Technology

The online data gathering solution consists of two parts, the back-end and the front-end. The back-end is used by the researcher to build, distribute and monitor the surveys while the front-end is the side of the solution that the participant will use to complete the survey.

Our software can run on both UNIX and Windows based servers giving our clients ultimate flexibility when it comes to utilising existing IT infrastructures.

Most important features

Surveys: Each survey has a set of options among which are the following: start/end date, number of responses, password, online/offline status, security properties such as which host to receive responses from and so on.

Survey Design: Each survey consists of areas, sections, groups, questions, questions options and questions keys. These are the building block that come together to create a survey. When starting building a survey the user is presented with the option of using a wizard to create the survey’s structure or create a very simple survey structure which can then be edited accordingly.

• Areas are used to separate question groups that cover different topics. For example, a survey may have an area covering the research topics and a second area with questions aiming to gather personal information about the participant. Areas can be thought of as pages in a paper questionnaire.
• Sections are used for conditional logic. A section carries the questions to be asked while the groups of questions, that the section contains, correspond to an answer of the conditional logic question. A section may contain one or more groups of questions, that may be part of the conditional logic or not.
• A group may contain one or more questions. If a group is part of the conditional logic, its questions will only be asked if the answer selected matches the answer associated with the group.
  Netservey supports 15 types of questions: Open ended: one line & multiple lines, Multiple choice: one answer, one answer-menu, dichotomous, multiple answers, Matrix: one answer, multiple answers, spreadsheet, Scales: Likert, rating, semantic differential, Stapel and finally type of questions that do not fall in any of the above categories such as rank order and constant sum.

  

• The survey options that the researcher selects are automatically applied to the modes based on their properties. For example, if a survey contains conditional logic this will make the web version of the survey to filter out the questions that do not apply. Also, on the web the solution can force the participant to complete all required solutions before continuing. These are not options that are available for paper surveys, as it is not an interactive mode!

Previews: Once the survey is ready, the researcher has the option to see how the survey will look like on the various modes, for example on the web, when it will be sent via email or even download a PDF version.

Sampling: The sampling manager is used to invite participants to complete a survey. The type of invitations sent depends on the participant’s preferred mode, which by default is the web. Clients who have signed up for an enhanced subscription will have access to databases, such as the electoral roll. Alternatively, the users can upload their own databases.

Analysis: Once the first response is received the researcher can perform univariate statistical analysis on the data. This includes means, modes, standard deviations and percentages. The results are also represented in graphical format.