Our Approach and Quality Control

Quality underpins all that we do

EDI is known for, and proud of, the emphasis it places on collecting high quality data.  In all studies we undertake, ensuring the highest quality data underpins all that we do.  From questionnaire design and rigorous testing of the questionnaires in electronic form and in real field conditions, through to selection and training of the best field staff, in-field stringent monitoring processes at all levels, and strong data cleaning and consistency check protocols, we ensure that we use internationally regarded best practice and we even add our own extra levels of quality control.

Using Electronic Methods to Improve Quality

Our ability to consistently collect and deliver high quality data, is assisted by our choice of  using Surveybe data collection (CAPI) software for minimising data error and inconsistencies and providing us with easy to use functionality for interviewing in multiple languages and accessing real-time data

  • It removes mistakes that arise when re-keying data from paper into a computer, as data is entered directly by the interviewer
  • It automates questionnaire skipping patterns that ensures only the right questions get asked
  • It performs hundreds of consistency checks on the responses given during the interview and ensures that only clean responses get recorded
  • It can automatically capture elaborate GIS data, which allows for more detailed, and ensures more accurate, GIS data

 

Data Quality Monitoring

Throughout the field phase we continually perform checks on the data we are collecting.  Interviewers perform data consistency checks on each screen in the electronic questionnaires.  Supervisors, perform both automated and visual data consistency checks of all interviews on a daily basis, as well as observing, and revisiting/re-interviewing a sample of respondents.  A QC Coordinator from HQ will also typically visit and conduct both random revisits and speak with area leaders for feedback on the performance of our teams.   When the data is received at HQ it goes through various further levels of data checks, from running internal Surveybe consistency checks , using bespoke Stata-do check files, to visual reviews and checks, all to ensure when the data is delivered to our clients, or used internally for our own analysis and research, we can be sure of the high level of accuracy of the data we are using.

We consistently use our observations and findings during the QC processes to improve upon the data being collected within the particular project through feedback to the teams, amendment of the questionnaires, adding further field consistency checks as well as using them to assist us in continually improving our internal systems, training programmes and monitoring protocols for future projects.

Data Confidentiality and Protection

We follow rigorous data confidentiality and privacy protocols in line with international standards.

Whilst the data that is collected can be available immediately to HQ and is regularly transmitted between the field teams and headquarters, the data-files are fully encrypted on the enumerators hardware unit to become immediately secure.  Data is then only externally accessible to those given specific access to use the decoder at HQ.

Data Delivery

Data collected by EDI is typically delivered to our clients by our internal researchers in both Stata and CSV format with full documentation.

What creates good quality?

An EDI Project Coordinator’s Perspective