bezoekadres Marnixkade 109 1015 ZL Amsterdam

postadres Postbus 15262 1001 MG Amsterdam E [email protected] T +31 (0)20 589 83 83 F +31 (0)20 589 83 00 W www.motivaction.nl

Answers to the 26 ESOMAR questions concerning online panels 2010

Contents
Page 2
Experience with online samples Sample quality The strength of the StemPunt research panel

Page 3

The use made of the StemPunt panel Sourcing hard-to-reach groups

Page 4

Communication with panel members Panel turnover The opt-in process for new panel members

Page 5
Confirmation of panellist identity and checking for fraud Profile data Capacity of the panel

Page 6

Sampling method Invitation to take part in a survey Respondent incentives Frequency of invitations

Page 7
Privacy policy Data security Quality management Research among children and young people

Page 8

Supplementing samples Panel overlap Expected response

Page 9

All panel membership data are available Data quality control Respondent satisfaction Sample debrief

By answering each of the 26 ESOMAR questions below we aim to demonstrate clearly the strengths of the Motivaction online access panel
More information concerning the 26 questions specified by ESOMAR can be found at www.ESOMAR.org

1. Experience with online samples

What experience does your company have providing online samples for market research? As a full-service research company Motivaction International has been investigating what motivates people for over 25 years. Where quantitative research is concerned we often take advantage of the strengths of online panels. Motivactions StemPunt panel, set up in 2002, provides the basis for this and we are able to guarantee the quality of the panel from a number of different disciplines. Motivaction was one of the first research companies in the Netherlands to be awarded the ISO-26362 quality hallmark certification for online access panels. In addition to the StemPunt panel Motivaction also operates individual panels commissioned by clients.

2. Sample quality
You supply samples for use in online market research. Please describe and explain the types of sources for these samples. Are these databases, actively managed panels, direct marketing lists, web intercept sampling, river sampling or other sources? The objective is to be able to draw a representative sample from the panel, with the quality of the data having top priority. In order to ensure the correct composition of the research panel, it is essential to recruit new panel members on a continual basis. Different methods are used for recruitment both online and offline. Using different sampling methods, such as propensity sampling, enables us to deliver the samples required (see Question 12). Motivaction is capable of supplying a representative sample of the Dutch population comprising 20,000 completes. Different disciplines, such as methodology, database management and panel management, are incorporated in the daily panel operations.

3. The strength of the StemPunt research panel

What do you consider to be the primary advantage of the samples you supply in comparison with samples generated by other suppliers? The respondent profiles from the StemPunt panel are supplemented with Mentality data. Mentality is Motivactions unique values and lifestyle segmentation which provides an effective method of classifying and approaching target groups. Based on the insight provided into peoples perceptions, our clients have a better perspective of what motivates people, what their lifestyle patterns are and how trends originate. It has been proven that Mentality provides a better explanation of peoples behaviour than the traditional classifications based on age, sex, education and income. This means that this segmentation is able to provide more tangible input for effective marketing and communication strategy than is the case for standard models.

4. The use made of the StemPunt panel

Is the panel used solely for market research? Motivaction uses the StemPunt panel solely for conducting market research. The panel is not used for other purposes such as direct marketing, in order to prevent any survey effects occurring whereby research questions could be influenced. In addition to completing online questionnaires the panellists may also take part in other forms of market research, e.g. group discussions, depth interviews and online web tracking.

5. Sourcing hard-to-reach groups

How do you source groups that may be hard-to-reach on the internet? It is a known fact that not all sub-groups within the Dutch population are prepared to take part in a research panel. In addition to this certain groups are relatively poorly represented on internet. For such hard-to-reach target groups Motivaction continually seeks ways of approaching these people as directly as possible. For this purpose a broad range of both online as well as offline sources are employed. In the event that it is considered necessary in order to ensure representativeness, the online sample will be a supplemented with a written or telephone component. A few examples of sourcing methods: Recruitment via postal surveys: based on a written questionnaire completed once every two years by 750 Dutch people. 21Minuten.nl: an independent initiative undertaken by McKinsey & Company and De Publieke Zaak. Participants in this survey are invited to become a member of the research panel. Respondents for the 21Minuten.nl survey are recruited via different (offline) media, such as radio, television, newspapers and magazines. Face-to-face recruitment: Motivaction literally goes out on the streets to ask people if they would like to become a member of the research panel. Online recruitment: in order to ensure sufficient diversity among the respondents Motivaction runs recruitment campaigns on various online networks. Making use of several online networks, with the recruitment campaign being run on many different types of websites. Recruitment from within the research panel: panellists often know the target groups that are harder for the panel recruitment to reach. With this sort of member-gets-member method we ask existing panel members to nominate new panellists. In addition to recruiting panellists Motivaction also invests time in working together with online communities when the difficult sub-groups prove hard to find. An example of this is the working partnership between Motivaction and Marokko Media.

In addition to special recruitment procedures designed to reach the hard-to-find groups, Motivaction uses the effective method of propensity sampling to ensure that the less frequent users of internet are also adequately represented in a sample.

6. Communication with panel members

What are people told when they are recruited? Motivaction ensures that all communication with panel members is completely transparent. The various aspects concerning panel membership are fully explained on the panel website, concerning for instance the surveys and the rewards system. On registration respondents agree to comply with the Terms of Participation and Privacy Declaration. These are completely in accordance with the current legislation defined by the Data Protection Act (Wet Bescherming Persoonsgegevens) and the Code of Ethics for Research and Statistics (Gedragscode voor Onderzoek & Statistiek).

7. Panel turnover
What is your annual panel turnover rate? Having access to up-to-date profile data is extremely important for various components in the research process, e.g. sampling, weighting and reporting. It is for this reason that Motivaction asks all panellists to update their profile data at least once a year. Panellists who do not comply with this request are inactivated and are no longer approached for research participation until an update of the data has taken place. The StemPunt panel has an annual turnover rate of approx. 35%. This percentage is based on the number of de-registered panellists in relation to the panel size. In general there are three reasons explaining panel turnover: Respondents become non-active and no longer participate in survey questionnaires. Respondents fail to update the necessary profile data. Respondents de-register themselves.

8. The opt-in process for new panel members

Please describe the opt-in process Respondents apply to join the StemPunt panel via a double opt-in procedure. The actions involved are roughly as follows: Step 1 - the respondent applies to join via the website or by accessing a link provided in an email invitation. Step 2 - the respondent agrees to the Terms of Participation and Privacy Declaration. Step 3 - the respondent answers the questions in the intake questionnaire on the basis of which his/her personal profile is compiled. Step 4 - those respondents who have not applied via email receive a confirmation email containing a link through which their membership can be confirmed.

9. Confirmation of panellist identity and checking for fraud

Do you have a confirmation of identity procedure on registration and/or any procedure to detect fraudulent respondents at the time of registration? In the application procedure for the StemPunt panel the email address is checked by means of a confirmation email. Membership is only activated once the potential panel member has clicked on the confirmation link. In addition to the checking procedure taking place on application, the panellists identity and response behaviour is also checked in the questionnaires (for more information about this check, see Question 24).

10. Profile data

What profile data is kept on panel members? For how many panellists is this data saved and how often is this data updated? A wide variety of profile data is available for panellists, e.g.: - Mentality profile, over 80% of the panellists are allocated to one of the eight Mentality segments (reference date: September 2009, see also Question 3). The Mentality segment profile is determined only once for each panellist in principle. - Socio-demographic profile data, defined according to the Golden Standard (Central Statistical Office-CBS/ Market Research Association-MOA) are available for every panellist. Panel members have access to these data at all times and are also permitted to update them. It is compulsory for panellists to update the most relevant socio-demographic data at least once every 12 months (see also Question 7). - Thematic background data deriving from Motivactions supplementary surveys, e.g. (print) media and health. These profile data can be updated periodically. In addition to these profile data the answers to all the research questions are stored for each panellist in the single-source database infrastructure.

11. Capacity of the panel

What is the size of the panel based on active members on a given date? Can you provide an overview of active panellists by category and/or on the basis of the recruitment method? Based on the StemPunt panel Motivaction has access to 80,000 active panellists (reference date: September 2009). The term active panellist implies panel members who have updated their profile data within the past twelve months (see also Question 7). Details of recruitment source (method) and panel composition are available on request.

12. Sampling method

Please describe your sampling process. Motivaction applies an advanced method for generating gross samples, based on propensity. In comparison with traditional sampling methods the propensity sampling method offers a significant advantage in that it enables a greater number and a wider variety of variables than socio-demographics alone to be incorporated in the sampling procedure. With propensity sampling it is normal for samples to be drawn which are representative of the whole Dutch population. Obviously it may happen that a specific sub-group is being sought, e.g. a sample consisting solely of women, or one which comprises only people aged between 18 to 29 years. This is not a problem with propensity sampling, which has been developed in such a way that it is possible to draw optimally representative samples for all relevant (sub)groups in the Dutch population. Using propensity sampling also provides Motivaction with a technique for correcting self-selection

13. Invitation to take part in a survey

Please explain how people are invited to take part in a survey. Motivaction is able to contact selected panellists quickly and easily for each survey by sending out invitation emails. In the email it is indicated how long it will take to complete the questionnaire and what the incentive for doing so is. The subject of the research study is normally described in general terms, in order to prevent any recruitment bias based on the research topic. In addition to issuing emails specific groups of respondents may also be notified by text messaging about the availability of an online questionnaire, depending on the nature of the individual study.

14. Respondent incentives

Please describe the nature of your incentive system. Does this vary by length of interview, specific characteristics of the respondents selected for the study and/or any other factors? A panel member receives what are called StemPunten as their reward for completing a questionnaire. These StemPunten are savings tokens which can be exchanged for products or gift vouchers from the StemPunt web shop. The number of tokens that a respondent can earn per survey depends on the length and complexity of the questionnaire.

15. Frequency of invitations

How often are individual members contacted for online surveys within a given time period? Do you keep data on panellist participation history and is a limit placed on the frequency that a member is contacted and asked to participate in a survey? On average a panel member is asked to participate twice a month. Motivaction keeps track of how often each panellist receives an invitation to participate in a survey and how often he/she actually accepts the invitation. Motivaction does not place a limit on the number of times a panellist is asked to participate in research, because this may reduce the representativeness.

16. Privacy policy

Is there a privacy policy in place? If so, what does it state? Is the panel compliant with all regional, national and local laws and regulations, including the those with respect to privacy, data protection and conducting research among children e.g. the EU Safe Harbour and COPPA in the US? What other research industry conditions (standards) do you comply with, e.g. ICC/ESOMAR Code, CASRO guidelines, etc.? The privacy of the respondent is of utmost importance to Motivaction. Therefore Motivaction without question ensures strict adherence to the Data Protection Act (C.B.P. Notification No. 1011350) and the guidelines issued by the Dutch Market Research Association (MOA) and ESOMAR. Clearly worded Terms of Participation and Privacy Declaration documents have been drawn up for the panellists. Participation is obviously completely voluntary and is only requested for the purposes of market research. Research results are analysed on the basis of anonymity unless the respondent has given clear permission for any different form of analysis.

17. Data security

What data protection/security measures do you have in place? Motivaction pays much attention to data security. Passwords and other IT-driven security techniques are used to prevent any form of unauthorised access to the data gathered by Motivaction. Motivaction has located its research infrastructure with EvoSwitch in Haarlem. EvoSwitch is a secure data centre with ISO-27001 and ISO-9001:2000 certification. The optimally designed network configuration makes it impossible for the database to be accessed directly via the internet. The online questionnaire instrument was screened 18 months ago by McAfee and received a positive assessment. The online research software used by Motivaction was audited 2 years ago by KPMG and approved.

18. Quality management

Do you apply a quality management system? Please describe it. The quality management system operated by Motivaction focuses on Motivactions core activity, i.e. conducting research. This process is defined in a Quality Manual and in specific Working Procedures for each type of research. The quality management system has been ISO-9001 certified since 1997.

19. Research among children and young people

Does Motivaction conduct surveys with children and young people? If so, please describe the process for obtaining permission. Motivaction is able to conduct research among youngsters aged twelve years or older. Motivaction offers the possibility of inviting youngsters via text messaging in addition to the traditional email method.

When conducting research among children and young people Motivaction adheres to the Code of Conduct for Research and Statistics. This stipulates that where research among children aged twelve years or older is concerned, no permission is required from parents or legal guardians. Despite this the parents of children aged under 16 years are informed if their child wishes to register with the StemPunt panel. Research among children aged under 12 years is also possible. In such cases respondents are approached via the parent/legal guardian. The presence and ages of any children in the household is recorded in the profile data for the panellists. In these cases it is possible for research to take place among these children once permission has been obtained from the parent/legal guardian.

20. Supplementing samples

Do you supplement your samples with samples from other providers? If so, how do you select these partners? Is it your policy to notify a client in advance when (partly) using a third party provider? Do you de-duplicate the sample when using multiple sample providers for a research project? For research within the Netherlands the StemPunt panel is generally adequate for the fieldwork required. For very specialist or for international target groups Motivaction may, in consultation with the client, make use of the resources of third party sample providers. External sample providers are always selected and evaluated with care. Obviously any overlap between the different samples is eliminated.

21. Panel overlap

Do you have a policy regarding multi-panel membership? What efforts do you undertake to ensure that survey results are unbiased given that some individuals belong to multiple panels? Motivaction participates in the MOA Panel Institute (MPI), a database in which the overlap between different panels in the Netherlands is registered. Information from the MPI can be used to prevent respondents from participating in a particular study. In addition to this Motivaction asks the StemPunt members for details of membership of any other panels. On recruitment these data are used to minimise overlap.

22. Expected response

What is the start rate percentage for the different types of research and what percentage drop out in the course of completing the questionnaire? What is the overall response rate? How are these percentages calculated? The response rates for research studies based on the StemPunt panel average between 25% and 40%. This percentage depends on the research and is influenced by factors such as the target group participating, the subject/length of the questionnaire and the length of the fieldwork period. The average number of drop-outs during a questionnaire is approx. 20%. Where representativeness is concerned Motivaction usually opts for a fieldwork period of one week. This means that respondents who do not respond within a day also have the chance to complete the questionnaire. The response rate is calculated by dividing the net sample by the gross sample.

23. All panel membership data are available

Do you maintain individual panel member data, such as recent participation frequency, date of membership entry, recruitment source, etc.? Are you able to supply your client with an analysis of such individual level data per project? Motivaction stores the profile data and research data in a single-source database. All historical data are available for analysis at individual panellist level. This enables Motivaction to conduct analyses based on panel characteristics, e.g. recruitment source, length of membership, frequency of participation and all other available profile data.

24. Data quality control

Do you use validation techniques to identify inattentive and fraudulent respondents? If yes, what techniques are used and at what point in the research process are they applied? Motivaction attributes top priority to the quality of the research. This means that research data from respondents who have given a different year of birth to that recorded on registering are eliminated from the analysis (the 'year of birth-check'). In addition to this Motivaction checks the research data per respondent for the recorded completion time, the number of times the answer dont know is given and for so-called 'straightlining' (opting for the same answer in grid questions). Data from respondents whose behaviour deviates from the norm on a combination of these points are recorded and subject to extra analysis. If it appears that a respondent has been fraudulent, then he/she is removed from the panel.

25. Respondent satisfaction

Do you measure respondent satisfaction? In every survey respondents are asked to complete a few evaluation questions. This enables respondents attitudes to be measured concerning the subject and the length of the questionnaire. The results from these questions are used in order to further improve the quality and can also be included in the sample debrief details if required. In addition to this Motivaction conducts panel satisfaction research on a regular basis.

26. Sample debrief

What information do you provide to debrief your client after the research project has been completed? Motivaction always includes full sample details for the client in its debrief reports. Included as standard in these reports are the response rate, the data collection period and the sampling method. Comments from respondents, the number of waves and the weighting specification are also included in the report, as appropriate. Any profile details not included on a standard basis in a research report can always be made available to the client on request.