Digital by Design
Digitisation is not moving a form onto a screen. Transferring a paper questionnaire to a tablet unchanged is not real digital research.
A good digital questionnaire protects quality while the data is being collected, through skip logic, validation rules, range checks, mandatory fields, automatic calculations, cascading selections, GPS and time stamps.
Some errors are then prevented rather than cleaned up afterwards.
CAPI, CATI & CAWI Infrastructure
ID can run different research models inside the same digital system: CAPI for face-to-face interviews on a tablet or phone, CATI for centralised telephone interviewing, CAWI for online self-completion — separately or in combination.
The target group, the sampling structure and the operating conditions determine which method is used.
KoBoToolbox & Digital Survey Systems
KoBoToolbox and similar digital collection infrastructures can be used in ID field operations.
These systems support questionnaire programming, cascaded select structures, automatic calculations, control rules, multi-language support, GPS and time information and data export.
Field data is then not merely collected; it is produced in a standardised form.
A realistic field story
"The Total Is Right But the Distribution Is Wrong"
Say a national study targets 5,000 interviews. The dashboard shows 4,750 completed. At first glance the field looks nearly finished.
Then the regional dashboard opens: Marmara has reached 115% of its target while the eastern region is only 63% complete. The total looks good, but the sample structure has started to break down.
This is where digital monitoring becomes critical. The question is no longer "how many surveys have we done?" but "have we done the right surveys in the right places?".
Real-Time Field Monitoring
Managing the field while it runs, not after it ends. Dashboards make it possible to monitor completed interviews, daily field production, provincial and regional distribution, quota achievement, refusal rates, partial interviews and enumerator performance.
That structure lets the operations team respond to field problems faster.
Dashboard is not Decoration
A dashboard is not just a nice chart. Its value is not measured by the number of graphics it holds. The real question is which decision it supports.
Dashboard design at ID therefore starts from the user. A project manager may ask whether the field is going to plan. A quality manager may ask which enumerators carry risk. Senior management may ask whether the core indicators are approaching target.
The same data needs different screens for different users.
Power BI & Decision Dashboards
Interactive dashboards can be developed with Power BI and similar tools as required: KPI cards, trend charts, regional breakdowns, target-versus-actual indicators, filters and drill-down analysis.
The aim is not to move a table of hundreds of rows onto a screen. It is to simplify the information a decision needs.
From Field Data to Management View
In a well-designed digital research system the process runs as one flow. Field data enters the system, control rules run, suspicious records are flagged, the dashboard updates and the manager sees the current position.
This turns research from a static report into a living information system.
- Data collection
- Validation
- Quality control
- Dashboard
- Decision
Paradata & Operational Intelligence
One of the significant advantages of digital research is that it measures not only the answers but the interview process itself.
Interview duration, start and end time, GPS, device information, question transitions and completion behaviour can all be examined through paradata.
These produce important early-warning signals in field quality control.
Automated Quality Rules
Some checks do not need to be made one by one by a person. If a value outside the age range is entered, if the total headcount does not match its breakdowns, if a date sequence is illogical or a mandatory field is left blank, the system can raise a warning.
That reduces the data cleaning burden and surfaces field errors earlier.
Multi-Mode Data Integration
Sometimes data arrives through different channels in the same study: face-to-face CAPI, telephone CATI, online CAWI.
The main problem then is not merging the data. It is merging it in the same variable structure, with the same coding system and in a comparable form.
ID handles that from questionnaire design through to database structure inside one methodological frame.
Live Monitoring for QA/QC
Digital dashboards serve quality control as well as operations. Very short interviews, high refusal rates, suspicious GPS density, unusual answer patterns and high missing data rates can all be flagged.
The quality team can then work risk-based alongside random checking.
A dashboard story
A Dashboard Story: 95% Complete May Really Be 70%
Say a system shows the total field target 95% complete. Look closer and the picture may differ: quotas may be unfilled in some target groups, some provinces may be short, some records may not have passed quality control.
If the dashboard only shows the "completed" count, the field looks 95% done. If it shows "validated completed", the real figure may be 70%.
That difference matters a great deal. The purpose of a digital system is not only to show data quickly; it is to show correctly defined data.
Data Visualization
Making complex data understandable. Good visualisation simplifies data without distorting it.
The right chart type, scale, comparison, filtering and indicator definition all matter. We know a percentage change can tell a different story from an absolute number, so visual design and statistical accuracy are handled together.
Dashboards for Monitoring & Evaluation
Dashboards are particularly strong in monitoring and evaluation projects: baseline values, targets, current achievement, regional performance and outcome indicators can be followed in one system.
Evaluation then stops being a report prepared at year end. The programme manager can follow the process while the project is running.
Role-Based Dashboards
Not everyone should see the same screen. Different users have different information needs, so dashboards can be designed by role.
- Field Coordinator
- Sees field production and team performance.
- Quality Manager
- Monitors risky records and QA/QC indicators.
- Project Manager
- Follows target-versus-actual and regional progress.
- Senior Management
- Sees summary KPIs and critical alerts.
Secure & Controlled Access
Data security is critical in research dashboards. User authorisation, role-based access, restriction of personal data, anonymised displays and access control can be applied as required.
Not every user needs to see all the raw data. A good digital system provides as much access as the role requires.
Reporting Automation
Digital systems can save considerable time in repetitive reporting. In weekly field reports, completed interviews, refusals, quotas, regional distribution and quality indicators can be updated automatically or semi-automatically.
The team then spends its time interpreting the result rather than rebuilding the same table every week.
From Dashboard to Action
The most important point: if a dashboard does not lead to a decision, it is only a screen.
We therefore try to establish a clear chain for every critical indicator. If the refusal rate passes a defined level, a warning is raised. If a region falls behind target, a warning is raised. If an enumerator’s interview duration drops unusually, a quality check is triggered.
This turns a dashboard from a passive reporting tool into an operational decision system.
- Indicator
- Threshold
- Alert
- Action
Digital Research Across Regions
In regional and multi-country research, digital infrastructure matters even more. While local teams work in different countries, a common data structure, a common questionnaire standard, common quality rules and a central dashboard can be used.
Local implementation is preserved while central coordination and comparability are maintained. This also supports our Global Standards + Local Expertise + Regional Perspective model technically.
