Question

Discuss the challenges a researcher would encounter in transforming an area probability face-to-face survey into an Online survey when attempting to estimate the same statistics.

Answer 1

Face-to-face surveys have several key strengths. These surveys are clearly structured, flexible and adaptable. They are based on personal interaction and can be controlled within the survey environment. Physical stimuli can be used and respondents are able to be observed. On the other hand, there are also some disadvantages, such as interviewer bias, high cost per respondent, geographical limitations and time pressure on respondents.

Online surveys have a number of strengths, such as lower cost and higher speed; they are visual, interactive, and flexible; they do not require interviewers to be present and busy people – often educated and well-off – who systematically ignore taking part in a telephone survey are willing to answer questions posted on their computer screens. Nevertheless, notes that relying on such modes, which require initiative from respondents, will likely lead to selective samples, raising concerns about nonresponse bias. Samples used for large national and international face-to-face and telephone surveys are considered representative of the general population, while online samples are currently regarded as representative of population subgroups only.

Challenges a researcher would encounter are:

• The research produces data based on real-world observations (empirical data).

• The breadth of coverage of many people or events means that it is more likely than some other approaches to obtain data based on a representative sample, and can therefore be generalizable to a population.

• Surveys can produce a large amount of data in a short time for a fairly low cost. Researchers can therefore set a finite time-span for a project, which can assist in planning and delivering end results.

• For internet-based methodologies focus mainly on sampling issues. However,other issues are raised around mode effects, where, for example, it is known that online respondents use scales differently from respondents in other modes. There is conflicting research on this, some showing that online respondents are more likely to choose midpoints in scales and ‘don’t know ’ options in general, and other research, in contrast, suggesting online respondents tend to choose extreme responses on these scales.

• These types of effect will be due to complex competing effects of response styles and do not necessarily make responses from online surveys less accurate, but they can cause problems when we attempt to switch to an online survey approach in tracking work. It is possible to correct for this to an extent through modelling, but this is likely to be viewed as less straightforward for those commissioning.

• Area based samples require listing of households in defined areas and selection from that list. If the listing process is not done properly then dwellings can be missed which causes under coverage. The main reasons for incorrect listing are poor maps, poor instructions, hard to find dwellings and incorrect classification of vacant dwellings.

• If listings are done well ahead of interviews taking place, and there are no processes in the field for adding new dwellings, then the list will not cover the very recent dwellings.