Surveys

I have been a bit anxious that perhaps as my question evolves it no longer sits with my methodology. In some ways methodology has been a centering point for the development of my question, I am interested in using this opportunity to do something different to evaluations that by definition have to focus on predetermined objectives. However, my supervisor keeps reminding me that I need to include the ‘local economy’ and encouraging me to make sure I ask questions about what people do in their suburb (shopping, hairdressers, etc). Clearly ticking boxes as to which activities out of my list people do in their suburb would be better addressed through a survey.

However, I feel a bit more relaxed this morning about the fact that there is no way I will be able to get such wide ranging data after I clicked through to do an online survey about smoking to see what filling in surveys feels like. One of the screening questions was whether or not you do certain things daily or almost daily. One of the check boxes on the list was drinking red or white wine and the next was drinking beer. I think that this was a major design flaw as they did not put a time frame (unless they are just interested in seeing if people consider themselves a daily drinker) and ignores the fact that for many people beer, wine and spirits are probably goods that can be substituted for each other.

If they were speaking with people they would be able to work out to what degree different alcoholic drinks are considered substitutable products. Now, they might not be interested in speaking to daily/ near daily drinkers as much as they have for separate quotas for beer drinkers and wine, which could be fair enough. However, if they have any interest in whether or not people are more likely to smoke when drinking beer then having data on whether or not there is a correlation between an individual’s smoking and beverage choices (i.e. if some people smoke when they drink beer but not wine).

Clearly I think talking to people is so much more interesting and I have no idea what data they survey designers need to collect, so I really cannot comment on their survey. However I like to think my methodology is not based on what I think is ‘fun’. Substitution, seasonal variability and contingencies do matter in my project. Hopefully my less systematic data collection will not leave me wishing I could go back in time when I come to write up.

Advertisements

Tags:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s