Arozin - Quantitative research, reflection

I have always found quantitative methods and their analytical potential interesting, especially in surveys and such so this weeks theme suited me well. The workshop we had with Olle Bälter was really interesting due to its interactivity and student participation.

I have worked with the creation and design of web surveys before, so i´ve heard a lot already, but he still made a lot of valid points, like the possibilities to create intelligent surveys, which means that the following questions can be regulated by your answers from the earlier questions.

Another interesting point he made was the one of respondent value, that before creating a survey and distribution it you shall make an extensive matrix of all your different respondent groups, with a good overview of their different values so you know which one will provide the best data for analyzing purposes. I would also argue that this is one of the most common mistakes while creating a survey, people think it is enough to create a good survey and throw it on the internet without evaluating what groups to target through their distribution.

When I started working with web surveys I learned that you should always have an even number of answers, so that people can't take the a neutral standing and have to think it through (a neutral answear doesnt give us anything) This is affected by the surveys theme of course, you should give the respondents a neutral choice if the survey is of political standings or similar due to personal and democratic integrity, but you shouldn't if the survey is a course evaluation or similar. Well this is all my opinion, and what i learnt from Bälters workshop is that this choice isnt black or white, theres always different cases that needs different approaches to answer the specific question.

We also talked about the distribution of surveys, and their sustainable aspects when it came to the environmental, economical and social aspects. One curious thing is that web surveys aren't always more environmental friendly than paper surveys, that web surveys have a higher initial cost due to servers, software and similar expenses.

One economical difference between paper surveys and web surveys is that expanding the survey itself has a higher impact on the individual production cost of each paper survey due to its constant increase in weight, actual paper and ink that is being used. Similiar kind of expansion does not affect the web surveys in the long run.

One interesting thing is the application on different demographic groups. Can the results of a survey taken place in sweden represent the danish population due to the similarities between the countries? I believe that this is fully possible, as long as you mention this, so people can take this in remembrance, this provides the possibility to apply this to more controversial demographic groups, and maybe not use the data as a result itself, but rather for comparison.

One thing i have learned from working with web surveys is the fact of precision, simplicity and accuracy. Don't ask a question within a question, don't use unnecessary and extensive formulation, don't have complex questions and other traps. I got this confirmed today at the workshop, that a survey shall always be effective, easy to understand, evaluate each question before publishing it and most important of all, it should not be to long (depends on the matter at hand)