Anda di halaman 1dari 30

Module E - Overview of Sampling

MODULE E
Overview of Sampling
LEARNING OBJECTIVES
Review
Checkpoints
Exercises, Prolems,
and Simulations
!" #nderstand the asic principles of sampling,
including the differences etween statistical and
nonstatistical sampling and sampling and
nonsampling risk"
!, $, %, &, ', ( '!, '$, '%, '&, ''
)parts a * c+, '(, (%
)parts a * c+, ((, (,,
,! )parts a * e+
$" #nderstand the asic steps and procedures used
in conducting a sampling plan"
,, -, ., !/, !!, !$,
!%
'' )part d+, ',, '-,
'., (/, (!, ($, (%
)parts d * g+, (&, (',
(-, ,/
%" 0dentif1 the two situations in which sampling is
used in an audit examination"
!&, !', !(, !,, !-,
!., $/, $!
,!
&" #nderstand how the asic steps and procedures
used in a sampling plan appl1 to an audit
examination"
$$, $%, $& (., ,$, ,%, ,&
SOLUTIONS OR REVIE! C"EC#$OINTS
E"! Sampling can e used 1 the auditor during the stud1 and evaluation of a client2s internal control
and the sustantive procedures"
E"$ Sampling risk is the possiilit1 that the decision made ased on the sample differs from the
decision that would have een made if the entire population had een examined, a sampling error"
Sampling error arises when the sample drawn from the population does not appropriatel1 represent
that population"
E"% 3onsampling risk represents the proailit1 that an incorrect conclusion is reached ecause of
reasons unrelated to the nature of the sample, a nonsampling error" 3onsampling error arises
ecause of errors in 4udgment or execution of the sampling plan"
E"& Sampling risk is controlled 1 the auditor 1 )!+ determining an appropriate sample si5e and )$+
evaluating sample results to consider the possiilit1 that the sample does not appropriatel1
represent the population"
MO6E-!
Module E - Overview of Sampling
E"' Statistical sampling plans appl1 the laws of proailit1 to select sample items for examination and
evaluate sample results" Statistical sampling plans differ from nonstatistical sampling plans in
terms of the methods used to determine the appropriate sample si5e and evaluate the sample
results" 0n a statistical sampling plan, these methods control exposure to sampling risk, whereas
the1 do not do so in a nonstatistical sampling plan"
E"( Either statistical sampling or nonstatistical sampling can e used under generall1 accepted
auditing standards" 3onstatistical sampling should not e used solel1 to reduce sample si5e"
E", !" 6etermine the o4ective of sampling
$" 6efine the characteristic of interest
%" 6efine the population
&" 6etermine the sample si5e
'" Select sample items
(" Measure sample items
," Evaluate the sample results
E"- 0t is important to carefull1 define the population of interest, since the results of the entire sampling
application will e ased on the population from which the sample is drawn"
E". Sampling risk has an inverse relationship with sample si5e7 that is, as a lower level of sampling
risk is necessar1, the individual needs to select a larger sample )and vice versa+"
E"!/ 8our methods commonl1 used to select sample items are )!+ unrestricted random selection, )$+
s1stematic random selection, )%+ hapha5ard selection, and )&+ lock selection"
9hen using unrestricted random selection, a series of random numers is identified and the
random numers are matched to numered items in the corresponding population"
9hen using s1stematic random selection, a random starting point is selected within the
population" : fixed numer of items are 1passed and the corresponding item in the population is
selected" ;his process is continued until a numer of items e<ual to the appropriate sample si5e is
selected"
=apha5ard selection identifies sample items in a nons1stematic manner, with no delierate effort
to match random numers to sample items"
>lock selection identifies a series of contiguous )ad4acent+ units for selection"
E"!! #nrestricted random selection or s1stematic random selection are used with statistical sampling
ecause these methods )!+ provide a reasonale likelihood of selecting a representative sample, )$+
allow the proailit1 of selecting sample items to e determined, and )%+ allow the sample
selection process to e replicated"
E"!$ ;he precision )or allowance for sampling risk+ is the numeric distance from the estimated
population value in which the true )ut unknown+ population value ma1 lie with a given
proailit1"
Reliailit1 )or confidence+ is the likelihood of achieving a given level of precision"
;he precision interval is a range around the sample estimate that has a likelihood e<ual to
reliailit1 )or !// percent minus the sampling risk+ of including the true population value"
MO6E-$
Module E - Overview of Sampling
E"!% ;he following are the asic steps in evaluating sample results?
!" Select and measure sample items to determine the sample estimate"
$" >ased on the acceptale sampling risk, determine the reliailit1 and related precision"
%" 8orm the precision interval 1 adding and sutracting the precision from the sample
estimate"
&" 6etermine whether the h1pothesi5ed )or acceptale+ value falls within the precision
interval"
E"!& :ttriute sampling is used to determine the extent to which a particular characteristic )or attriute+
exists within a population" 0n an audit examination, attriute sampling is used in the stud1 and
evaluation of internal control and suse<uent assessment of control risk"
E"!' ;he tolerale deviation rate is the maximum rate of deviations from a control that an auditor will
permit without reducing the planned reliance on internal control" ;he auditor will compare an
@ad4ustedA sample deviation rate )upper limit deviation rate+ to the tolerale deviation rate to
determine the extent to which the auditor can rel1 on internal control"
E"!( ;he risk of assessing control risk too high )risk of underreliance+ occurs when the auditor2s sample
indicates that the control is not functioning effectivel1 when, in fact, it is doing so" 9hen this risk
occurs, the auditor2s upper limit deviation rate exceeds the tolerale deviation rate" =owever,
unknown to the auditor, the true population deviation rate is less than the tolerale deviation rate"
;he risk of assessing control risk too low )risk of overreliance+ occurs when the auditor2s sample
indicates that the control is functioning effectivel1 when, in fact, it is not" 9hen this risk occurs,
the auditor2s upper limit deviation rate is less than the tolerale deviation rate" =owever, unknown
to the auditor, the true population deviation rate exceeds the tolerale deviation rate"
;he assessing control risk too high results in an efficienc1 loss for the auditor, since more
extensive sustantive procedures are performed than is necessar1" ;he assessing control risk too
low exposes the auditor to an effectiveness loss, since the auditor2s sustantive procedures will not
reduce audit risk to the acceptale level"
E"!, ;he risk of assessing control risk too low is of greater concern to the auditor, since it ma1
eventuall1 result in the auditor issuing an un<ualified opinion on financial statements that are
materiall1 misstated"
E"!- Bariales sampling is used to examine a population when the auditor wants to estimate the amount
)or value+ of some characteristic of that population" Bariales sampling is used 1 the auditor
when performing sustantive procedures to evaluate the fairness of an account alance or class of
transactions"
E"!. ;olerale error is the amount of misstatement that the auditor is willing to allow in an account
alance or class of transactions without concluding that it is materiall1 misstated" ;he auditor will
compare an @ad4ustedA sample misstatement )upper error limit+ to the tolerale error to determine
whether the account alance is materiall1 misstated"
E"$/ ;he two sampling risks associated with variales sampling are the risk of incorrect acceptance and
the risk of incorrect re4ection" ;he risk of incorrect acceptance is the likelihood that the sample
results indicate the account alance is fairl1 stated when, in fact, it is materiall1 misstated" ;he
risk of incorrect re4ection is the likelihood that the sample results indicate the account alance is
materiall1 misstated when, in fact, it is fairl1 stated"
MO6E-%
Module E - Overview of Sampling
0ncorrect acceptance exposes the auditor to an effectiveness loss, ecause the auditor will make an
incorrect conclusion and issue an inappropriate opinion on the financial statements"
0ncorrect re4ection exposes the auditor to an efficienc1 loss, ecause additional transactions or
components will e examined 1 the auditor prior to proposing an ad4ustment to the client2s
account alance"
E"$! ;he auditor is more concerned with the risk of incorrect acceptance ecause it ma1 result in
issuing an un<ualified opinion on financial statements that are materiall1 misstated"
E"$$ ;he o4ective of attriute sampling is to assess the operating effectiveness of a ke1 control" ;he
o4ective of variales sampling is to estimate the amount of misstatement in an account alance or
class of transactions"
E"$% ;he factors that affect the sample si5e in an attriute sampling application )as well as their
relationship to sample si5e+ are shown elow?
Population si5e )direct relationship+
Expected deviation rate )direct relationship+
;olerale deviation rate )inverse relationship+
Sampling risk )inverse relationship+
E"$& ;he factors that affect the sample si5e in a variales sampling application )as well as their
relationship to sample si5e+ are shown elow?
Population si5e )direct relationship+
Expected error )direct relationship+
;olerale error )inverse relationship+
Sampling risk )inverse relationship+
Population variailit1 )direct relationship+
SOLUTIONS OR MULTI$LE%C"OICE &UESTIONS
E"$' a" 0ncorrect 9hen sampling, the auditor performs procedures on less than !//
percent of the items in a alance"
" Correct 9hen sampling, the auditor performs procedures on less than !//
percent of the items in a alance to form a conclusion aout the entire
alance"
c" 0ncorrect >ecoming familiar with an accounting s1stem is not an application of
audit sampling"
d" 0ncorrect :nal1tical procedures are not an application of audit sampling"
E"$( a" 0ncorrect Statistical sampling is characteri5ed 1 statistical calculation of the
results"
" 0ncorrect Statistical sampling is characteri5ed 1 representative sample selection"
c" Correct Statistical sampling is characteri5ed 1 oth representative sample
selection and statistical calculation of the results"
d" 0ncorrect Statistical sampling is characteri5ed 1 oth representative sample
selection and statistical calculation of the results"
MO6E-&
Module E - Overview of Sampling
E"$, a" 0ncorrect Sampling is t1picall1 most appropriate for populations consisting of a
large numer of items"
" Correct Sampling is appropriate when the need for precise information aout
the population is less important"
c" 0ncorrect More critical decisions would t1picall1 increase the need to examine
the entire population"
d" 0ncorrect Sampling is not appropriate when the costs of an incorrect decision are
extremel1 high"
E"$- a" 0ncorrect :udit risk is the risk that the auditor issues an un<ualified opinion on
financial statements that are materiall1 misstated"
" 0ncorrect ;here is no term known as examination risk"
c" Correct ;his response represents the correct definition of sampling risk"
d" 0ncorrect 3onsampling risk is related to errors in 4udgment and execution and not
to the nature of the sample"
E"$. NOTE TO INSTRUCTOR: Since this question asks students to identify the statement that will not
assist in controlling the auditors exposure to sampling risk, the response labeled correct will
not assist in controlling the auditors exposure to sampling risk and the those labeled incorrect
will assist in controlling the auditors exposure to sampling risk
a" 0ncorrect ;his method assists in controlling the auditor2s exposure to sampling
risk"
" Correct Performing the appropriate audit procedure is related to nonsampling
risk, not sampling risk"
c" 0ncorrect ;his method assists in controlling the auditor2s exposure to sampling
risk"
d" 0ncorrect ;his method assists in controlling the auditor2s exposure to sampling
risk"
E"%/ a" 0ncorrect Onl1 statistical sampling methods measure the auditor2s exposure to
sampling risk"
" 0ncorrect Cenerall1 accepted auditing standards permit the use of either statistical
sampling or nonstatistical sampling methods"
c" 0ncorrect Samples can e selected either randoml1 or 4udgmentall1 under
nonstatistical sampling methods"
d" Correct 3onstatistical sampling is t1picall1 less complex than statistical
sampling"
E"%! a" 0ncorrect >lock selection identifies a series of contiguous units for examination"
" 0ncorrect =apha5ard selection is characteri5ed 1 the auditor selecting sample
items in a nons1stematic fashion"
c" 0ncorrect S1stematic random selection uses a random starting point and then
1passes a fixed numer of items in selecting sample items"
d" Correct #nrestricted random selection uses a series of random numers to
identif1 sample items"
E"%$ a" 0ncorrect See the response to )c+ elow"
" 0ncorrect See the response to )c+ elow"
c" Correct ;he random starting point )!/+ is the first item selected" ;he sampling
interval would e calculated as ' )!// $/ D '+" :dding the sampling
interval to the random starting point would result in item !' eing
selected )!/ E ' D !'+, followed 1 item $/ )!' E ' D $/+"
d" 0ncorrect See the response to )c+ aove"
MO6E-'
Module E - Overview of Sampling
E"%% a" 0ncorrect S1stematic random selection provides a relativel1 high likelihood of
1ielding a representative sample"
" 0ncorrect ;here is no difference in the sample si5e provided 1 s1stematic
random selection and other selection methods"
c" Correct >ecause s1stematic random selection 1passes items etween selection,
a limitation ma1 occur if the population is not randoml1 ordered"
d" 0ncorrect S1stematic random selection can e used with statistical sampling
plans"
E"%& a" 0ncorrect >lock selection cannot e used with statistical sampling applications"
" Correct Of these two methods, onl1 unrestricted random selection can e used
with statistical sampling applications"
c" 0ncorrect #nrestricted random selection can e used with statistical sampling
plans7 lock selection cannot e used with statistical sampling
applications"
d" 0ncorrect #nrestricted random selection can e used with statistical sampling
applications"
E"%' a" 0ncorrect >lock selection cannot e used with statistical sampling applications"
" 0ncorrect 3either lock selection nor hapha5ard selection can e used with
statistical sampling applications"
c" 0ncorrect =apha5ard selection cannot e used with statistical sampling
applications"
d" Correct #nrestricted random selection and s1stematic random selection can e
used with statistical sampling applications"
E"%( a" 0ncorrect ;he confidence is the likelihood that the precision interval contains the
true )ut unknown+ population value"
" 0ncorrect ;he mean is the average of the oservations in the sample"
c" Correct ;he precision represents a range around the sample estimate that
has a certain likelihood )e<ual to reliailit1+ of including the true
population value"
d" 0ncorrect ;he precision interval is the sample estimate plus and minus the
precision"
E"%, a" Correct ;he confidence is the likelihood that the precision interval contains the
true )ut unknown+ population value"
" 0ncorrect ;he mean is the average of the oservations in the sample"
c" 0ncorrect ;he precision represents the closeness of a sample estimate to the true
population value"
d" 0ncorrect ;he likelihood that the interval contains the true population value is
confidence, or one minus sampling risk"
MO6E-(
Module E - Overview of Sampling
E"%- NOTE TO INSTRUCTOR: Since this question asks students to identify the statement that is not
true, the response labeled correct is not true and those labeled incorrect are true
a" 0ncorrect >ecause there is a ./ percent )! minus sampling risk+ proailit1 that
the interval contains the true population value, there is a !/ percent
proailit1 that the true population value lies outside of this interval"
" Correct ;here is a !/ percent proailit1 )sampling risk+ that the true
population value is less than (/ or greater than ,/"
c" 0ncorrect Reliailit1 e<uals one minus sampling risk, or ./ percent )! minus !/
percent D ./ percent+"
d" 0ncorrect Since the precision interval is s1mmetrical around the sample estimate,
the sample estimate is the average of the precision interval, or (' F)(/ E
,/+ $ D ('+G" ;he precision can e determined as the distance
etween either ound of the precision interval and the sample estimate,
or ' )(' * (/ D '+"
E"%. a" 0ncorrect Sampling risk can occur in either statistical or nonstatistical sampling
applications"
" 0ncorrect ;his response is the definition of nonsampling risk"
c" 0ncorrect ;his response is the definition of inherent risk"
d" Correct ;his response is the definition of sampling risk"
E"&/ a" Correct >oth the risk of incorrect acceptance and risk of assessing control risk
too low relate to the effectiveness of an audit"
" 0ncorrect ;he risk of incorrect re4ection and the risk of assessing control risk too
high relate to the efficienc1 of the audit"
c" 0ncorrect Onl1 the risk of assessing control risk too low is related to control risk
assessments"
d" 0ncorrect Onl1 the risk of incorrect acceptance is related to evidence aout
assertions in financial statements"
E"&! a" Correct :ttriute sampling is most fre<uentl1 used during the auditor2s stud1 of
internal control"
" 0ncorrect Control sampling is not a t1pe of sampling"
c" 0ncorrect Proailit1 proportional to si5e sampling is a form of variales
sampling, which is used in the auditor2s sustantive procedures"
d" 0ncorrect Bariales sampling is used in the auditor2s sustantive procedures"
E"&$ a" 0ncorrect 9hile attriute sampling occurs as part of the use of the audit risk
model, it is most closel1 associated with the assessment of control risk"
" Correct :ttriute sampling is most closel1 associated with the assessment of
control risk"
c" 0ncorrect 9hile attriute sampling will allow the auditor to determine the
acceptale level of detection risk, it is most closel1 associated with the
assessment of control risk"
d" 0ncorrect :ttriute sampling is unrelated to inherent risk"
MO6E-,
Module E - Overview of Sampling
E"&% a" 0ncorrect 0f the upper limit deviation rate exceeds the tolerale deviation rate,
auditors will reduce their reliance on controls" ;hese responses are
reversed"
" Correct 0f the upper limit deviation rate exceeds the tolerale deviation rate, the
auditor will reduce their reliance on controls"
c" 0ncorrect ;he auditor2s evaluation of controls is not related to the expected
deviation rate"
d" 0ncorrect ;he auditor2s evaluation of controls is not related to the expected
deviation rate"
E"&& a" 0ncorrect Statistical sampling methods do not necessaril1 provide the auditor
with greater assurance"
" Correct Statistical sampling methods allow the auditor to
<uantitativel1 measure the exposure to sampling risk"
c" 0ncorrect Either statistical or nonstatistical sampling methods can
convert samples into dual-purpose tests for sustantive procedures"
d" 0ncorrect Hudgments are re<uired to assess various factors that affect
sample si5e under oth statistical and nonstatistical sampling"
E"&' a" 0ncorrect See the response to )d+ elow"
" 0ncorrect See the response to )d+ elow"
c" 0ncorrect See the response to )d+ elow"
d" Correct ;he auditor would compare an estimation of the deviation rate to the
tolerale deviation rate when using sampling in the stud1 of internal
control"
E"&( NOTE TO INSTRUCTOR: Since this question asks students to identify the item that would not
expose an individual to nonsampling risk, the response labeled correct would not expose an
individual to nonsampling risk and the responses labeled incorrect would expose the individual
to nonsampling risk
a" 0ncorrect Measuring the characteristic in an inappropriate manner would result in
exposure to nonsampling risk"
" Correct 0f items are selected that are not representative of the population, this
represents sampling risk, not nonsampling risk"
c" 0ncorrect : mistake in measurement )whether intentional or unintentional+ would
result in exposure to nonsampling risk"
d" 0ncorrect >ecause )+ would not result in exposure to nonsampling risk, this
choice would not e appropriate"
E"&, NOTE TO INSTRUCTOR: Since this question asks students to identify the statement that is not
true with respect to nonstatistical sampling, the response labeled correct is not true and those
labeled incorrect are true
a" Correct Either statistical or nonstatistical sampling can e used in an audit
conducted in accordance with generall1 accepted auditing standards"
" 0ncorrect 3onstatistical sampling considers various factors in determining sample
si5e"
c" 0ncorrect 3onstatistical sampling does provide an estimate of the characteristic of
interest"
d" 0ncorrect 3onstatistical sampling re<uires the use of 4udgment in estalishing
factors that will e used to determine sample si5e, among other areas"
MO6E--
Module E - Overview of Sampling
E"&- a" 0ncorrect >ecause a ./ percent proailit1 exists that the average weight is
etween !!/ and !%/ pounds, the likelihood that the average weight is
greater than !%/ pounds is less than !/ percent, since the average
weight could e less than !!/ pounds"
" 0ncorrect >ecause a ./ percent proailit1 exists that the average weight is
etween !!/ and !%/ pounds, the likelihood that the average weight is
less than !!/ pounds is less than !/ percent, since the average weight
could e greater than !%/ pounds"
c" 0ncorrect ;he likelihood that the average weight is less than !!/ pounds or
greater than !%/ pounds is one minus reliailit1, or !/ percent )not ./
percent+"
d" Correct : reliailit1 )./ percent+ proailit1 exists that the average weight is in
the interval ounded 1 the sample estimate I precision )in this case,
!$/ pounds I !/ pounds, or !!/ pounds to !%/ pounds+"
E"&. a" Correct ;his step would e performed last )see responses elow+"
" 0ncorrect ;he desired level of reliailit1 must e determined prior to selecting the
sample, which precedes examining sample items"
c" 0ncorrect 6etermining the o4ective of the sampling application is the first step in
the sampling application"
d" 0ncorrect 6etermining the appropriate sample si5e would occur prior to
examining sample items"
E"'/ a" 0ncorrect 0n oth cases, the correct decision with respect to the client2s internal
control or account alances should ultimatel1 e made 1 the auditor"
" Correct >oth risks ma1 result in the auditor performing more extensive sustantive
procedures than necessar1 to control audit risk to acceptale levels"
c" 0ncorrect Onl1 the risk of incorrect re4ection is related to preliminar1 estimates of
materialit1"
d" 0ncorrect Onl1 the risk of incorrect re4ection is related to tolerale error"
SOLUTIONS OR E'ERCISES( $ROBLEMS( AND SIMULATIONS
E"'! Sampling Risk
a" Sampling risk is the risk that the decision made ased on the sample is different from the
decision that would have een made if the entire population were examined"
;he two t1pes of sampling risk are if 1ou conclude that )!+ the temperature will e aove
'/ degrees when it will e elow '/ degrees and )$+ the temperature will e elow '/
degrees when it will e aove '/ degrees"
" ;he costs of committing sampling risk if 1ou conclude that the temperature will e aove '/
degrees is that 1ou will not pack heav1 clothing and will e uncomfortale or forced to
purchase heav1 clothing during 1our trip"
MO6E-.
Module E - Overview of Sampling
;he cost of committing sampling risk if 1ou conclude that the temperature will e elow
'/ degrees is that 1ou will unnecessaril1 pack heav1 clothing"
c" 0f 1ou use statistical sampling, 1our sampling plan would consider the acceptale level of
sampling risk in determining the appropriate sample si5e and evaluating sample results"
3onstatistical sampling methods would not consider sampling risk in determining sample
si5e or evaluating sample results"
d" ;he primar1 advantage of statistical sampling is that it allows 1ou to explicitl1 control 1our
exposure to sampling risk" ;he primar1 disadvantage of using statistical sampling is that
it is t1picall1 more complex and time-consuming that nonstatistical sampling"
E"'$ Sampling and Nonsampling Risk
a" 3onsampling risk )misplaced distance markers would result in the failure to accuratel1
record 1ardage+"
" Sampling risk )golfers pla1ing in a clu championship would t1picall1 e of a higher skill
level than those not pla1ing in a clu championship+"
c" 3onsampling risk )ecause a %-wood will not hit the all as long as a driver, 1our
conclusion ma1 e affected 1 the clu used and not the golf all+"
d" Sampling risk )a sample of golfers that includes onl1 females will not e representative of
the population+"
e" 3onsampling risk )the conclusions ma1 e affected 1 the fact that the comparison golf
alls are older, not inferior in <ualit1 to the 9ilson golf alls+"
f" 3onsampling risk )failure to accuratel1 record 1ardage ma1 influence 1our conclusions+"
NOTE TO INSTRUCTOR: Some students might classify items !c" and !e" as neither sampling
risk nor nonsampling risk, since they do not apparently relate to an error made in recording
results #owever, one could contend that the failure to ensure that the proper golf club or a new
golf ball is being used as a form of error $he important point to raise is that these issues do not
relate to the representativeness of the sample, while items !b" and !d" do relate to the
representativeness of the sample
E"'% Sampling and Nonsampling Risk
!" ;his situation is characteristic of sampling risk if the passengers seated in the rows 1ou
selected had either higher or lower income than that of the average passenger on the
airplane" ;his could occur if one of the rows 1ou randoml1 selected was in the first class
cain or usiness class cain )in this case, 1our sample would t1picall1 provide a higher
average income than the average passenger on the plane+" 0n addition to this example, the
possiilit1 exists that individuals of relativel1 high and low incomes could e seated in
different rows throughout the airplane"
$" ;his situation is characteristic of nonsampling risk ecause of the error made 1
including swimmers from events other than the freest1le race" ;his error is not related to
the representativeness of the sample of swimmers, ut including events other than that of
interest in the sample"
MO6E-!/
Module E - Overview of Sampling
%" ;his situation is characteristic of sampling risk, ecause various characteristics of honors
students )intelligence, extent of preparation, diligence in completing the examination,
etc"+ ma1 not e representative of the population of students" 0nterestingl1, it is not clear
whether the sample of honors students will complete the examination more <uickl1
)ecause of superior intelligence and examination preparation+ or more slowl1 )ecause
of more diligence in completing the examination+ than non-honors students"
&" ;his situation is characteristic of sampling risk, ecause the sample of students attending
schools in small, college towns are likel1 to differ from the overall population of sixth
grade students" Specificall1, these students are likel1 to e more aware of college
education in general7 in addition, their parents are more likel1 to have received a college
education, as the1 are emplo1ed 1 the universit1" :s a result, it is likel1 that including
these students in 1our sample would provide a relativel1 high sample estimate of the
percentage of students who plan on attending college"
E"'% Sampling and Nonsampling Risk (Continued)
'" ;his situation is characteristic of nonsampling risk" :ssuming 1ou selected a
representative sample, 1our sampling procedure might 1ield an inappropriate conclusion
ecause of mistakes in converting various currencies into #"S" dollars" ;hese mistakes
are unrelated to the representativeness of 1our sample"
E"'& Basic Sampling
a" Sampling is the process of making a statement aout a population of interest ased on
examining onl1 a suset )or sample+ of that population"
" ;he primar1 advantage of sampling is efficienc17 1ou could make 1our conclusion in a
fraction of the time )1 examining onl1 a suset of all students enrolled at 1our
universit1+ than would e necessar1 if the entire population were examined"
;he primar1 disadvantage of sampling is related to effectiveness" ;he decision reached
ased on the sample might differ from the decision that would e made if the entire
population were examined"
c" Sampling is more likel1 to e used if )!+ the need for exact information is less important
)for example, it is less important to know if 1ou are taller than the average student at 1our
universit1+ or )$+ if the numer of students enrolled at 1our universit1 is larger"
E"'' Basic Sampling
a" Some possile methods of estimating the numer of patrons on the evenings 1ou visit
include?
Counting the patrons at one show time for one screen and assuming that all other
show times and screens are similar in terms of attendance"
Counting the patrons at all show times for one screen and assuming that all other
screens are similar in terms of attendance"
Counting the patrons at all screens for one show time and assuming that the
other show times are similar in terms of attendance"
Counting all patrons attending on a given evening for all show times and
screens"
Count patrons at a random sample of shows taken from the ,-/ )$! x $/ E .x&/+
shows"
MO6E-!!
Module E - Overview of Sampling
" ;o ensure a representative sample, 1ou should consider the following events or
characteristics that ma1 result in nonrepresentative levels of attendance )students ma1 list
others+?
@>lockusterA or other popular movies that ma1 have an unusuall1 large
audience"
;he existence of other local events )such as a high school graduation or sporting
event+ that ma1 result in unusuall1 low attendance at the theater"
Poor weather, which has an indeterminate effect on attendance )some patrons
ma1 prefer to attend movies during poor weather7 others ma1 find it too
uncomfortale to leave their homes+"
Cood weather, which ma1 result in unusuall1 low attendance if patrons
preferred other t1pes of activities to attending movies"
;he time of the show )matinee versus evening versus late+, as overall attendance
ma1 differ 1 time"
;he da1 of the week )popular television shows airing on certain weeknights ma1
influence attendance+"
9eekend versus weekda1s )weekends would t1picall1 have more attendance+"
E"'' Basic Sampling (Continued)
c" ;he two t1pes of sampling risk are )!+ concluding that the average attendance exceeds
!',/// patrons per month when it is less than !',/// patrons per month and )$+
concluding that the average attendance is less than !',/// patrons per month when it
exceeds !',/// patrons per month"
Of these two risks, most people would conclude that the first of these is greater" 0f 1ou
elieve that the average attendance will support the theater and it will not, 1ou will lose
1our investment and have an unsuccessful usiness" 0f 1ou elieve that the average
attendance will not support the theater and it will, 1ou will have missed a potentiall1
profitale opportunit1"
d" 0n all of these cases, 1ou will need to convert the dail1 estimates to monthl1 estimates 1
multipl1ing 1 %/ da1s"
!" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D (// %/ D ',/ to (%/ )dail1+
Precision 0nterval D !,,!// to !-,.// )monthl1+
$" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D (-/ !'/ D '%/ to -%/ )dail1+
Precision 0nterval D !',.// to $&,.// )monthl1+
%" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D &./ %' D &'' to '$' )dail1+
Precision 0nterval D !%,('/ to !',,'/ )monthl1+
0n the first two cases, 1ou would conclude that the movie theater would e profitale,
since the lower ound of the precision interval exceeds the criterion level of attendance
)!',/// patrons+" 0n the third case, 1ou would e unale to reach this conclusion, since
the lower ound of the precision interval )!%,('/+ is less than the criterion level of
attendance )!',/// patrons+"
MO6E-!$
Module E - Overview of Sampling
E"'( Basic Sampling
a" Sampling is the process of making a statement aout a population of interest ased on
examining onl1 a suset )or sample+ of that population"
;he primar1 advantage of sampling is the time savings of examining onl1 a suset of
flights instead of all flights 1 competing airlines" ;he primar1 disadvantage of sampling
is that, if a nonrepresentative sample of flights is selected, the results ma1 not trul1 reflect
the population of interest and 3ortheast :irlines ma1 reach an incorrect conclusion"
" ;o identif1 the population of flights, 1ou would first need to clearl1 define the
airlines against which 3ortheast wishes to compete" 0n this case, it appears that there are
four other airlines" Once these airlines have een defined, 1ou could otain a flight
schedule or other information that contained a listing of all flights into the proposed
@huA airline" ;his list would represent the population of flights from which 1ou would
select 1our sample"
E"'( Basic Sampling (Continued)
c" Sampling risk is the possiilit1 that the decision made ased on the sample differs
from the decision that would have een made if the entire population were examined" ;he
primar1 cause of sampling risk is the selection of a sample that is not representative of
the population from which it is drawn"
3onsampling risk is the proailit1 that an incorrect conclusion is reached ecause of
reasons unrelated to the nature of the sample" ;he primar1 cause of nonsampling risk is a
mistake in evaluating sample items or interpreting sample results"
Sampling risk could occur if the sample of flights 1ou select differs from the population
of flights into the proposed airport )see )d+ for a more complete discussion of wh1 this
might occur+" 3onsampling risk could occur if some error was made in classif1ing a
flight )an on-time flight was erroneousl1 classified as late, or vice versa+ or an error was
made in compiling the sample results or calculating the sample estimate"
d" 9hile not comprehensive, some characteristics that might influence the on-time
arrival of flights )and, therefore, exposure to sampling risk+ are?
:irline )ecause of relative operating advantages and disadvantages, some
airlines will have a greater percentage of on-time arrivals than others+"
Origin of flight )flights from us1 airports are more likel1 to e dela1ed7 flights
from areas with more severe weather are more likel1 to e dela1ed+"
;ime of da1 of flight )flights with peak departure and arrival times are more
likel1 to e dela1ed+"
Jength of flight )longer flights ma1 have more opportunit1 to compensate for
ground dela1s and e more likel1 to arrive on time+"
6a1 of week )flights during peak travel da1s are more likel1 to e dela1ed+"
MO6E-!%
Module E - Overview of Sampling
Capacit1 of flight )more crowded flights are more likel1 to e dela1ed as a result
of longer passenger oarding and seating times+"
e" :fter estimating the on-time arrival rate for the airlines currentl1 serving this airport,
1ou would form a precision interval ased on the desired level of reliailit1" ;he on-time
arrival rates included in this interval could e compared to 3ortheast2s rates )-$ percent+
to see if 3ortheast could e competitive at this airport"
E"', Sample Evaluation
a" Sample Estimate? ;he estimate of the true population value ased on the sample drawn to
represent that population"
Precision? ;he precision is the numeric distance from the estimated population value in
which the true )ut unknown+ population value ma1 lie with a given proailit1"
Reliailit1? ;he proailit1 the true )ut unknown+ population value lies within the
precision interval"
" ;he preferred candidate could have etween &$ percent and '& percent of the vote )&-
percent ( percent+"
c" : .. percent likelihood exists that the preferred candidate2s share of the vote is etween
&$ percent and '& percent"
d" Since onl1 two candidates are seeking office, it is uncertain as to whether the preferred
candidate will e ale to attract '/ percent of the vote, since the share could e as low as
&$ percent"
e" 0n this case, the preferred candidate2s vote could range from &, percent to &. percent )&-
percent ! percent+" 9hile this suggests that the candidate has at least &, percent of the
vote, it also means that the candidate almost certainl1 cannot expect a ma4orit1 of the
vote"
E"'- Sample Evaluation
:" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D '( $/ D %( to ,(
>" Sampling Risk D ! - Reliailit1
Sampling Risk D ! - /".' D /"/' or 'K
C" ;he precision can e determined 1 taking the difference etween either ound of the
precision interval and the sample estimate )for example, -' * -/ D '+"
6" Sampling risk D ! - Reliailit1
/"!/ D ! * Reliailit1
Reliailit1 D /"./ or ./K
E" ;he sample estimate can e determined 1 taking the difference etween either ound of
the precision interval and the precision )for example, !%! * !/ D !$!+"
MO6E-!&
Module E - Overview of Sampling
8" Sampling Risk D ! - Reliailit1
Sampling Risk D ! * /".- D /"/$ or $K
E"'. Sample Evaluation
a" Sample !?
Precision 0nterval D Sample Estimate Precision
Precision 0nterval D $( ' D $! to %!
Sample $?
Precision 0nterval D Sample Estimate Precision
Precision 0nterval D %& % D %! to %,
Sample %?
Precision 0nterval D Sample Estimate Precision
Precision 0nterval D &/ - D %$ to &-
" Sample !? ;here is a .' percent likelihood that the average age of an 38J fan is
etween $! and %! 1ears"
Sample $? ;here is a .' percent likelihood that the average age of an 38J fan is
etween %! and %, 1ears"
Sample %? ;here is a .' percent likelihood that the average ago of an 38J fan is
etween %$ and &- 1ears"
c" 8or sample !, since the upper ound of the precision interval )%! 1ears+ is less than %' 1ears,
the 38J could relial1 conclude that the average age of its fan ase is less than %' 1ears"
8or samples $ and %, ecause the upper ound of the precision interval )%, 1ears and &-
1ears, respectivel1+ exceeds %' 1ears, the 38J could not relial1 conclude that the
average age of its fan ase is less than %' 1ears at the specified confidence"
d" 0f sampling risk is increased, the reliailit1 )or confidence+ decreases, since reliailit1 e<uals
one minus sampling risk" :s a result, the precision interval would tighten and the 38J
would e more likel1 to conclude that its fan ase has an average age of less than %'
1ears )assuming that the sample estimate is an accurate representation of the true
population average+"
On the other hand, if sampling risk is decreased, the precision interval would widen and
the 38J would e less likel1 to conclude that its fan ase has an average age of less than
%' 1ears )assuming that the sample estimate is an accurate representation of the true
population average+"
MO6E-!'
Module E - Overview of Sampling
E"(/ Sample Evaluation
a" Precision is the numeric distance from the estimated population value in which the true
)ut unknown+ population value ma1 lie with a given proailit1" Reliailit1 is the
proailit1 the true )ut unknown+ population value lies within the precision interval"
;hese terms are related in that a level of precision is associated with a given level of
reliailit1" :s a higher )lower+ level of reliailit1 is desired, a wider )tighter+ precision
interval results"
" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D $"' /", D !"- to %"$
c" : lower sampling risk will correspond with a higher reliailit1" >ecause the precision
interval will need to have a larger likelihood of including the true population value
)reliailit1+, the level of precision will increase, resulting in a wider precision interval"
d" !" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D $"' /", D !"- to %"$
>ecause the lower ound of the precision interval is greater than !"', Cloria
would conclude that the average numer of children per household exceeds !"'
children"
$" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D $"' !"& D !"! to %".
>ecause the lower ound of the precision interval is less than !"', Cloria would
not e ale to conclude that the average numer of children per household
exceeds !"' children"
%" Precision 0nterval D Sample Estimate Precision
Precision 0nterval D $"' !"- D /", to &"%
>ecause the lower ound of the precision interval is less than !"', Cloria would
not e ale to conclude that the average numer of children per household
exceeds !"' children"
e" ;he differences noted in )d+ aove are the direct result of different levels of reliailit1" 0f
a given interval needs to have a higher likelihood of including the true )ut unknown+
population value )higher reliailit1 or confidence and lower sampling risk+, that interval
needs to e larger )wider+" ;his relationship can e seen examining the differences in the
three scenarios noted aove"
MO6E-!(
Module E - Overview of Sampling
E"(! Sample Evaluation
a" >ased on the sample results, :lice can conclude that there is a .' percent proailit1 that
her level of support is etween &- percent and '- percent )'% percent I ' percent+"
;herefore, while a reasonale proailit1 exists that she might win in the election, it is
less than the desired .' percent level and :lice could not have .' percent confidence of
receiving a ma4orit1 of the allots cast" ;his is ecause the precision interval includes
levels elow '/ percent"
" !" 0n this case, the level of support would e etween &' percent and (! percent )'% percent
I - percent+" Since the precision interval includes levels elow '/ percent, :lice
could not have .. percent confidence of receiving a ma4orit1 of the allots cast"
$" 0n this case, the level of support would e etween '! percent and '' percent
)'% percent I $ percent+" Since the precision interval does not include levels
elow '/ percent, :lice could have ./ percent confidence of receiving a
ma4orit1 of the allots cast"
c" :s one wishes to e more confident that the precision interval includes the true )ut unknown+
population value )i"e", higher level of reliailit1+, the precision interval needs to e wider
)i"e", larger precision+" 0n this particular case, as :lice desires higher levels of confidence
regarding her ailit1 to receive a ma4orit1 of the allots cast, the lower ound of the
precision interval will e lower"
d" Summari5ed elow is a rief description of how some of the issues in 1our notes of the
sampling process affect :lice Evans2 ailit1 to rel1 on the sample evidence )numers
correspond to numers from 1our sampling notes+" Man1 of these issues relate to whether
the individual)s+ surve1ed were representative of the population of voters in :lice2s
district )sampling risk+" ;here are no definitive @rightA or @wrongA answers7 the main
o4ective is to have students egin to understand how various facets of a sampling plan
ma1 introduce sampling and nonsampling risk into the evaluation process"
!" :lice2s ailit1 to rel1 on the results could e affected if )!+ the four
neighorhoods chosen were not representative of the population of eight
neighorhoods comprising :lice2s district and )$+ individuals who responded to
the door-to-door in<uir1 were not representative of those who did not respond"
$" :lice2s ailit1 to rel1 on the results could e affected if the individuals who were
home during the hours in which the neighorhood was canvassed were not
representative of voters in :lice2s district"
%" 9hile this <uestion does not introduce sampling risk, respondents ma1 view it
differentl1 )and respond differentl1+ than a more direct <uestion aout whether
the1 would vote for :lice Evans"
E"(! Sample Evaluation (part d, Continued)
&" ;he use of telephone surve1s in one neighorhood ma1 result in a
nonrepresentative sample, particularl1 if individuals screened their phone calls
or otherwise chose not to participate )although the1 could also do so for a door-
to-door surve1+"
MO6E-!,
Module E - Overview of Sampling
'" Cases where a household indicated support for :lice while displa1ing signs for
her rival ma1 reflect an undecided voter or a desire to minimi5e the amount of
time spent with the pollster" 0n an1 case, these instances should e interpreted
carefull1"
(" Clinton2s misunderstanding will oviousl1 have an effect on the results, ecause
trul1 undecided voters ma1 provide responses that do not trul1 reflect their
feelings" Similar to )'+, these responses should e interpreted carefull1"
," >ush2s mistake ma1 result in nonregistered voters2 responses eing included in
the sample results" Since these individuals will not vote, the1 are not a part of
the population of voters and their inclusion in the sample will misrepresent
:lice2s level of support )this misrepresentation could e in either a favorale or
unfavorale direction+"
E"($ Sample Evaluation
a" ;he sample estimate would e L!,,'/,/// )'/,/// x L%' D L!,,'/,///+"
" Since the sample estimate )L!,,'/,///+ exceeds the targeted amount to e raised
)L!,'//,///+, it appears that the fundraising campaign could e successful"
c" ;he primar1 limitation of rel1ing onl1 on the sample estimate is that the sample of
citi5ens ma1 not e representative of the population of the College >r1an area" More
specificall1, these individuals ma1 e willing to provide a greater level of financial
support to the fundraising campaign than citi5ens who were not included in the sample"
;his would result in Marts elieving that the campaign would e successful when, in fact,
it would not e successful"
d" Sampling risk is the possiilit1 that the decision made ased on the sample differs from
the decision that would have een made if the entire population were examined"
8actors that could result in Marts2 exposure to sampling risk are characteristics of sample
items that result in the sample not eing representative of the population from which it is
drawn" Since this example represents an individual2s willingness to participate in a
fundraising campaign to assist in uilding a new recreation center, the following should
e considered?
=ousehold income )those with higher incomes could e more likel1 to support the
campaign+"
:ge and composition of household, particularl1 the numer of children )those with
children could e more likel1 to support the campaign+"
Past support of various fundraising efforts )those who have supported efforts in the
past could e more likel1 to support the campaign+"
Proximit1 of address to proposed recreation center )those living closer to the facilit1
could e more likel1 to support the campaign+"
MO6E-!-
Module E - Overview of Sampling
E"($ Sample Evaluation (Continued)
e" Sample Estimate I Precision D Precision 0nterval
!" L!,,'/,/// I L!//,/// D L!,('/,/// to L!,-'/,///
$" L!,,'/,/// I L$//,/// D L!,''/,/// to L!,.'/,///
%" L!,,'/,/// I L%//,/// D L!,&'/,/// to L$,/'/,///
f" >ased on the precision intervals calculated in )e+, it appears that the fundraising campaign
is likel1 to e successful" 0n cases )!+ and )$+, the lower ound of the precision interval
exceeds the fundraising target of L!,'//,///" 0n the one case where the lower ound is
less than the fundraising target, it is ver1 close to this target )L!,&'/,///+" :ssuming that
these precisions corresponded to a reliailit1 of .. percent, the likelihood that Marts will
exceed its fundraising target and have a successful fundraising campaign is <uite high"
g" :s the reliailit1 decreases, the precision will also decrease" 0n this instance, since the
sample estimate exceeds the fundraising target, a decrease in the precision will result in
the lower ound of the precision interval eing higher" :s a result, Marts would e more
likel1 to conclude that its fundraising campaign would e successful"
E"(% Basic Sampling: Comprehensive
a" ;he primar1 advantage of sampling is efficienc17 Reagan could make his conclusion in a
fraction of the time )1 examining onl1 a suset of all households in :n1town, #S:+
than would e necessar1 if the entire population were examined"
;he primar1 disadvantage of sampling is related to effectiveness" Reagan2s decision
ased on the sample might differ from the decision that he would make 1 examining the
entire population"
" ;he advantage of statistical sampling is that it allows Reagan to explicitl1 control his
exposure to sampling risk" ;he primar1 disadvantage of statistical sampling is that it is
t1picall1 more costl1 and time-consuming that nonstatistical sampling"
c" !" Sampling risk is the possiilit1 that the decision made ased on the sample differs
from the decision that would have een made if the entire population were
examined"
3onsampling risk is the risk that an incorrect conclusion is reached ecause of
reasons unrelated to the nature of the sample"
$" Reagan can control his exposure to sampling risk 1 mathematicall1
determining the appropriate sample si5e and mathematicall1 evaluating sample
results" =e can control his exposure to nonsampling risk primaril1 1 exercising
care during the sampling and evaluation process"
%" :n example of sampling risk would e selecting a nonrepresentative sample"
;his ma1 occur if Reagan limited his selection of households to those in either
unusuall1 high or low income neighorhoods"
MO6E-!.
Module E - Overview of Sampling
Examples of nonsampling risk would e failing to properl1 record household
income levels or making a mathematical error in taulating the sample results"
E"(% Basic Sampling: Comprehensive (Continued)
d" Reagan would identif1 !// random numers )either using a random numer tale or
computer program+ and match those random numers to items in the population" Once
these items were identified, Reagan would then determine the income for these
households"
e" Selecting all of the houses on a small numer of streets ma1 not provide a representative
sample, since income levels would proal1 e fairl1 similar on these streets" Essentiall1,
Reagan would e sampling a small numer of items )streets+ and would risk selecting
streets that did not represent the population of :n1town, #S: with respect to average
levels of household income" ;his is an example of lock selection"
f" ;he precision interval would e L%(,/// to L&$,/// )L%.,/// L%,/// D L%(,/// to
L&$,///+"
>ased on this information, 1ou would tell Reagan that there is a ./ percent )! minus
sampling risk of !/ percent+ proailit1 that the true average household income is
etween L%(,/// and L&$,///" :s a result, Reagan could relial1 conclude that the
average household income exceeds L%',///, since the lower ound of the precision
interval )L%(,///+ exceeds this criterion level"
g" ;he precision interval would e L%$,/// to L'$,/// )L&$,/// L!/,/// D L%$,/// to
L'$,///+"
>ased on this information, 1ou would tell Reagan that there is a ./ percent )! minus
sampling risk of !/ percent+ proailit1 that the true average household income is
etween L%$,/// and L'$,///" 9hile the average household income could e as high as
L'$,///, it could also e as low as L%$,///" :s a result, Reagan could not relial1
conclude that the average household income exceeds L%',///"
E"(& Sample Selection
a" Sampling risk is the possiilit1 that the decision made ased on the sample differs from the
decision that would have een made if the entire population were examined" :rthur can
control sampling risk 1 )!+ mathematicall1 determining the appropriate sample si5e and
)$+ mathematicall1 evaluating the sample results"
" ;he population would e defined as all registered voters residing in =oops Count1" ;hese
voters can e defined using the appropriate 5ip codes in the electronic data file
maintained in the State Commissioner2s office"
c" >ecause the population is maintained in an electronic format, :rthur can use a computer
program with an unrestricted random selection method or a s1stematic random selection
method" Prior to doing so, he needs to identif1 the appropriate 5ip code)s+ for voters
registered in =oops Count1"
MO6E-$/
Module E - Overview of Sampling
9hen using an unrestricted random selection method, :rthur would identif1 a series of
random numers and match those numers to voters in the data file" ;his can e done
using voter registration numers or simpl1 renumering the voters in the data file
eginning with the numer !" 0f he chooses a s1stematic random selection method, :rthur
would identif1 a random starting point in the population and then select ever1 n
th
voter in
the data file"
:lternativel1, :rthur could use hapha5ard selection methods 1 nons1stematicall1
selecting voters from a printed cop1 of the electronic data file )for example, 1 picking
some numer of pages and sampling the identified voters+"
8inall1, :rthur could use a lock selection method to select a series of contiguous items
)a series of contiguous voter registration numers, addresses, names, etc"+"
E"(& Sample Selection (Continued)
d" One of the limitations of using hapha5ard selection and lock selection is that the1 are less
likel1 to provide a representative sample compared to unrestricted random selection and
s1stematic random selection" 9ith respect to the use of an1 of these methods, care should
e taken to ensure that :rthur selects a sample of voters representing different genders,
ethnicities, income levels, famil1 si5es, and other factors that ma1 influence their
propensit1 to support or oppose the funding for a sports stadium" One wa1 to do this
would e to sort the electronic data file alphaeticall1 1 last name"
E"(' Sample Selection Methods
a" ;o ensure a representative sample, 1ou should consider factors that ma1 influence
customer satisfaction" ;hese factors include t1pe of service, length of service, and name
of carrier"
" 8our ma4or methods used to select samples are?
#nrestricted random selection )or random selection+ involves numering items in the
population and selecting items for examination ased on random numers picked from a
random numer tale or generated 1 a computer program"
S1stematic random selection )or s1stematic selection+ involves randoml1 selecting a
starting point in a population and 1passing a fixed numer of items etween selections"
=apha5ard selection involves selecting sample items in a nons1stematic manner"
>lock selection involves selecting sample items 1 choosing a series of contiguous )or
ad4acent+ items"
c" 9ith an appropriate sample si5e of !'/ items, 1ou would select the sample as
follows using the aove methods?
#nrestricted random selection? 3umer the customers in the dataase )from ! to !/,'//+,
identif1 !'/ random numers etween ! and !/,'//, and select the corresponding
customer record"
MO6E-$!
Module E - Overview of Sampling

S1stematic random selection? Select a random starting point within the population and
then select ever1 ,/th customer record thereafter )!/,'// customers M sample si5e of !'/
D ,/ items+"
=apha5ard selection? Select !'/ customer records in a nons1stematic manner )for
example, selecting the first !'/ even numered customers in the dataase+"
>lock selection? Select !' customers on ten streets"
E"(' Sample Selection Methods
d" ;his portion illustrates the impact of a randoml1-ordered population when using
various selection methods" 0n this case, the fact that the population is arranged ased on
length of service is prolematic, since one could assume that longer-term suscriers
have a higher level of satisfaction than more recent suscriers" 0n most instances, the use
of unrestricted random selection and hapha5ard selection are not influenced 1 the
ordering of the population"
!" >ecause this population is not randoml1 arranged and cannot e sorted, the use
of s1stematic random selection ma1 e prolematic )since the fixed numer of
items 1passed will e similar in terms of length of service+" 0n addition, since
contiguous items selected under a lock selection method will also e similar in
terms of length of service, this method ma1 result in a nonrepresentative sample
eing selected )for example, all customers could e either long-term suscriers
or recent suscriers+"
$" 6espite the fact that the population is maintained electronicall1, the issues noted
in )a+ aove still exist"
%" >ecause the population can e sorted, 1ou would sort the population to ensure it
is in random order with respect to factors influencing customer satisfaction"
#sing customer name or telephone numer would appear to e most appropriate,
since neither factor appears to e related to satisfaction"
e" !" Nou would not e ale to determine whether 1our sample includes suscriers
residing throughout the suscription area"
$" 9hile the classification of length of service into these three categories will allow
1ou to ascertain that 1our sample includes a variet1 of suscription lengths, it
will not allow 1ou to determine whether 1our sample includes a representative
numer of customers that have suscried for relativel1 long periods of time )for
example, five 1ears or longer+"
%" ;his feature might result in long-term suscriers eing misclassified as recent
suscriers, making it more difficult for 1ou to determine that 1our sample
includes a representative numer of oth t1pes of suscriers"
MO6E-$$
Module E - Overview of Sampling
E"(( Sampling and Nonsampling Risk
a" ;his characteristic ma1 affect sampling risk, since :rthur needs to consider the time of
da1 during which he conducts his sample to ensure that he can include these households"
>ecause man1 dual-income career couples ma1 not have children )or ma1 have a smaller
numer of children+, the failure to include these households ma1 overstate the average
numer of famil1 memers"
" ;his characteristic ma1 affect sampling risk, ecause :rthur would need to select homes
from each of the different price ranges for his sample" 0t is uncertain whether more
expensive homes would e occupied 1 larger families or smaller families, ut the t1pes
of families choosing these homes ma1 differ from those in smaller, newer homes"
c" ;his characteristic ma1 affect sampling risk, ecause homes near the park would e
expected to have a larger numer of famil1 memers" 9hile :rthur would need to
include some of these homes in his sample, if a disproportionate numer of these homes
were included in his sample, the sample estimate ma1 overstate the average numer of
famil1 memers"
d" ;his characteristic ma1 affect sampling risk, ecause homes near the wooded area would
e expected to have a smaller numer of famil1 memers" 9hile :rthur would need to
include some of these homes in his sample, if a disproportionate numer of these homes
were included in his sample, the sample estimate ma1 understate the average numer of
famil1 memers"
e" ;his characteristic ma1 affect nonsampling risk, ecause visitors ma1 e mistakenl1
recorded as famil1 memers"
NOTE TO INSTRUCTOR: !a" % !d" above may also be sub&ect to nonsampling risk if errors in
recording observations !number of family members" or calculations were made #owever, the
characteristics described in !a" % !d" do not by themselves provide exposure to nonsampling risk
E"(, Sampling and Nonsampling Risk
a" !" Sampling risk would e affected if individuals who work out in the morning are
more highl1-motivated and disciplined than those who work out during other
times of the da1" Presumal1, these higher levels of motivation and discipline
would result in a greater weight loss"
3onsampling risk would not e affected 1 this characteristic of the
methodolog1"
$" Sampling risk would e affected if males are more or less likel1 than females to
experience a weight loss as a result of the workout regimen or dietar1
restrictions"
3onsampling risk would not e affected 1 this characteristic of the
methodolog1"
%" Sampling risk would e affected if the lifest1les of these t1pes of individuals
provide them with a greater opportunit1 to exercise or oserve a more restricted
diet, resulting in a greater weight loss"
MO6E-$%
Module E - Overview of Sampling
3onsampling risk would not e affected 1 this characteristic of the
methodolog1"
&" Sampling risk would e affected, since individuals who fre<uentl1 walk to their
offices are more likel1 to e in good ph1sical condition and some of the
oserved weight loss could e the result from this additional ph1sical activit1"
3onsampling risk would not e affected 1 this characteristic of the
methodolog1"
'" 3onsampling risk would e affected if participants did not provide 1ou with
accurate self-reported weights"
Sampling risk would not e affected 1 this characteristic of the methodolog1"
(" 3onsampling risk would e affected if participants did not completel1 oserve the
workout regimen or dietar1 restrictions"
Sampling risk would not e affected 1 this characteristic of the methodolog1"
" Some improvements to the sampling methodolog1 are as follows?
!" 0f a population of individuals etween the ages of %/ to %' from within the
communit1 can e identified and otained, select participants from this listing as
opposed to memers of =ealth >usters" 0f such a listing is not availale, select
memers attending =ealth >usters at various times throughout the da1" ;his
might overcome some of the issues with respect to the composition of 1our
current sample )mostl1 males, single or married with no children, individuals
working out at =ealth >usters earl1 in the morning, etc"+ and provide 1ou with a
more representative sample of individuals aged %/ to %' 1ears" :s a result, this
improvement will reduce 1our exposure to sampling risk"
$" Either oserve workouts and dietar1 restrictions in some manner or have
participants provide 1ou with a log of their workouts and diet during the two-
week period" 9hile using a log introduces issues with respect to self-reporting
of information, it is not feasile for 1ou to oserve all workouts and meals and is
an improvement compared to the current methodolog1" ;his will reduce 1our
exposure to nonsampling risk"
E"(, Sampling and Nonsampling Risk (part b, Continued)
%" Ph1sicall1 weigh the individuals efore and following the two-week workout
regimen and dietar1 restriction" ;his will overcome issues related to self-
reporting of information and reduce 1our exposure to nonsampling risk"
E"(- Sample Selection Methods
a" Precautions that should e considered to enhance the likelihood of a representative
sample include ensuring that the sample includes oservations from different groups of
students, as noted elow )student responses ma1 include other considerations as well+?
MO6E-$&
Module E - Overview of Sampling
Students taught 1 different professors"
Students having different ailities, as measured 1 grade point average and other
characteristics"
Students enrolled in classes at different times of the da1 )morning, afternoon,
and evening classes+"
Students enrolled in different t1pes of usiness courses )accounting, finance,
management, and marketing+"
Students having different personal characteristics )gender, weight, ethnicit1+"
Selecting sample items from the different groups as defined aove will allow 1ou to
ensure that an1 oserved results are ecause of the intended treatment of interest )coffee+
and not other factors" 8or example, if 1ou provide coffee to students in Professor 91nn2s
class and withhold coffee from students in Professor Selleck2s class, 1ou couldn2t
determine if the difference was ecause of the coffee or the teaching methods used 1
Professors 91nn and Selleck"
" )!+ 0dentif1 !// random numers using a random numer tale or computer
program and @matchA the student from the listing that corresponds to the random
numer selected"
)$+ Select a random starting point )sa1 the third student+, which is the first
oservation" Calculate the sampling interval 1 dividing the population si5e
)$,$// students+ 1 the appropriate sample si5e )!// students+, otaining a
sampling interval of $$ students )$,$// !// D $$+" >1pass $$ students to select
the $'
th
student as the second oservation and continue to proceed through the
listing of students until a total of !// have een selected"
)%+ Select !// students in some sort of a nons1stematic 1et noniased wa1" 8or
example, 1ou could pick the first, third, fifth, seventh, and ninth students on
each of the first $/ pages of the listing, for a total of !// students )' students x
$/ pages D !// students+"
)&+ Select one page at random and use that page )which contains !// names+ as 1our
sample"
E"(- Sample Selection Methods (Continued)
c" #nrestricted random selection and s1stematic random selection are advantageous in that
the1 are more likel1 to provide a representative sample and that another individual can
replicate the sample given the criteria of the selection" >oth of these methods have a
disadvantage in that the1 are more time consuming )and, therefore, costl1+ to perform" :n
additional disadvantage of s1stematic random selection is that this method 1passes a
fixed numer of items etween selections" ;his feature could result in a nonrepresentative
sample eing selected if the population is not arranged in a random order"
=apha5ard and lock selection are advantageous in that the1 are less time consuming
)and, therefore, costl1+ to perform" =owever, a primar1 disadvantage of these methods is
that the1 are less likel1 to provide a representative sample" 0n addition, it is t1picall1
more difficult for another individual to replicate the sample given the criteria of the
search"
MO6E-$'
Module E - Overview of Sampling
E"(. actors !""ecting Sample Si#e
:ttriute
Sampling
Bariales
Sampling
Population Si5e 6 6
Expected deviation rateOExpected error 6 6
;olerale deviation rateO;olerale error 0 0
Sampling Risk 0 0
Population Bariailit1 # 6
E",/ $eneral Sampling
a" :lex2s decision to use sampling would e ased on the need for exact information and the
si5e of the population eing considered" 0n this instance, he would e more likel1 to use
sampling if the need for exact information is less important" 8or example, if the store
would e successful if the average numer of children per household is exactl1 !"', he
would e more likel1 to examine the entire population than if less exact information is
re<uired )such as the average numer of children exceeding !"% per household+" :lso,
:lex would e more likel1 to use sampling if the numer of households within a one-
mile radius was larger, since examining the entire population would e less viale in this
situation"
" ;he ma4or advantage of using sampling in this situation is efficienc17 that is, :lex can
attempt to draw a conclusion ased on examining onl1 a suset of the entire population"
;he primar1 disadvantage of using sampling is related to effectiveness" >ecause :lex2s
decision will e ased on onl1 a suset of the items in the population, there is a chance
that sampling will not provide him with the correct answer to his <uestion"
c" Sampling risk is the risk that the decision made ased on the sample is different from the
decision that would have een made if the entire population had een examined"
One outcome that ma1 reflect sampling risk is if :lex2s sample results indicate that the
average numer of children per household is less than !"% when, in fact, it is greater than
!"%" 0n this situation, :lex would decide not to open the ice cream and cand1 shop at that
location" ;he loss to :lex would e the annual income that is foregone from a successful
usiness"
: second outcome that ma1 reflect sampling risk is if :lex2s sample results indicate that
the average numer of children per household is more than !"% when, in fact, it is less
than !"%" 0n this situation, :lex would decide to open the ice cream and cand1 shop at that
location, ut it would e unsuccessful" ;he loss to :lex would e his initial investment in
the unsuccessful usiness"
;here is no one @correctA answer as to which risk is of more concern to :lex" 8or the first
outcome, :lex would fail to reali5e income related to a successful usiness" 0n the second
outcome, :lex would lose the L!'/,/// related to his initial investment" ;he student2s
answer would depend upon their perceptions of the relative effect)s+ of the L!'/,///
initial investment and the L'/,/// annual income on :lex2s net worth" 0f :lex could
easil1 withstand the loss of L!'/,///, the first outcome ma1 e of greater concern7 if not,
the second outcome would e of greater concern"
MO6E-$(
Module E - Overview of Sampling
d" :lex would e far greater concerned with the first outcome, since the annual income
foregone )L'/,/// per 1ear+ is far more sustantial than the initial investment that could
potentiall1 e lost from an unsuccessful venture )L!/,///+"
e" 3onsampling risk is the proailit1 that an incorrect conclusion is reached ecause of
reasons unrelated to the nature of the sample" Potential nonsampling risks that ma1 e
encountered 1 :lex are incorrectl1 recording the numer of children in sample
household)s+, mistaking pla1mates for children residing in sample household)s+, or
making a mathematical error in determining the average numer of children per
household"
E",! !udit Sampling: %&pes o" !udit Samples
a" 'ttribute sampling is a form of sampling used to determine the extent to which some
characteristic )attriute+ exists within a population of interest7 used 1 auditors during
tests of controls"
(ariables sampling is a form of sampling used to examine a population to estimate the
amount or value of some characteristic of that population7 used 1 auditors during their
sustantive procedures"
: dual)purpose test is an audit procedure that can e used as oth a test of controls and a
sustantive test"
b. 1. This is not an audit sample because the results will not be
projected to the population. The procedure is only used to gain
an understanding of the internal controls in operation.
$" ;his is an example of attriute sampling" Nou are not looking to estimate a
dollar amount, ut the rate of error occurrence in internal control"
%" ;his is an example of a dual-purpose test" Nou are testing for internal control
errors, and 1ou will use the sample to pro4ect a dollar amount of an account
alance"
&" ;his is an example of variales sampling" Nou are using the sample to estimate
the monetar1 amount of an account alance"
'" ;his is not an audit sample as 1ou are examining the entire population"
(" ;his is an attriutes sample ecause 1ou are not looking for monetar1 errors"
," Even though 1ou will not in<uire of all personnel, this is not sampling as
considered 1 S'S *+"
MO6E-$,
Module E - Overview of Sampling
E",$ !udit Simulation: 'arious Sampling Concepts
a" ;his statement is not correct >oth statistical and nonstatistical sampling re<uire the
auditor to use professional 4udgment in planning, performing, and evaluating a sample
and in relating the audit evidence produced 1 the sample to other audit evidence when
forming a conclusion aout the related account alance of class of transactions ):#
%'/"/%+"
" ;his statement is not correct Either statistical or nonstatistical sampling methods are
permissile under generall1 accepted auditing standards ):# %'/"/&+" =owever, it is true
that statistical sampling is more time-consuming than nonstatistical sampling"
c" ;his stateement is not correct 3onsampling risk ma1 result in an error in evaluating
sample results, regardless of the representativeness of the sample" 3onsampling risk ma1
result in the failure to reach an appropriate conclusion with respect to the sample ):#
%'/"!!+"
d" ;his statement is correct : lower sampling risk will result in a larger numer of items
eing selected7 that is, sample si5e varies inversel1 with sampling risk ):# %'/"!/+"
e" ;his statement is not correct Sample items should e selected in such a wa1 that the
sample is representative of the population and all items in the population have an e<ual
opportunit1 to e selected ):# %'/"%.+" :s a result, despite the difficult1 in examining
transactions with 9imledon, these transactions cannot e automaticall1 excluded from
the items su4ect to selection"
f" ;his statement is not correct 0f the purpose of the auditor2s application of a procedure to
less than !// percent of the population is something other than evaluating a trait of the
entire account alance or class of transactions, the application is not considered sampling
):# .%'/"/$+" : walk-through that is conducted to understand the nature of transactions
is an example of such an exclusion"
E",% !udit Simulation: $eneral Sampling
;O? Mason P Harr, CP:s
8ROM? Consultant-:dvisor
S#>HEC;? :pplication of audit sampling standards )S'S *+, :# %'/, .%'/+
:t 1our re<uest, 0 have reviewed the audit work in the case files 1ou provided" =erein are m1
conclusions aout proper application of the audit sampling standards in S'S *+ ):# %'/+"
a" 9ork to #nderstand the :ccounting S1stem
;he sample of three purchase orders and suse<uent tracing the cash disursement
documents and procedures is not considered @audit sampling,A and S'S *+ does not
appl1" ;he work was properl1 done for the purpose of otaining a preliminar1
understanding of internal control, not for making a 4udgment aout the effectiveness of
control procedures" :udit sampling standards appl1 to samples taken for the purpose of
reaching a conclusion aout a whole population of dataQin this case the cash
disursements controlsQand not to work done to otain a general understanding of a
client2s internal control" ):# .%'/"/$+
MO6E-$-
Module E - Overview of Sampling
" 0nventor1 Count :ccurac1 ;est
;he sample of inventor1 items for recounting is a sampling application covered 1 S'S
*+" Ms" Harr took the sample for the purpose of making an overall 4udgment of the
accurac1 of the counting procedure" =owever, the sample did not meet S'S *+
re<uirements ecause it does not appear to have een representative ):# %'/"$&+" Onl1
the largest-<uantit1 items were chosen, and the others were ignored" ;hese items were
proal1 the most likel1 to e miscounted"
0n addition to the aove concern, Ms" Harr did not appropriatel1 consider sampling risk in
pro4ecting the results to the population ):# %'/"$(+" Simpl1 stated, the rate of miscount
)- percent+ assumed that the entire population was misstated to the same extent as the
sample )!( $// D - percent+" ;his is not appropriate"
c" Short-;erm 6et Outstanding
;he audit of all of the outstanding commercial paper notes is not a sampling application"
:udit sampling is the application of audit procedures to less than !// percent of the items
in a alance ):# %'/"/!+" Ms" Harr2s work examined all items comprising the short-term
commercial paper amount on the alance sheet"
d" Management Representations
:udit sampling is not involved in the procedure of otaining management
representations" 9hile it ma1 initiall1 appear that Ms" Harr @sampledA client officers to
otain management representations, this evidence is considered to represent the entire
od1 of evidence under examination"
NOTE TO INSTRUCTOR: $he following professional guidance is more closely related
to management representations than audit sampling
Management representations are ordinaril1 otained from those memers of management
with overall responsiilit1 for financial and operating matters, such as the chief executive
officer, chief financial officer, and others with e<uivalent positions whom the auditor
elieves are responsile for and knowledgeale aout, directl1 or through others in the
organi5ation, the matters covered 1 the representations" Such memers of management
normall1 include the chief executive officer and chief financial officer or others with
e<uivalent positions in the entit1" ):# %%%"/.+
E",% !udit Simulation: $eneral Sampling (Continued)
e" Ms" Harr divided the Repairs and Maintenance account into two @populations"A #nder
S'S *+, this is acceptale and, in fact, the recorded or ook value of the items is
specificall1 mentioned as an example of a characteristic that ma1 e used to sudivide a
population ):# %'/"$$+
9hile the process of sudividing the population appears to have een appropriate, Ms"
Harr2s selection of sample items within these groups was not" She audited all the items in
one population )the L$,-,/// expense entries each over L',///+, and this was not a
sampling application" ;he second population )L,',/// for which each entr1 was less than
L',///+ also was not a sampling application7 Ms" Harr applied anal1tical procedures with
respect to these entries ):# .%'/"/$+"
MO6E-$.
Module E - Overview of Sampling
0nstructors should feel free to oserve that the @anal1tical comparisonA is ver1 casual, and
the L,',/// of smaller entries is actuall1 not audited ver1 well" ;he purpose of auditing
the larger entries was to reach a conclusion aout the entire account, even though ver1
little audit effort was given to the population of smaller entries"
E",& (aplan C)! E*am Simulation: Sampling %erminolog&

C Sampling Risk ;he chance that auditorRs conclusion will e wrong ecause
onl1 a portion of the population was examined"
Sampling risk is the chance that auditorRs conclusion will e
wrong ecause onl1 a portion of the population was examined"
6 Hudgment sampling Estimates the amount of sampling risk that the auditor faces
purel1 1 human guess"
Hudgment sampling estimates the amount of sampling risk that
the auditor faces purel1 1 human guess"
S ;olerale misstatement Measures the sufficienc1 of the evidential matter otained"
Statistical sampling measures the sufficienc1 of the evidential
matter otained"
8 :ttriutes sampling Estimates a percentage and is often used in tests of controls"
:ttriutes sampling estimates a rate of occurrence and is often
used in the tests of controls"
0 Bariales sampling Estimates a total and is often used in sustantive tests"
Bariale sampling estimates a total and is often used in
sustantive tests"
E ;olerale deviation rate ;he maximum error rate that the auditor is willing to accept
without modif1ing the planned assessed level of control risk"
;olerale deviation rate is the maximum error rate that the
auditor is willing to accept without modif1ing the planned
assessed level of control risk"
= ;olerale misstatement ;he si5e of the largest misstatement in the account eing
examined that )when comined with misstatements in all other
accounts+ would still not cause the financial statements to e
materiall1 misstated"
;olerale misstatement is the si5e of the largest misstatement
in the account eing examined that )when comined with
misstatements in all other accounts+ would still not cause the
financial statements to e materiall1 misstated"
C 3on-sampling risk ;he chance that auditorRs conclusion will e wrong for reasons
that would happen even if ever1 item had een tested"
0ncludes human errors such as the failure to recogni5e a
misstatement and the misinterpretation of results"
3on-sampling risk is the chance that the auditorRs conclusion
will e wrong for reasons that would happen even if ever1
item had een tested" 0ncludes human error such as the failure
to recogni5e a misstatement and the misinterpretation of
results"
MO6E-%/