CEN/TS 14773:2004

SLOVENSKI STANDARD
01-januar-2005
3RãWQHVWRULWYH±.DNRYRVWVWRULWYH±0HUMHQMHL]JXELQ]QDWQLKXSRþDVQLWHYSUL
SUHGQRVWQLKLQSUYRUD]UHGQLKSRãLOMNDK]XSRUDERSUHVNXVQLKSLVHP
Postal services - Quality of service - Measurement of loss and substantial delay in
priority and first class single piece mail using a survey of test letters
Postalische Dienstleistungen - Dienstqualität - Messung des Verlustes und der
erheblichen Verzögerungen von Vorzugssendungen und Einzelsendungen erster Klasse
unter Verwendung von Testbriefen
Services postaux - Qualité du service - Mesure des pertes et des retards importants pour
le courrier prioritaire égrené et de 1ére classe, a partir d'une enquete utilisant des lettres
témoin
Ta slovenski standard je istoveten z: CEN/TS 14773:2004
ICS:
03.240 Poštne storitve Postal services
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

TECHNICAL SPECIFICATION
CEN/TS 14773
SPÉCIFICATION TECHNIQUE
TECHNISCHE SPEZIFIKATION
September 2004
ICS 03.240
English version
Postal services - Quality of service - Measurement of loss and
substantial delay in priority and first class single piece mail using
a survey of test letters
Services postaux - Qualité du service - Mesurage des Postalische Dienstleistungen - Dienstqualität - Messung
pertes et des retards importants pour le courrier prioritaire des Verlustes und der erheblichen Verzögerungen von
égrené et de 1ére classe, à partir d'une enquête utilisant Vorzugssendungen und Einzelsendungen erster Klasse
des lettres d'essai unter Verwendung von Testbriefen
This Technical Specification (CEN/TS) was approved by CEN on 26 January 2004 for provisional application.
The period of validity of this CEN/TS is limited initially to three years. After two years the members of CEN will be requested to submit their
comments, particularly on the question whether the CEN/TS can be converted into a European Standard.
CEN members are required to announce the existence of this CEN/TS in the same way as for an EN and to make the CEN/TS available
promptly at national level in an appropriate form. It is permissible to keep conflicting national standards in force (in parallel to the CEN/TS)
until the final decision about the possible conversion of the CEN/TS into an EN is reached.
CEN members are the national standards bodies of Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France,
Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia,
Slovenia, Spain, Sweden, Switzerland and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre: rue de Stassart, 36  B-1050 Brussels
© 2004 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TS 14773:2004: E
worldwide for CEN national Members.

Contents Page
Foreword. 4
Introduction . 5
1 Scope. 6
2 Normative references . 6
3 Terms and definitions. 8
4 Loss as a quality of service indicator. 14
4.1 General. 14
4.2 Calculation and presentation of loss and substantial delay . 15
4.2.1 General. 15
4.2.2 Period after which an item is considered to be lost or substantially delayed. 15
4.2.3 Service performance indicator . 15
5 Methodology. 16
5.1 General. 16
5.2 Estimators. 16
5.2.1 General. 16
5.2.2 Accuracy . 16
5.2.3 Measurement results . 17
5.2.4 Calculation of the accuracy . 17
5.2.5 Aggregation of results. 17
5.3 Validation and control of panellist information . 17
5.3.1 General. 17
5.3.2 Panellist requirements. 18
5.3.3 Validation of information. 18
5.4 Treatment of clusters of lost items . 18
6 Test mail characteristics . 19
7 Report. 19
8 Quality control and auditing . 20
Annex A (normative) Calculation of the accuracy of the estimator of the performance
indicator . 21
Annex B (normative) Panellist management and validation . 24
B.1 Management, guidance, training and monitoring of panellists . 24
B.2 Item validation . 24
B.3 Panellists validation. 25
Annex C (normative) Quality control and auditing. 27
C.1 General. 27
C.2 Sending test items . 27
C.3 Receiving test items . 27
Annex D (informative) Use of different rules for defining substantial delay . 28
D.1 General. 28
D.2 Circumstances when additional analysis is appropriate. 28
D.3 Reporting requirements for additional analyses . 29
Annex E (informative) Experience of measuring loss. 30
E.1 General. 30
E.2 Experience in Portugal . 30
E.2.1 General. 30
E.2.2 Brief description of the Portuguese system . 30
E.3 Experience in United Kingdom. 31
E.4 Experience in Germany . 31
E.5 Comments from International Post Corporation . 31
E.5.1 General. 31
E.5.2 Response Rates (Referring to requirements under 5.3) . 32
E.5.3 Clustering (Referring to requirements under 5.4) . 33
E.5.4 Sample sizes (Referring to Annex A). 33
E.5.5 Matching reality. 33
E.6 Comments from France. 33
E.6.1 General. 33
E.6.2 Panel efficiency rate . 34
E.6.3 Confirmation procedures . 34
E.7 Concluding Remarks . 35
Bibliography . 36

Foreword
This document (CEN/TS 14773:2004) has been prepared by Technical Committee CEN/TC 331
"Postal Services", the secretariat of which is held by NEN.
This European Standard shall be given the status of a national standard, either by publication of an
identical text or by endorsement, at the latest by March 2005, and conflicting national standards shall
be withdrawn at the latest by March 2005.
According to the CEN/CENELEC Internal Regulations, the national standards organizations of the
following countries are bound to announce this CEN Technical Specification: Austria, Belgium, Cyprus,
Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland,
Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia,
Slovenia, Spain, Sweden, Switzerland and United Kingdom.
For compatibility between CEN and UPU versions of this document, the term 'document' is used. In a
CEN context, this should be interpreted as being equivalent to the deliverable mentioned on the title
page of this document. In a UPU context, this should be interpreted as being equivalent to 'standard'."

Introduction
In the Green paper on postal services in 1992 the European Commission emphasised the need to
establish common rules for the development of community postal services and the improvement of
quality of service. The Commission identified requirements for quality of service measurement that
included the reliability of services and treatment of problems of loss, theft and damage.
CEN was asked to draw up standards related to the methodology for the measurement of loss of mail
at reasonable cost for some mail types within the universal service.
There has been only limited experience in using sample surveys to measure the level of loss in postal
services and it is not possible to define a standard for measurement at present. However it is possible
to set out the main requirements that would probably be included in such a standard in order that the
methodology may be applied and tested. Once this has been done and experience has been gained
this document may be developed into a standard.
This document sets out a methodology for a more precise measure of the level of loss and substantial
delay of single piece priority letters than can be obtained from the measurement of complaints. This
document refers to a number of principles and minimum requirements to be applied for the
measurement of the level of loss and substantial delay of national and cross-border priority and first
class letter mail.
Annex E of this document describes the experience of measuring loss and contains comments
relating to the implementation of the Technical Specification which have already been reported to
CEN. Since the Technical Specification is to some extent experimental, CEN would particularly like to
collect experience on its use from those who have implemented it. This will be valuable in determining
how the Technical Specification may best be developed into a standard.
1 Scope
This document specifies methods for measuring the level of loss and substantial delay, using a survey
of test letters, of domestic and cross-border priority and first class single piece letter mail, collected,
processed and distributed by postal service operators.
It is impossible in practice to distinguish between items which will never arrive and items which have
been delayed for a very long time. A minimum period is therefore defined after which an item that has
been sent shall be treated as if it has been lost or substantially delayed for the purposes of
measurement.
The resulting overall figure for loss and substantial delay is to be expressed as a percentage of the
total posted priority and first class single piece mail. This indicator does not measure the postal
operator’s overall performance in a way which provides direct comparison of postal service operators,
and does not include other service performance indicators other than those related to loss and
substantial delay.
According to this document, loss and substantial delay is estimated within specified accuracy limits.
When the level of loss is very small it may only be possible to estimate an upper limit for the level of
loss.
This document has been developed from and is compatible with the requirements of EN 13850 for the
measurement of the transit time of end-to-end services for single piece priority mail and first class mail,
so that the same survey may be used to measure loss and substantial delay and on-time performance.
Thus it comprises a set of requirements for the design of a measuring system involving the selection
and distribution of test letters sent and received by selected panellists, according to a specification
that ensures that the test letter sample design is representative of the real mail flows.
This document relates to the measurement of the so called “normal” services given to private persons
/ households and businesses that post mail at street letter boxes, over the counter at post offices,
have pick ups at their offices or give their mail directly at postal service operators’ sorting centres.
For technical reasons this document may not in all parts be suitable for the measuring of very small
volumes of mail and for operators with limited coverage.
This document is not applicable for measuring the level of loss or substantial delay for large bulk
mailers’ services and hybrid mail which require different measurement systems and methodologies.
This document includes specifications for the quality control and auditing of the measurement system
and for the reporting of loss and substantial delay of mail.
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
EN 13850:2002, Postal services - Quality of service - Measurement of the transit time of end-to-end
services for single piece priority mail and first class mail
EN 14012, Postal services - Quality of service - Measurement of complaints and redress procedures.
EN ISO 9000, Quality management systems — Fundamentals and vocabulary (ISO 9000:2000).
ISO 3534-1:1993, Statistics — Vocabulary and symbols — Part 1: Probability and general statistical
terms.
ISO 3534-2:1993, Statistics — Vocabulary and symbols — Part 2: Statistical quality control.
ISO 19011, Guidelines for quality and/or environmental management systems auditing
International Chamber of Commerce / ESOMAR International Code of Marketing and Social Research
1)
Practice, with attachments.
1)
This document can be obtained from www.esomar.org and from www.iccwbo.org . ESOMAR Central
Secretariat (email@esomar.nl)
3 Terms and definitions
For the purposes of this document the following terms and definitions apply.
3.1
accuracy
closeness of agreement between a test result and the accepted reference value
NOTE 1 The term accuracy, when applied to a set of test results, involves a combination of random
components and a common systematic error or bias component [ISO 3534-1:1993].
NOTE 2 In this document the accuracy is given by the length of the confidence interval at the confidence level
95 % for the parameter being estimated, namely the probability of being lost or substantially delayed
3.2
aggregation
compounding of primary data into an aggregate for the purpose of expressing them in a summary
form
3.3
audit
systematic and independent examination to determine whether activities and related results comply
with planned arrangements and whether these arrangements are implemented effectively and are
suitable to achieve objectives [EN ISO 9000]
3.4
average (arithmetic mean)
sum of values divided by the number of values [ISO 3534-1:1993]
3.5
business
bodies, other than households, that receive and send mail
3.6
characteristic
property which helps to identify or differentiate between items of a given population
NOTE 1 The characteristics may be either quantitative - by variables, or qualitative - by attributes
[ISO 3534-1:1993]
NOTE 2 In this document the population is postal test items and the characteristics are related to type of
senders, type of receivers, times and types of induction or delivery, physical aspects of test items, franking, etc.
3.7
conformity
fulfilment of specified requirements [EN ISO 9000]
3.8
corrective action
action taken to eliminate the causes of an existing non-conformity, defect or other undesirable
situation in order to prevent recurrence [EN ISO 9000]
3.9
correctly addressed letter
letter with an address that conforms to a recognized international or national standard on the postal
item concerned
NOTE The basic elements of the address are; identification information, delivery address and post code,
and postal address of the addressee, and, in the case of cross-border mail, also the name of the country of
destination.
3.10
correlation
relationship between two or several random variables within a distribution of two or more random
variables
NOTE Most statistical measures of correlation measure only the degree of linear relationship.
[ISO 3534-1:1993]
3.11
correlation coefficient
ratio of the covariance of two random variables to the product of their standard deviations
[ISO 3534-1:1993]
3.12
country
territory of a nation with its own government
3.13
cross-border mail
mail distributed from one country to another country
3.14
customer
natural or legal person having a business relation with a postal operator
3.15
date of delivery
date on which a postal item is delivered to the address or the addressee
3.16
date of deposit
J
day on which a postal item is posted, provided posting takes place before the last collection of that
day
NOTE 1 The term date of deposit is used in relation to quality of service measurement. The definition is based
on the Postal Directive 97/67/EC.
NOTE 2 Last collection refers to the advertised last time for collection (not the actual time).
3.17
delivery point
physical location at which delivery of postal items delivered by a postal operator takes place and
where they leave the operator’s responsibility
3.18
design factor
ratio of the variance of the estimator in the given sample design by the variance of the estimator in an
elementary sample design of the same size
NOTE The design factor is related to a given sample design and estimator.

3.19
discriminant (characteristic)
characteristic affecting the outcome
NOTE In this document a characteristic is discriminant when transit time differs according to the different
modes of the characteristic.
3.20
distribution
process from collecting mail at collecting points through sorting at the distribution centre to delivery of
mail items to the addressee
3.21
domestic mail
mail items sent and received within one country
3.22
end-to-end (transit time)
from the point mail is placed into the collection/acceptance system under the responsibility of the
collecting postal operator to the final delivery point under the responsibility of the delivering postal
operators
3.23
estimate
value of an estimator obtained as a result of an estimation [ISO 3534-1:1993]
3.24
estimator
statistic used to estimate a population parameter [ISO 3534-1:1993]
NOTE In this document, a function of the observed values of test item transit times allowing the estimation
of the probability of the item being lost or substantially delayed.
3.25
expectation
for a discrete random variable X taking the values x , with the probabilities p the expectation if it
i i,
exists, is:
µ==EXpΣx
()
ii
the sum being extended over all the values x which can be taken by X [ISO 3534-1:1993]
i
3.26
field of study
total mail flow between defined postal areas
NOTE 1 Field of study could be defined for example as:
- Domestic — one operator in one country
- Domestic — one operator in one part of a country
- Domestic — a group of operators in one country
- Cross-border — one operator to one country
- Cross-border — one operator to a group of countries
- Cross-border — one country to one country
- Cross-border — one country to a group of countries
NOTE 2 Some mail flows between postal operators may not meet the technical requirements in this document
to qualify as fields of study
3.27
geographical coverage
spread of postal services within a pre-defined geographical area
3.28
independent performance monitoring organisation
body charged with the monitoring of the quality of service according to the methodology specified in
this document and which body is external to, and having no links of ownership or control with the
postal service provider thus monitored
3.29
independence
two random variables are independent if their distribution functions are related by
F(x, y) = F(x, ∞) × F(∞,y) = G(x) × H(y)
where F(x, ∞) = G(x) and F(∞ , y) =H(y) are the marginal distribution functions of X and Y, respectively
for all pairs (x, y) [ISO 3534-1]
3.30
induction
deposition of mail into the postal network
3.31
inspection
activity such as measuring, examining, testing or gauging one or more characteristics of an entity and
comparing the results with specified requirements in order to establish whether conformity is achieved
for each characteristic [EN ISO 9000 and ISO 3534-2:1993]
3.32
item arriving late
postal item delivered to the recipient in a time longer than the service provider’s promised service
standard but within a specified period of time [EN 14012]
NOTE The specified period is defined in 4.2.2
cf. item lost or substantially delayed (3.33)
3.33
item lost or substantially delayed
postal item that is not delivered to the recipient or is only delivered after substantial delay
NOTE For the purposes of this document, lost items include items with substantial delay.
cf. item arriving late (3.32), substantial delay (3.57)
3.34
observed value
value of a characteristic obtained as the result of a single observation [ISO 3531-1:1993]
3.35
on-time performance
proportion of postal items within a given period of time t with transit times meeting the specification
Nt()
Formal definition: X where N()t is the total number of postal items inducted during
i
∑
Nt()
i=1
the period, and X =1 if postal items i fulfils the specification and otherwise X = 0
i i
3.36
on-time probability
probability of the event that the transit time of a mail item meets the specification
EXAMPLE If the transit time of a mail item is T and the specification is that the transit time does not exceed s
days, then the on-time probability is Pr ( T ≤ s).
3.37
performance indicator
expression derived from postal transit time statistics and data used to characterise the performance of
a postal operator
3.38
postal service
services involving the clearance, sorting, transport and delivery of postal items
3.39
priority item
first class item
A-class item
letter post item sent with priority by air or by surface and receiving different processing compared to
items classified as non-priority
3.40
probability distribution
function giving the probability that a random variable takes any given value or belongs to a given set
of values [ISO 3534-1:1993]
3.41
quality
totality of characteristics of an entity that bear on its ability to satisfy stated and implied needs
[EN ISO 9000]
3.42
quality assurance
all the planned and systematic activities implemented within the quality system and demonstrated as
needed, to provide adequate confidence that an entity will fulfil requirements for quality [EN ISO 9000]
3.43
quality control
operational techniques and activities that are used to fulfil requirements for quality [EN ISO 9000 and
ISO 3534-2:1993]
3.44
quality evaluation
systematic examination of the extent to which an entity is capable of fulfilling specified requirements
[EN ISO 9000]
3.45
random sample
sample of n items taken from a population of N items in such a way that each item in the population
has a strictly positive probability of being selected
3.46
random variable
variable that may take any of the values of a specific set of values and which is associated with a
probability distribution [ISO 3534-1:1993]
3.47
real mail
postal items sent by customers
3.48
real mail flow
number and types of postal items distributed end-to-end by postal operators
3.49
real mail study
studies on real mail flows or real mail characteristics
3.50
sample
one or more sampling units taken from a population and intended to provide information on the
population [ISO 3534-1:1993]
3.51
service standard
standard that specifies requirements to be fulfilled by a service, to establish its fitness for purpose
3.52
single piece mail
postal items posted and distributed individually
3.53
sorting centre
place where the main sorting of mail is done
3.54
standard
document, established by consensus and approved by a recognised body, that provides, for common
and repeated use, rules, guidelines or characteristics for activities or their results, aimed at the
achievement of the optimum degree of order in a given context
NOTE Standards should be based on the consolidated results of science, technology and experience, and
aimed at the promotion of optimum community benefits [EN 45020].
3.55
stratification
division of a population into mutually exclusive and exhaustive subpopulations (called strata) which
are thought to be more homogeneous with respect to the characteristics investigated than the total
population [ISO 3534-1:1993]
3.56
study domain
subpopulations for which separate results can be appropriate
NOTE Study domains could be defined for example by geographic segmentation or by product.
3.57
substantial delay
postal item delivered to the recipient with a transit time greater than a specified period of time
NOTE The specified period is defined in 4.2.2
3.58
test item
postal item produced in the test measurement system for the purpose of measuring real mail
performances
3.59
test period
period of time under which measurement has been carried out and for which the results are presented
in a separate test report
3.60
transit time
number of days elapsed between induction and delivery of mail item
3.61
upper limit
highest value within a confidence interval for an estimate of a parameter
3.62
variance
measure of dispersion, which is the sum of the squared deviations of observations from their average
divided by one less than the number of observations [ISO 3534-1:1993]
3.63
weight
value to compensate for the difference between the distribution of characteristics in the measurement
system compared to real mail
4 Loss as a quality of service indicator
4.1 General
The system for measuring the loss and substantial delay of mail items shall be robust and shall give
statistical measures within estimated levels of accuracy.
The system for measurement shall comply with the requirements of EN 13850 for the measurement of
the transit time of end-to-end services for single piece priority mail and first class mail. In particular the
following requirements of EN 13850 shall apply, specifically:
 sample design shall be representative of real single piece mail flows;
 measuring system shall provide one annual figure for each relevant field of study;
 measuring methodology shall be objective and shall be audited;
 all measurements of transit time shall refer to the end-to-end transit time;
 measurement system shall be continuous;
 field of study shall be defined and used consistently throughout the measurement.
4.2 Calculation and presentation of loss and substantial delay
4.2.1 General
The following subjects are related to the definition, calculation and presentation of the performance
indicator.
4.2.2 Period after which an item is considered to be lost or substantially delayed
The transit time of a postal item shall be measured in units of days, expressed as (J + n) days, and
shall be calculated in accordance with the requirements of EN 13850:2002, Clause 4.2.
NOTE This allows the possibility of variations in the definitions used for domestic mail.
For the purposes of measurement, a test item shall be classified as lost or substantially delayed if the
item has not been received by the recipient with a transit time that is less than or equal to the number
of days shown in Table 1.
Table 1 — Definition of substantial delay
Service Standard
J + 1 J + 2 J + 3 Over J + 3
or no standard
Domestic mail 7 13 20 27
Cross-border mail 10 20 30 40
As the figures in Table 1 show, the minimum time required is related to the service standard, e.g.
J + 1 for domestic first class mail. The column headed "Over J + 3" should be used for postal items for
which no service standard has been defined. Exceptionally, for cross-border mail on routes where
delays of 40 days or more are not unusual a longer period than 40 days may be used to define
substantial delay.
Test items with imprecise transit times shall always be excluded from the calculations.
The survey operator shall specify and report the transit time after which a domestic or cross-border
mail item has been classified as lost or subject to substantial delay.
Loss/substantial delay may also be estimated using a different rule from that set out in Table 1 if this
is considered appropriate.
NOTE Annex D gives guidance on how the results may be reported and the circumstances under which this
additional analysis might be appropriate.
4.2.3 Service performance indicator
The indicator used in the presentation of the service performance result shall be the percentage of
test items not delivered within the defined period. The result shall be presented as the percentage of
test items not having arrived by J + n, whereby J represents the day of deposit and n the number of
qualified days for the service standard.
NOTE If the level of loss and substantial delay is very small it is sufficient to report an upper limit on the level
of loss (cf 5.2).
5 Methodology
5.1 General
The methodology shall be based on test mail representative of the real mail and shall provide
indicators that are representative of the real mail levels of loss and substantial delay in the period
measured for correctly addressed postal items which do not appear to contain valuables. As required
by EN 13850:
 The test mail method shall consist of a process in which panellists acting as senders post test
items into the postal operator’s mail network and register date and time of posting, to the
destination of panellists acting as receivers who register the day of delivery.
 The sample design shall be representative of the postal flows and of the discriminant
characteristics of the real mail.
 Real mail studies for domestic and cross-border mail measurement shall be implemented prior to
the set up of the test mail measurement system in order to obtain the information needed for the
sample design.
 Geographical stratification shall be the basis of the test mail sample design.
 Senders and receivers shall be spread all over the field of study in order to fulfil the specifications
of stratification and geographical coverage. The distribution of the panel shall be done according
to random sampling.
 The performance monitoring shall be carried out by an independent performance monitoring
organisation.
5.2 Estimators
5.2.1 General
The probability that a letter will be lost or substantially delayed shall be estimated based on a sample
and an estimator. According to the two different types of measurement results, two different
estimators are necessary. Let:
ˆ
 P be the estimator for p the true probability of loss or substantial delay for domestic mail.
ˆ
 be the estimator for p the true probability of loss or substantial delay for cross-border mail.
P 2
5.2.2 Accuracy
ˆ
For the estimators Pi,1=,2 the accuracy per year is given by the maximum absolute difference
i
ˆ
ε between P and the respective upper and lower confidence limits of the confidence interval for
i
i
p for a given confidence level 1−α .
()
i i
ˆ
P shall be given explicitly provided that the upper and lower limits of its confidence interval satisfy
i
the following conditions:
 for a domestic field of study, the confidence level is set to be 19−=α 5%, and the maximum
ˆ
absolute deviation ε is less than or equal to 0, 4P
1 1
 for a country-to-country field of study, the confidence level is set to be 19−=α 5%, and the
ˆ
maximum absolute deviation ε is less than or equal to 0,8 P
2 2
ˆ
If P cannot be estimated with these levels of accuracy because the level of loss and substantial delay
i
ˆ
or the sample size is too small then P shall not be given explicitly but instead an upper limit for p
i
i
shall be stated. The confidence level for this upper limit is also set to be 19−=α 5%
NOTE The upper limit is given by the length ε of the confidence interval [0, ε] for p at the confidence level
i
95 %.
Example 1: Suppose that the level of loss and substantial delay for a domestic field of study is
0,005 (or 0,5 %). The maximum absolute deviation should be no greater than 0,002 giving a
maximum confidence interval of [0,003. 0,007].
Example 2: Suppose that the level of loss and substantial delay for a cross border field of study is
0,01 (or 1 %). The maximum absolute deviation should be no greater than 0,008 giving a maximum
confidence interval of [0,002. 0,018].
Annex A describes how the accuracy limits of the estimator of the probability of loss and substantial
delay shall be calculated, including the situation where only an upper limit is calculated.
5.2.3 Measurement results
The domestic measurement system shall provide one annual result or an upper limit for the overall
level of loss and substantial delay for the field of study.
The cross-border measurement system between two countries shall provide one annual result or an
upper limit for the overall level of loss and substantial delay for the field of study.
5.2.4 Calculation of the accuracy
The accuracy of the probability that a letter will be lost or substantially delayed shall be assessed by
calculating the variance of the estimator and the design factor in accordance with EN 13850.
5.2.5 Aggregation of results
The strata weights shall be determined based on real mail studies and shall contribute to the overall
accuracy.
The correct proportions of test items for the geographical strata or for the highly discriminant mail
characteristics may not be achieved, due to cases of non-response or invalid test items. Weighting
shall be implemented to correct these according to the procedures of EN 13850.
5.3 Validation and control of panellist information
5.3.1 General
Measurement errors shall be identified and assessed, and action taken to reduce them to a minimum
level.
5.3.2 Panellist requirements
The panel of senders and receivers shall be independent of postal service operators and shall be
managed according to the International Chamber of Commerce/ESOMAR International Code of
Marketing and Social Research Practice including its attachment.
After selection, the qualified panellists shall sign a confirmation of their participation in and approval of
general rules for conducting the survey. To ensure proper and correct conduct of actions agreed,
individual or group training shall be arranged, for example, by telephone or face-to-face, and written
instructions shall be distributed to the panellists.
Panellists shall be encouraged to report honestly.
EXAMPLE Panellists who do not post letters on the day specified or do not report letters promptly, should not
be penalised in a way that could encourage them to lie about what actually happened.
The exact location of the active panel shall remain unknown to the postal operator. If panellists
become known they shall be excluded.
5.3.3 Validation of information
Before a letter is recorded as lost or substantially delayed, a definite statement regarding the exact
date on which the letter was sent, and a definite statement that the letter had not been received or
had been received on a date which constituted substantial delay, shall be verified. The sending
panellist shall be contacted and asked to confirm that all the test letters he or she was supposed to
send recently have actually been sent. The receiving panellist shall be contacted and asked to
confirm that all test items received recently have been handled correctly and that the panellist has not
been away, changed their address, and so on. A test item shall only be counted as lost or
substantially delayed if these records have been confirmed with both panellists. Where an item
appears to be lost the time between the dispatch of a given test item and the initial contact for
evaluation of the work of the panellists with regard to that test item shall not exceed 14 days.
All items removed from the survey shall be recorded and the reason for the removal shall be
documented and made available for audit.
The operator shall ensure that the removal of test items from the survey does not bias the survey
results. Thus if there are doubts about whether a sender has correctly recorded some dates of
induction in a particular period, all the items sent by that panellist in the same period shall also be
removed from the survey. For example, if there is evidence that a receiver was probably not able or
willing to confirm some dates of delivery in a particular period, all the items sent to that panellist in the
same period shall also be removed from the survey.
EXAMPLE A panellist has been working in the survey for a year and has previously been satisfactory. Two
items sent to the panellist recently have not been received. The panellist has been contacted and has stated that
he has been in hospital for the last two months and members of his family have been returning the test items;
some items may have been overlooked. In the last two months 11 items were sent to the panellist of which 9
were reported as received. All 11 items should be removed from the survey.
NOTE Some items, such as lost items for example, might be eligible for the measurement of loss and
substantial delay which cannot be used in the transit time measurement. Conversely there are usually items
eligible for the transit time measurement which must be excluded from the measurement of loss and substantial
delay, for example from panellists who would be unable to confirm a potential non-delivery.
5.4 Treatment of clusters of lost items
A
...