An estimator is said to be efficient if it has relative smaller variance or standard
deviation.
 A consistent estimator is the one who’s SD decreases with the increase of sample size.
 An estimator is said to be sufficient if it utilizers all the information in the sample to
arrive to the estimate.
 An unbiased estimator is that on in which expected value is exactly equal to the
parameter being estimated. E.g. x is unbiased estimator of µ since (x )=µ . The following
table shows population parameter and its unbiased estimator

Population parameter Unbiased estimator

Population mean µ Sample mean µ
Population variance δ 2. Sample variance s2
Population SD δ Sample SD s
Population proportion P Sample proportion p ’=x / n , x is number of
observation under category of interest
Total value in population of size N is Nµ Nx
Total number of observation falling under NP ’
category of interest in a pop NP
There are two types of estimators; A point estimator and An interval estimator.
A point estimator draws inference about population by estimating the value of unknown
parameter using a single value. E.g. sample mean x is a point estimator of pop mean µ.

Example 1.
A marketing research analyst collects data for a random sample of 100 customers of the 500who
purchased a particular item under promotion. The 100 people in the sample spent an average
(mean) of Tshs 27,000 in the store with a standard deviation of Tshs 8,000 and 70% of the
customers in the sample made at least one other purchase in addition to the item under
promotion. Based on these results, estimate the values of the following;
a) Mean purchase amount by all 500 customers who purchased the item under promotion.
b) The standard deviation of the distribution of purchase amounts by 500 customers.
c) Total amount of purchases made by 500 customers.
d) The number of customers out of 500 who made at least one other purchase in addition to
the product under promotion.

An interval estimator draws inference about a population by estimating the value of unknown
parameter using an interval. When we use x as an estimator of a population µ we prefer not to
give this as a single value but as an interval within which we are confident that the true value
lies. The interval will be associated with the probability which measures how confident we are
that the true value lies within this interval. Let the confidence level, the probability that the
interval includes the actual value of µ be given by 1−α then the interval estimator of µ is given
δ
by µ=x ± Z α where x is sample mean, n is sample size, δ is pop sd and Z α is the value of z in
2 √n 2

the normal distribution curve. We usually use 1−α between 90 % and 99 % . The following are
commonly used confidence levels and Z
1−α α α Zα -Z α
2 2 2

0.9 0.1 0.05 1.645 -1.645

0.95 0.05 0.025 1.96 -1.96
0.98 0.02 0.01 2.33 -2.33
0.99 0.01 0.005 2.575 -2.575
Example 2
A mining company in Zambia needs to estimate the average amount of copper ore per ton mined.
A random sample of 50 tones gives a sample mean of 146.75kg. The population standard
deviation is assumed to be 35.2kg.
a) Provide a 90% C.I for the average amount of copper per ton.
b) Provide a 95% C.I for the average amount of copper per ton.
c) Provide a 99% C.I for the average amount of copper per ton.

Example 3:
TTCL wants to estimate the average length of long distance calls during weekends. A random
sample of 50 calls gives a mean of 14.5 minutes and standard deviation of 5.6 minutes. Give
95% C.I for the average length of a long distance phone call during weekends.
Example 4:
The makers of a medicated facial cream are interested in determining the percentage of people in
a given age group who may benefit from the ointment. A random sample of 68 people results in
42 successful treatments. Contract a 95% C.I for the proportion of people in the given age group
who may be successfully treated with the facial cream.
Example 5:
A marketing research firm wants to conduct a survey to estimate the average amount spent on
entertainment by each person visiting a popular resort. The people who panned the survey would
like to be able to determine the average amount spent by all people visiting the resort to within
Tshs. 2000/=, with 95% confidence. From past operation of the resort, an estimate of the
population standard deviation is Tshs. 4,000/=.what is the minimum required sample size?
Example 6:
Find the minimum required sample size of accounts if the proportion of accounts in error is to be
estimated to within 0.02 with 95% confidence. A rough guess of proportion of accounts in error
is 0.1.
 TUTORIAL 1 QUESTIONS
QUESTION 1:
A management consulting firm needs to estimate the average number of years of experience of
executives in a given branch of management. A random sample of 32 executives gives a mean of
6.7 years and standard deviation of 2.5 years. Give a 95% C.I for the average number of years of
experience for all executives in this branch.
QUESTION 2:
In a department store an auditor finds that 30 randomly chosen charge accounts of a total of
3,000 such accounts have a mean debit balance of Tshs. 32,800/= with a standard deviation of
Tshs. 8,400/=.
a) Estimate the mean balance of all accounts using 99%
b) Estimate the total balance due for all 3,000 charge accounts using 99%

QUESTION 3:
As an aid in the establishment of personnel requirements, the director of Muhimbili National
Hospital (MNH) wishes to estimate the mean number of people who are admitted to the
emergency room during 24-hour period. The director randomly selects 64 different 24-hour
periods and determines the number of admissions for each. For this sample mean is 19.8 and
variance 25. Estimate the number of admissions per 24-hour period with 95% confidence interval
at MNH.

QUESTION 4:
In checking the reliability of bank’s records, auditing firms sometimes ask a sample of bank’s
customers to confirm the accuracy of their savings account balances as reported by the bank.
Suppose an auditing firm interested in estimating P, the proportion of bank’s savings accounts on
whose balances the bank and its customers disagree. Of 200 saving accounts customers
questioned by the auditors, 15 said their balance disagree with that reported by the bank.
a) Use 95% confidence interval to estimate the actual proportion of the bank’s savings
accounts on whose balances the bank and customers disagree.
b) How many savings account customers should the auditors question if they want to
estimate P to within 0.02 with probability equal to 0.95?

QUESTION 5:
In a small firm employing 100 people a random sample shows that 35 employees earn an average
hourly wage of Tshs. 475/=with a standard deviation of Tshs. 25/=. Suppose the wages are
approximately normally distributed, estimate the average hourly wage of all 100 employees,
using 95% confidence limits.
QUESTION 6:
An analyst wishes to estimate the mean monthly wages of workers in a particular company
within Tshs. 2500/= and with 95% confidence. The standard deviation of the wages rates is
estimated as being Tshs. 10,000/=. What is the number of personnel records that should be
sampled as the minimum to satisfy this research objective?
QUESTION 7:
A marketer of candy products wants to estimate the total number of candy bars sold in a given
month in all 500 candy stores in a given area. A random sample of 32 stores gives a mean of
1220 and standard deviation of 550. Construct a 95% C.I for the total number of candy bars sold
in all 500 stores.