|
This assessment is conducted as data is being collected and during data processing once a project closes. Data which clearly indicates poorly considered, inconsistent, rushed and confused input is flagged and may be removed from the final data set.
In addition, participants who are supplied to us via external sources are also screened thoroughly. Those contributions that are deemed invalid on two or more occasions are added to a black-ban listing managed within our system, thus preventing them from participating in further research conducted by i-Link.
In the same way, we compile a history on all i-Link panel respondents and can determine those who have been invited to participate in a set number of our surveys. Although rare, our system will block entry to respondents who have previously completed more than 3 surveys in a 6 month period.
Within a survey, we will also measure a participant's response time for total interview length as well as time taken to complete various questions. A participant whose response times are consistently shorter than is reasonably expected indicates a failure by that person to contribute a valid and considered response. These invalid interviews are then flagged and removed from the final data set.
|
|
We also adhere to best practices for minimising inadvertent respondent error as well as the potential for respondents to be intentionally misleading. For example, in our system, once a respondent has been screened out they are unable to alter their response and continue with the survey. We also deploy smart screening techniques which minimise lie-responses.
We can provide assistance with survey designs that enable the authenticity of responses to be verified and assist with techniques for applying questions in ways that will garner greater consideration by the respondent.
Our Data Processing department examines open-ended data in each project looking for nonsense or 'junk' responses. In addition, we can cross-check closed (quant) questions and flag respondents who demonstrate obvious and repeated inconsistencies in their responses.
As a final measure, with projects where lists and contact numbers are provided we will undertake a random telephone validation of up to 10% of the respondent details given. |
|
 |
 |
|
 The ancillary services provided by the i-Link bureau
team include: |
|
 |
 |
|
