Nfield’s Quality Control Options for Rock-Solid CAPI Data

Quality control is a process when fieldwork executives look into various source files to evaluate whether the interviews have been conducted with honesty and accuracy. It is in the researchers’ best interests to ensure that the data delivered to customers comply with the highest quality standards.

Nfield CAPI for smart quality control

In Nfield CAPI you can use location tracking and silent recording with ease to verify data and the credibility of your interviewers.

Silent recording enables you to record either the whole interview or parts of it without the interviewer’s knowledge. It is set up via scripting by inserting a simple command.

Location tracking means collecting unbiased information about the locations where your interviews have been conducted. To allocate an interviewer, the device has to be equipped with:

  • the GPS receiver which is the most popular and more accurate option and/or
  • Wi-Fi and mobile network – GSM, 3G or 4G which, on average, provides less accuracy than the GPS.

Things to remember about GPS

  • The interviewer can activate both the GPS and Wi-Fi and mobile network options in device settings, and let the system decide which one will capture the location best given the current conditions. It is very convenient, though activating both options consumes battery life very quickly.
  • How accurately the GPS is able to capture a location depends on various factors: number of satellites in view, mountains and ionospheric disturbances, signal interference caused by buildings and number of channels on the receiver. Specifications for most GPS receivers indicate their accuracy will be within 5-15 meters, 95% of the time. This assumes the receiver has a clear view of the sky and has finished acquiring more than three satellites. If the conditions are not ideal then the accuracy declines rapidly.
  • Nfield CAPI automatically attempts to capture the current location of an interviewer when the survey is started. This process happens discreetly in the background and without interviewer intervention. If the required accuracy is not achieved, the interviewer will need to confirm the best available location match manually after completing the interview. The location is saved in paradata and only persons with assigned rights are permitted to download this to view the locations of the interviews.

More options: measure twice, cut just once

In addition to the aforementioned methods, you can utilize Nfield CAPI for:

  • Speeding – measure the average time needed to take the survey and check if some interviews have been completed way faster than is realistically possible. You can download the interview duration easily.
  • Straightening – analyze the data to see if the same answer category for a series of questions appears too frequently.
  • Pictures – look into the data and view the pictures snapped during the interviews.
  • Call back – use the gathered contact information of respondents to conduct random calls to verify the data.

You can reject interviews that do not meet the quality standards in Nfield CAPI which will then  eliminate them from the complete target counts and exclude their data from the overall survey data.

Care about the best quality for your customers

False or insincere interviews provide useless data that cannot be delivered to customers, nor do they support the researchers’ objectives. The situation can even tarnish your reputation if the wrong data is used for expensive, important analyses that influence customers’ business decisions. Whilst the majority of interviewers respect their job and the nature of their work, there will always be a small percentage who will try and find shortcuts.
A lack of quality control usually results in more cases of false interviews. Pen and Paper researchers struggle with quality the most. Based on our experience, a few dishonest interviewers are revealed at the start when switching from Pen and Paper to CAPI, and as soon as the interviewers realize that the quality of their work is being thoroughly checked, the number of unfortunate incidents decline. Further features of the quality control system then prevents a come-back of dishonest practices.