SHARE THE ARTICLE ON
Internet surveys typically take the form of computerized self-administered questionnaires that are accessed through the web. Internet surveys can take many forms as they can be conducted through multiple different web channels including through emails, websites, and even social media. Respondents can read the questions from the screen of their electronic device and manually record their own responses. Once a respondent completes their survey and submits it, the surveyor receives the response in real-time.
Get ready to uncover the how, when, what, and where questions in a research problem
Technology has influenced two major aspects of surveys:
Although internet surveys have drastically broadened the potential of surveys, they also come with their own complications. The increasing number of interfaces through which surveys can be conducted (mobile phones, computers, etc) can make the process of surveying more complicated, increasing the reliance on skilled professionals and specialised software to implement.
In this section, we will take a look at a few prominent methodological issues associated with internet surveys.
Some prominent advantages of internet surveys are that they are extremely easy to implement, inexpensive when compared to other methods of surveying, and that the process of recruiting respondents is relatively easy. However, survey practice has shown that sampling for internet surveys may be just as complicated as with other modes, if not more difficult.Â
As most web surveys generally have a pool of self-selected respondents, these surveys are inherently associated with non-probability sampling. It is important to note, however, that the issues related to probability vs non-probability samples are independent of the survey mode used.Â
The inferential problems associated with non-probability sampling can be approached using advanced calibration methods and propensity score weighting to improve adjustment procedures. Research has shown a lot of potential when employing multiple imputations, data fusion, and proper philosophy of causality.Â
When probability sampling is used to choose a sample group for an internet survey, we come across the following two issues; frame problems and coverage issues.Â
The coverage issues stem from the fact that not all of the population uses the internet. This is an important problem to address as the users and non-users of the internet are likely to have pivotal differences that may skew the results of a study if not accounted for.Â
The sampling frame issues are caused due to a lack of access to the contacts of those in the target population. For instance, if email surveys are conducted, there is no email directory consisting of email IDs of everyone in the general population.Â
Another key issue associated with internet surveys is the complexity of the survey response process. Even after an email invitation is sent to a respondent, a researcher can never know whether it was actually delivered to the respondent or not. Typing errors, spam filters, or simply recipients overlooking their emails can all lead to respondents never coming across the survey at all. Additionally, in internet surveys, respondents can easily choose to quit the survey at any moment in time if they decide they do not want to continue.Â
When compared to other survey modes, internet surveys tend to have a 6% to 15% lower response rate. These low response rates reinforce the importance of incentives in internet surveys.
All in all, there are four key components that influence the response process of internet surveys:
Conducting exploratory research seems tricky but an effective guide can help.
Measurement error refers to the difference between the values reported by respondents and the true values of the variables. This error can be caused by respondents intentionally or unintentionally.Â
In self-administered survey modes, such as internet surveys, there are two main sources of measurement error:
See Voxco survey software in action with a Free demo.
Internet surveys typically take the form of computerized self-administered questionnaires that are accessed and filled out by respondents through the web.
Some key advantages of internet surveys are that they are inexpensive, quick to conduct, and are generally easy to implement.
The internet has influenced surveys in two major ways: multimedia components can now be incorporated into surveys and surveys are significantly more interactive.
Measurement error occurs when the values reported by respondents are different from the true values of the variables. Respondents may do so intentionally or even unintentionally.
Some key methodological issues associated with internet surveys are caused due to sampling issues, measurement error, and solicitation and nonresponse issues.
Read more
English
French
