No Bitter Notes: Ensuring Data Quality in Stiegl’s Brand Tracking Study with ReDem
Julia Mittermayr1, Christoph Auböck2
1Redem GmbH, Austria; 2Media1
Relevance & Research Question In an increasingly complex global landscape, Germany’s Federal Foreign Office sought to gain accurate insights into how Germany is perceived worldwide across various languages and cultures. Ensuring genuine, high-quality responses was paramount to answering the question, "How is Germany perceived?". However, the rising sophistication of survey fraud and AI-generated responses were identified as a significant potential threat to reliable insights. INFO GmbH, the agency leading the study, recognized that traditional quality checks are likely to be insufficient for a project of this scale and complexity. Hence, ReDem’s AI-powered and human-refined tool was chosen to tackle the challenge of identifying and cleaning bad survey data, ensuring high data quality across diverse cultural and linguistic contexts.
Methods & Data INFO GmbH used ReDem’s integration with the survey tool keyingress to apply multiple layers of real-time checks as the data was collected. Each response underwent a thorough, automated assessment: time checks ensured attentive engagement, grid question patterns were analyzed for unusual answer behaviour, and sophisticated algorithms reviewed open-ended answers in multiple languages. These reviews automatically flagged off-topic, duplicated, copy-pasted, or poorly worded responses, etc.
Results By the end of the study, only high-quality data remained in the dataset for the Federal Foreign Office, empowering INFO GmbH to deliver precise and reliable insights with minimal manual intervention. ReDem’s technology flagged and removed X% of responses in real-time due to low-quality indicators, catching fraudulent and inattentive data that traditional methods may likely have overlooked. INFO GmbH estimates that ReDem enabled them to capture an additional X% of poor-quality data, raising the overall quality of the dataset. Beyond ensuring data quality for the Federal Foreign Office, ReDem’s solution saved INFO GmbH approximately X hours of manual data review, allowing the agency to reallocate resources strategically and enhance the project’s efficiency.
The Teilhabe-Community. An infrastructure for research projects involving individuals with disabilities
Lea Thönnes1, Matthies Tobies2, Armgard Zindler3, Robert Grimm4
1Aktion Mensch e.V., Germany; 2Ipsos; 3Ipsos; 4Ipsos
Relevance & Research Question
People with various impairments or disabilities are often not adequately represented in (population) surveys, and their specific subgroups are not identifiable in result analyses. Similarly, participatory product development with people with disabilities rarely takes place.This group wants a voice, a right protected by the UN Disability Rights Convention. Aktion Mensch develops innovations for an inclusive society, including the idea of a special panel. Ipsos has been brought on board for a collaboration to establish this panel. Ipsos and Aktion Mensch launched the "Teilhabe-Community" in early 2023. This panel of around 900 individuals with disabilities is available for diverse research about their life realities and everyday experiences. Due to barriers, these individuals often require specialized communication or customer engagement, necessitating an additional panel.
Methods & Data
Ipsos hosts the "Teilhabe-Community" on its own online platform. Following a participatory approach, we work with people with disabilities together to ensure the community is as barrier-free as possible.We aim to capture a diverse range of impairments in our panel. In addition to online panels and social media, we leverage contacts through leading social welfare organizations (Aktion Mensch members). Workshops and events, involving advocates like facility directors and support staff, help convey our message.The introduced exchange feature on the platform, requested by individuals with disabilities, also serves the purpose of direct exchange with this group. Moderators assist inexperienced panelists in expressing and forming opinions.
Results
The "Teilhabe-Community" serves researchers and companies for varied inquiries, supporting both quantitative and qualitative methods, including offline activity recruitment. Applications span from usual social research to customer and user experience, including customer journey and more.The "Teilhabe-Community" enables companies to integrate diverse experiences of people with disabilities into product or service development, fostering an inclusive society. Aktion Mensch and others have already conducted surveys and website tests.
Added Value
The unique feature of the "Teilhabe-Community" is its easy and quick access to this group. Panelists register in advance, sharing information on their impairments and demographics for targeted sampling for research purposes. The community also allows panelists to interact. Involving them in research significantly contributes to fostering inclusivity.
Developing a detailed understanding of patient journeys and treatment pathways through a co-created digital-first approach, to rapidly collect real-world patient experience data.
Daniel Simon Chapple, Reiner Lehmann
DontBePatient Intelligence, Germany
Relevance & Research Question The pharmaceutical client entered the prostate cancer indication for the first time, introducing innovative treatments deviating from standard care pathways. To ensure patient access, diagnostic and therapeutic pathways needed assessment and potential reshaping. The client aimed to identify healthcare professionals (HCPs) involved in various treatment stages across the USA, UK, and Germany and sought strategies to reach and inform patients effectively. A key goal was to strengthen engagement with key opinion leaders (KOLs) and patient organizations, aiming to produce at least three publications for collaboration. Methods & Data A detailed survey, developed with KOLs and patient organizations in the target countries, combined clinical questions with validated scales (BPI, FACT-P) and adaptive designs. This facilitated the comparison of survey results to established metrics. Recruitment employed a digital pathway with social media, search engine ads, patient organizations, and databases while upholding data privacy and participant anonymity. Surveys were conducted in English, German, and Spanish for respective populations. Data analysis applied descriptive statistics and machine learning to uncover clusters and develop predictive models for identifying patients needing enhanced support. Results Data collection from over 15,800 prostate cancer patients was rapid, taking less than 60 days, with participants showing a median age of 70 and a representative distribution of disease severity. Notably, most participants had no prior interaction with patient organizations but expressed interest in anonymous digital platforms for support and information. Urologists in Germany, clinics in the USA, and clinical oncologists in the UK were key influencers in patient decisions. Despite diverse healthcare journeys, patients shared similar attitudes and expectations across countries. Added Value The project provided crucial data to refine communication strategies and define messaging channels for target segments. Findings facilitated physician engagement, reinforced early screening advocacy, and led to the creation of a digital prostate cancer community. Ongoing collaborations with KOLs and patient organizations yielded several research publications and informed treatment launch strategies, including expansion into Japan.
|