› Overview
› Team
› Careers
› News
› Events
› Press
› Publications
› References
ODC Services is a young and cutting-edge enterprise primarily dedicated to up the ante in terms of quality and keep it at a top-notch level. To be able to extend our market position with brilliant ideas, innovations and know-how ODC Services frequently publishes studies in the field of online marketing research - especially with regard to exchanging recent trends and developments in theory and application.
If you are interested in one of our publications, please contact us via email! We look forward to send you the lecture in question.
Zeig mir, wie du tippst und ich sag dir, wer du bist (Planung & Analyse 1/2008)
It has long been an unsolved problem of online market research done by means of Access Panel sample surveys, that there has been no possibility to verify whether a participant in a survey is really the same who was requested to take part in the corresponding survey via e-mail.
A biometrical identification process developed by the University of Regensburg recognizes the user on the basis of his typing habits. It is possible to identify a person by means of the analysis of these characteristics of typing habits.
The technical viability for the authentication of test cases was verified with the help of a study, as was the acceptance and the willingness of the panelists to participate.
(This article is available in German only.)
A Panelists Scoring Model for an even allocation of surveys with different incidences (GOR 2008)
If panel participants do not belong to the target groups of surveys several times consecutively, this may have a negative effect on the willingness to participate and on the quality of answers in surveys. With a scoring model, panelists, who didn't qualify for a survey due to target group restrictions, were invited to high incidence surveys preferably. This way several consecutive screenouts shall be avoided. A test of this method, conducted by ODC Services, showed a more even allocation of different surveys, thus leading to a more constant motivation level of panelists. This way the willingness to participate in surveys can be stabilized and improved. There is less reason and provocation to give untrue answers, and there are less unsubscribers.
Respondent Authenticity (ESOMAR Panel Research 2007)
An as yet unsolved problem in online market research via random samples derived from access panels is that there is no means of checking whether a respondent is really the person who was invited by email to participate in the survey in question. The university of Regensburg/Germany developed a biometric identification process that recognises users based on their typing behaviour when using a PC keyboard. This presentation would like to present the results of a study testing the technical feasibility of the authentication of panellists via this biometrical approach as well as the panellists' acceptance and willingness to participate.
Paneloverlap and learning effects (GOR 2007)
At this year General Online Research Conference in Leipzig Christoph Irmer and Mathias Peichl (ODC) presented, together with Dr. Olaf Wenzel (Skopos GmbH & Co.KG), a poster dissertation about Paneloverlap. It is important to ensure that members of a panel as possible are not registered in other panels. This is so that we can minimise the risk of multiple answers (if clients conduct their studies with several panel suppliers) and keep the number of so-called "incentive-hunters" in the panel to a minimum.
In order to find out how high the panel-overlap with other panel suppliers actually is, ODC Services conducted an internal study. Using the RRT method (Randomised Response Technique), a panel-overlap of 4% is found with 707 respondents in the German opinion people panels. With the RRT method, we don't ask about participation in our panels in the usual way. In fact, we use special question technology to achieve realistic values about sensitive topics. This low score of 4% confirms our belief that we should continue to recruit actively and use as varied a range of recruiting sources as possible (offline and online).
ODC-Workshop "Lerneffekte in online access panels" (R&R 2007)
The workshop of Christoph Irmer and Mathias Peichl held within the framework of this year's Research & Results Marketing Research Fair received very affirmative feedback. Broaching the issue of learning effects in online access panels the workshop pinpointed with a case study, how learning effects can be quantised and consequently be avoided or at least alleviated.
If you are interested in one of our publications, please contact us via email! We look forward to send you the lecture in question.
Zeig mir, wie du tippst und ich sag dir, wer du bist (Planung & Analyse 1/2008)It has long been an unsolved problem of online market research done by means of Access Panel sample surveys, that there has been no possibility to verify whether a participant in a survey is really the same who was requested to take part in the corresponding survey via e-mail.
A biometrical identification process developed by the University of Regensburg recognizes the user on the basis of his typing habits. It is possible to identify a person by means of the analysis of these characteristics of typing habits.
The technical viability for the authentication of test cases was verified with the help of a study, as was the acceptance and the willingness of the panelists to participate.
(This article is available in German only.)
A Panelists Scoring Model for an even allocation of surveys with different incidences (GOR 2008)If panel participants do not belong to the target groups of surveys several times consecutively, this may have a negative effect on the willingness to participate and on the quality of answers in surveys. With a scoring model, panelists, who didn't qualify for a survey due to target group restrictions, were invited to high incidence surveys preferably. This way several consecutive screenouts shall be avoided. A test of this method, conducted by ODC Services, showed a more even allocation of different surveys, thus leading to a more constant motivation level of panelists. This way the willingness to participate in surveys can be stabilized and improved. There is less reason and provocation to give untrue answers, and there are less unsubscribers.
Respondent Authenticity (ESOMAR Panel Research 2007)An as yet unsolved problem in online market research via random samples derived from access panels is that there is no means of checking whether a respondent is really the person who was invited by email to participate in the survey in question. The university of Regensburg/Germany developed a biometric identification process that recognises users based on their typing behaviour when using a PC keyboard. This presentation would like to present the results of a study testing the technical feasibility of the authentication of panellists via this biometrical approach as well as the panellists' acceptance and willingness to participate.
Paneloverlap and learning effects (GOR 2007)At this year General Online Research Conference in Leipzig Christoph Irmer and Mathias Peichl (ODC) presented, together with Dr. Olaf Wenzel (Skopos GmbH & Co.KG), a poster dissertation about Paneloverlap. It is important to ensure that members of a panel as possible are not registered in other panels. This is so that we can minimise the risk of multiple answers (if clients conduct their studies with several panel suppliers) and keep the number of so-called "incentive-hunters" in the panel to a minimum.
In order to find out how high the panel-overlap with other panel suppliers actually is, ODC Services conducted an internal study. Using the RRT method (Randomised Response Technique), a panel-overlap of 4% is found with 707 respondents in the German opinion people panels. With the RRT method, we don't ask about participation in our panels in the usual way. In fact, we use special question technology to achieve realistic values about sensitive topics. This low score of 4% confirms our belief that we should continue to recruit actively and use as varied a range of recruiting sources as possible (offline and online).
ODC-Workshop "Lerneffekte in online access panels" (R&R 2007)The workshop of Christoph Irmer and Mathias Peichl held within the framework of this year's Research & Results Marketing Research Fair received very affirmative feedback. Broaching the issue of learning effects in online access panels the workshop pinpointed with a case study, how learning effects can be quantised and consequently be avoided or at least alleviated.
