SIST EN 15707:2009

2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.Raziskave tiskanih medijev - Slovar in zahteve za storitvePrintmedienanalysen - Begriffe und DienstleistungsanforderungenMesure d'audience de la presse écrite - Vocabulaire et exigences de servicePrint media surveys - Vocabulary and service requirements03.080.20Storitve za podjetjaServices for companies01.040.03Storitve. Organizacija podjetja, vodenje in kakovost. Uprava. Transport. Sociologija. (Slovarji)Services. Company organization, management and quality. Administration. Transport. Sociology. (Vocabularies)ICS:Ta slovenski standard je istoveten z:EN 15707:2008SIST EN 15707:2009en01-januar-2009SIST EN 15707:2009SLOVENSKI
STANDARD



SIST EN 15707:2009



EUROPEAN STANDARDNORME EUROPÉENNEEUROPÄISCHE NORMEN 15707November 2008ICS 01.040.03; 03.080.20 English VersionPrint media surveys - Vocabulary and service requirementsMesure d'audience de la presse écrite - Vocabulaire etexigences de servicePrintmedienanalysen - Begriffe undDienstleistungsanforderungenThis European Standard was approved by CEN on 25 September 2008.CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this EuropeanStandard the status of a national standard without any alteration. Up-to-date lists and bibliographical references concerning such nationalstandards may be obtained on application to the CEN Management Centre or to any CEN member.This European Standard exists in three official versions (English, French, German). A version in any other language made by translationunder the responsibility of a CEN member into its own language and notified to the CEN Management Centre has the same status as theofficial versions.CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland,France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal,Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.EUROPEAN COMMITTEE FOR STANDARDIZATIONCOMITÉ EUROPÉEN DE NORMALISATIONEUROPÄISCHES KOMITEE FÜR NORMUNGManagement Centre: rue de Stassart, 36
B-1050 Brussels© 2008 CENAll rights of exploitation in any form and by any means reservedworldwide for CEN national Members.Ref. No. EN 15707:2008: ESIST EN 15707:2009



EN 15707:2008 (E) 2 Contents Page Foreword.3 1 Scope.4 2 Normative references.4 3 Terms and definitions.4 4 Requirements.6 4.1 Cooperation with clients.6 4.1.1 Client briefing for research service provider.6 4.1.2 Changes in the specification of work.6 4.1.3 Use of subcontractors.6 4.1.4 Client review of questionnaire.6 4.1.5 Checking of data quality.6 4.1.6 Code frame and data analysis.7 4.1.7 Presentation of results to the client.7 4.1.8 Copyrights.7 4.1.9 Declaration of own interests and/or relevant parallel work.7 4.2 Privacy.7 4.2.1 General.7 4.2.2 Privacy rights of the respondents.7 4.2.3 Privacy rights of the interviewers.7 4.2.4 Privacy rights of the clients.8 4.3 Universe and sample.8 4.3.1 Definition and description of the universe.8 4.3.2 Samples.9 4.4 Measurement procedures.14 4.4.1 Types of surveys.14 4.4.2 Audience measurement surveys.14 4.4.3 Media profile surveys.16 4.4.4 Measurement of page exposure.16 4.4.5 Calculation of reading probabilities and methods of accumulation.16 4.5 Fieldwork management.17 4.5.1 General.17 4.5.2 Interviewer recruitment.17 4.5.3 Interviewer training.17 4.5.4 Interviewer validation.19 4.5.5 Subcontracting of fieldwork.19 4.6 Documentation/questionnaire.19 4.6.1 General.19 4.6.2 List of specific details to be documented.20
SIST EN 15707:2009



EN 15707:2008 (E) 3 Foreword This document (EN 15707:2008) has been prepared by Technical Committee CEN/TC 376 “Project Committee - Print media analyses services”, the secretariat of which is held by DIN. 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 May 2009, and conflicting national standards shall be withdrawn at the latest by May 2009. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. CEN [and/or CENELEC] shall not be held responsible for identifying any or all such patent rights. According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following countries are bound to implement this European Standard: Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and the United Kingdom.
SIST EN 15707:2009



EN 15707:2008 (E) 4 1 Scope This European Standard specifies the vocabulary and service requirements for media surveys in the field of print media. 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. ISO 20252:2006, Market, opinion and social research — Vocabulary and service requirements 3 Terms and definitions For the purposes of this document, the following terms and definitions apply. 3.1 client individual, organization, department or division, internal or external, which requests or commissions a research project [ISO 20252:2006, 2.8] 3.2 code numeric or alpha character or combination of characters associated with each response category [ISO 20252:2006, 2.9] 3.3 code frame list of categories with associated codes for classifying responses [ISO 20252:2006, 2.10] 3.4 first reading yesterday (FRY) variation of the “Recent Reading” model in which for all titles being read a question follows if a particular copy (issue) has been read yesterday for the first time 3.5 first reading in publishing interval (FRIPI) variation of the FRY-model in which for all titles being read a question follows if the particular titles has been read within the last publication interval for the first time 3.6 frequency frequency of being exposed to a medium within a specific time period 3.7 interview activity of collecting information from respondents SIST EN 15707:2009



EN 15707:2008 (E) 5 3.8 print medium newspaper or magazine either sold or freely distributed 3.9 questionnaire structured tool for collecting data consisting of a series of questions [ISO 20252:2006, 2.44] 3.10 quota sampling method of constructing a sample so that it conforms to a predefined structure with respect to certain variables [ISO 20252:2006, 2.45] 3.11 random sampling procedure such that each member of the sampling frame has a specific positive chance of being included in the sample [ISO 20252:2006, 2.40] 3.12 reach share of target persons within a universe exposed to a medium within a specific time period 3.13 readership projection projection in absolute figures of the share of target persons exposed to a medium within a specific time period 3.14 readership profile structure of the readership of a particular title 3.15 recent reading reading within a specific time period 3.16 representativeness extent to which a sample represents the defined target population [ISO 20252:2006, 2.47] 3.17 research service provider individual, organization, department or division, internal or external, which conducts research projects or parts of them 3.18 respondent person from whom data are collected 3.19 sample subset of the target population from which data are to be collected [ISO 20252:2006, 2.52] SIST EN 15707:2009



EN 15707:2008 (E) 6 3.20 sampling frame list of population elements or other appropriate sources from which sample members are selected [ISO 20252:2006, 2.53] 3.21 subcontractor individual, organization, department or division, internal or external, which takes responsibility for executing an element of the research project on behalf of the research service provider 3.22 target population population of interest to which inferences are to be made [ISO 20252:2006, 2.60] 3.23 universe synonym for target population 4 Requirements 4.1 Cooperation with clients 4.1.1 Client briefing for research service provider If the objectives of the research projects and/or other requirements do not appear to be described fully enough in the client’s documents, the research service provider shall request a more detailed briefing from the client. Any briefing shall be documented. 4.1.2 Changes in the specification of work The research service provider shall be responsible for agreeing and confirming in writing with the client any changes that will affect the project or client requirements, which may include the cost, timing and any other agreements stated in the research proposal, cost quotation or contract. 4.1.3 Use of subcontractors The research service provider shall inform the client which subcontractors, if any, are to be used, for any part of the research process, before these subcontractors undertake the work. Any subsequent changes in subcontractors to be used shall also be communicated to the client. 4.1.4 Client review of questionnaire The research service provider shall give the client the opportunity to participate in the design and/or the review of the questionnaire. The research service provider shall obtain the approval of the questionnaire unless it is waived by the client. This approval shall be documented. 4.1.5 Checking of data quality The research service provider shall allow the client to check the quality of data collection as well as data editing. Clients shall be informed if their checking could affect the quality of the data collected. The research service provider shall take adequate steps to ensure that the identity of respondents and interviewers is protected and that clients agree to respect this. SIST EN 15707:2009



EN 15707:2008 (E) 7 4.1.6 Code frame and data analysis The research service provider shall make available to the client the code frame and the methods used to analyse the data, if requested by the client. 4.1.7 Presentation of results to the client The research service provider shall agree with the client the way the research results are presented. 4.1.8 Copyrights It shall be discussed in advance and in writing with each client individually the rights of the research service provider and the client concerning the method, questionnaire, materials and results. The General Business Terms of the research service provider shall not contain anything contradictory to this requirement. 4.1.9 Declaration of own interests and/or relevant parallel work The research service provider should inform the client if he is already working for a direct competitor of the client or if he has own interests in the field he tenders for or if he is already working on a similar research project for another client or self. 4.2 Privacy 4.2.1 General The research service provider shall observe the privacy of respondents, interviewers and clients according to applicable European and national laws and relevant principles and rules of professional conduct. 4.2.2 Privacy rights of the respondents When asking for participation in a research project it shall be obvious for the respondents that their participation is completely voluntary and that they can break off the interview at any stage. The respondents shall be told the general purpose of the research project and should not be misled about the presumed duration of the interview and the kind of questions to be asked. If any follow-up or repeated interviews are intended, the respondents shall be asked for permission to re-contact them at the end of the first interview at the latest. This permission shall include consent to the necessary storage of address details. The research data and the address data shall be separated from each other as soon as possible and stored accordingly. The list of keys that link research and address data shall be stored separately. The address data shall be destroyed or deleted at the earliest possible moment. 4.2.3 Privacy rights of the interviewers The research service provider shall provide the interviewers in an appropriate way with the general information that their work will be reviewed. The research service provider shall inform the interviewer about any discrepancies or problems identified by the review and shall give him the opportunity to comment on them, before any action is taken. SIST EN 15707:2009



EN 15707:2008 (E) 8 4.2.4 Privacy rights of the clients The identity of the client shall not be revealed to the interviewers and respondents unless the client requests for this to be revealed. If this might cause methodological problems the research service provider shall inform the client accordingly. The research service provider shall not reveal the identity of a client to any third party including other clients without prior permission. The research service provider shall not reveal documents, materials, products, records etc. which have been provided to him by the client or have been prepared by him to conduct a research project to any third party including other clients without prior permission. 4.3 Universe and sample 4.3.1 Definition and description of the universe The universe shall be limited regionally and temporally as well as comprehensibly defined. Possible universes can be:  physical persons;  private households;  institutions/companies;  recipients of specific media. The universe description shall be extensive. The description shall be objective as well as regionally and temporally defined. The following specifications shall be included:  verifiable characteristics which define the members of the universe;  geographic area of the survey;  date of establishment of the reference data base for addresses and projection. In the case of audience measurement surveys the universe shall be defined independently from the media whereas in the case of media profile surveys the universe shall include the recipients of specific media. The following preconditions are essential to build and describe the universe:  inclusion of a comprehensible definition and its absolute size (projection);  most recent available and reliable data shall be used (including source and date);  characteristics and data used to define the universe and its projection shall be described in detail. SIST EN 15707:2009



EN 15707:2008 (E) 9 4.3.2 Samples 4.3.2.1 Sampling procedures: probability vs. non-probability samples 4.3.2.1.1 General A sample shall be representative either for the defined universe or for a relevant subset of it in order to draw a conclusion from the sample to the universe. The size of the sample shall be chosen in such a way that the sampling variation achieved both in the overall sample and in the subgroups relevant to the analysis is acceptable for the purposes of the research project. Not only the distribution of socio-demographic variables should be evaluated but also other variables which may influence the results. NOTE Representativeness is only possible when special methods are applied, e.g. strict probability selection or strict adherence of specified quotas. 4.3.2.1.2 Stratification The quality of random samples and quota methods can be improved by stratification. NOTE 1 These samples are distinguished by procedures in which the universe is divided into disjunctive subsets and a sample of elements is drawn independently from each subset. The sample quality is improved to the degree that the characteristics of stratification correlates with the survey subject. In order to make a statement about the total sample all under claimed and over claimed groups shall be balanced according to the distribution within the universe from which the sample was drawn. A disproportionate selection for different groups should enable the realisation of a specific case size that is independent of the real ratio within the total universe. NOTE 2 If the selection leads to regional and/or temporal sample clusters this clustering enlarges the sampling error. 4.3.2.1.3 Characteristics of random/quota samples In case of random sampling each element of the universe shall have a calculable non-zero chance of being included in the sample and without selection bias on the determination of the sampling unit (e.g. through the interviewer). NOTE 1 This probability theory model allows the calculation of sampling errors. In the case of quota sampling the elements of the universe shall be selected in such a way that the proportion of sample elements are reflecting the proportion of these elements in the universe. NOTE 2 The theoretical basis to calculate sampling errors is missing. NOTE 3 The differentiation between random and quota is also valid for the selection and recruiting of panels. 4.3.2.1.4 Methods of random sampling 4.3.2.1.4.1 General Basis for a random sample can be registers, regional units (area sampling), telephone number circles, public places, event places or entrances (for certain time slots if necessary) or measured operations. In order to consider the selection of survey units in a random selection all units shall have the same non-zero or calculable chance to be part of the sample. This shall be true for the respondents or operations out of these units that are to be selected. SIST EN 15707:2009



EN 15707:2008 (E) 10 Each stage of the selection process shall be documented exactly. This shall include which and how many units where selected with which method and which units were interviewed or analyzed. 4.3.2.1.4.2 File sampling/register sampling Prerequisite for the usage of a file as base for the selection of a sample should be the collection of all elements in the universe. If the file does not cover all elements of the universe there should be a qualitative and quantitative description of all subsets within the universe which are insufficiently represented. The source file should include at least 85 % of all elements of the universe. These elements can either be the survey units themselves or units from the last selection level (e.g. household or company). The sample used for reporting the results should have the same structure as the source file in order to make a statement about the universe. This should be documented on the basis of familiar characteristics in the universe. If there is a bias in the file for the sample building, then a reliable representation of the sample shall be assured before it is selected. 4.3.2.1.4.3 Area sampling The base material for area sampling shall be regionally defined units that distinctively cover all the inhabited areas of the universe. Each area unit shall have a calculable non-zero chance to be chosen. 4.3.2.1.4.4 Telephone sampling The basis for the selection process and sampling procedure shall be files with telephone numbers deriving either from telephone directories or from RDD (Random-Digit-Dialing). If a statement is made for all elements in the universe then the 85 % coverage rule shall be valid (see 4.3.2.1.4.2). In the case of RDD sampling it shall be assured that the selection chance for registered numbers and not-registered numbers is calculable. Otherwise the RDD sample can not be considered representative. If a regional stratification is not possible then the method (screening etc.) of attaining a representative regional basis shall be documented in detail. 4.3.2.1.4.5 Other random sampling NOTE For many surveys other random methods are possible, this includes the possibility of interviewing pedestrians or visitors. Regardless of random method all persons belonging to the defined universe shall have a known and calculable non-zero chance to be chosen. The universe definition and selection procedure shall be documented. All other selection methods can not be called random. 4.3.2.1.5 Selection of survey units If the address (households, institutions or functions) is not identical with the targeted respondent, then a traceable and defined random method shall be applied in order to chose the respondent (e.g. registry of all household members older than 14 years, all decision makers in a company, the audience of magazines in a company etc.). In the case where the interviewer selects the interviewee from this registry a random method shall be applied and this method shall be documented. 4.3.2.1.6 Quota sampling In case of the quota method the interviewer shall choose the potential respondent if certain quota criteria are known and fulfilled. SIST EN 15707:2009



EN 15707:2008 (E) 11 The following conditions/terms shall apply to a representative quota selection:  variables used to define the quota cells shall comply with the survey objective;  quoting shall be objective, independent from media, and specified;  interviewer network shall be directed in such a way that differences regarding work methods and selection are minimized. The complete survey units in a quota method should be arranged in such a way that the distribution of single quota characteristics correspond to the structure of the total universe. Thereby it shall be differentiated between  combined quotas, where the person to be interviewed is individually defined through a combination of quota characteristics and  un-combined quotas, where it is up to the interviewer to decide how the separate cells overlap. Prerequisite for the appliance of a quota method shall be the knowledge of the structure of the essential characteristics within the total universe. In order to judge the quality of a quota sample the following criteria shall be necessary:  statistical reference sources for the distribution of quota characteristics;  distribution of combinations of quota characteristics;  control through comparison of the distribution of quota characteristics with the corresponding statistical reference sources;  additional characteristics not controlled by quota shall be compared to other available statistical reference sources and shall be documented. 4.3.2.1.7 Web-based sampling NOTE Web-based sampling approaches in general do not yet comply with the needs of sampling methods and it is not yet clear if web-based sampling methods will be able to meet the requirements for random and/or quota sampling procedures as described above. If web-based sampling is used, a full description of method and validity shall be provided. Self-selection approaches (e.g. including pop-up questionnaires appearing to each nth visitor of specific websites) shall not be considered as methods of random sampling according to this European Standard. 4.3.2.1.8 Mixed procedures For all mixed procedures e.g. random quota, quota panel, random panel or random panel quota it is compulsory to document each partial method as described above. The response rate shall be documented at each level, not only at the last level (e.g. fieldwork via panel). In the case of mixed procedures a sample shall not be defined as random. 4.3.2.2 Weighting NOTE Weighting is a technique to assign importance weights to cases and/or subgroups of a survey via mathematical methods. This weight represents the impact - which is concretized and manifested by appropriate
pre-assigned criteria - within the population represented by the survey. SIST EN 15707:2009



EN 15707:2008 (E) 12 The weighting process shall be documented in detail:  source and date of weighting target data shall be provided;  techniques: weighting of cells and/or iterative weighting of marginals;  weighting matrices (e.g. gender x age groups);  comparison of the weighted and un-weighted sample (in case of disproportionate sampling the correction shall be applied first);  documentation of the structures of variables also for those which were not used as weighting criteria;  distribution of weighting factors (e.g. percentiles). In order to evaluate the weight distribution the weighted total shall have the same size as the un-weighted total. The effective sample size should be calculated. If the sample is clustered the effective sample size should be calculated for each subset (cluster) separately. The weighting process shall be documented in a way that a third party is able to understand the weighting process without further information. 4.3.2.3 Data integration methods: fusion, ascription and imputation NOTE 1 Fusions are standard methods in print media analyses. Main areas of fusions are: — completion of missing information within a data file for some of the sample (e.g. separation between personal interview and leave behind questionnaire with incomplete return); — completion of missing information within a data if certain themes where run in a random selection or skipped systematically in order to stay within an adequate interview time; — merging of independent samples. NOTE 2 Fusion methods assume the same hypothesis that people with similar demographic and/or attitude and consumption characteristics also have similar values in other characteristics that are only available in the other study or sample. NOTE 3 Prerequisite for a successful fusion is a large number of common variables in order to find the ‘best’ donor for each recipient and an appropriate ratio between donors and recipients. In order to judge the quality of fusion the method and its verification shall be documented in detail:  comparison of donor and recipi
...