Miss Gonzalez's [attempts at doing a] doctorate II
April 2015
“Germany is the country that all other countries should aspire to
become, especially one that also starts with G and rhymes with cheese”
Motivation
questionnaire respondent #222
Figure 1; "And
here you will find a thrilling set of likert scales where you can rate your
assessment of the different parties' performance at the televised election debate"
Googleforms has saved my cheese
with questionnaire work, but it has also
showed me, on the other hand, how much effort and wasted time lies behind
questionnaire based research when one hasn’t got a research assistant – also
known as minion who does all the legwork for you. I cannot deny though that I
find completing my dataset a rather rewarding task filled with pearls of wisdom
such as the above response to why respondent #222, quoted above, chose to take up the language
– thankfully, this student also included a more serious answer, but the cheese
rhyme (really?) concept had to be shared and seemed like a fitting prelude to
this post.
When I decided to use
questionnaires one of the things I was clear about was that our students are
over-surveyed and that there may be much dismay in having to fill in yet one
more form. But, much to my surprise, not only did I get quite a good
participation rate – bearing in mind the German team were excellently pushy, I suspect,
at getting their students to complete my questionnaire, seeing the completion
rate – but also a number of enthusiastic and candid responses to the open
questions. These open questions required students to explain, in one line,
their motivation to study their degree & their motivation to study a
language. What many seem to have done is to pour their heart out, to show their
passion or even their doubts about the subject in such a way that it almost seems
they were grateful to be asked about the topic. I did have an inkling about how
the questionnaire piqued their interest after some students stopped me at the
end of the class, once questionnaires had been handed out and completed, to
learn more about the research. Interest was clear among all the language
teachers who volunteered to help; I had to almost pry the completed
questionnaires in closed envelopes from some since they wanted photocopies to
see directly what their students had said.
Now, back to Googleforms. Handing
out the questionnaires in printed form required much photocopying and
preparation from me (apologies to the many trees that go into research), and a
lot of training other teachers to make sure information was delivered
adequately and to avoid any “forced completion” from students. After three
weeks of madness, the task was completed. But I was unable to reach all courses
and groups, so a form was designed online with Google tools, and then forwarded
to students that had not been reached in person. The completion rate was, as
could be expected, extremely low. But I discovered, though, that such
questionnaire tool generates an Excel form and reports on the data (not
necessarily doing all the hard work for you, but allowing to have an initial
overview of numerical data with pie and bar charts).
The Excel form is nowhere near
the finished product, this will have to be transformed into a more suited
Excel, or, directly, an SPSS form, which can be “read” by the statistics
programme, but it is one first step in automatized data collection and,
importantly, can be accessed from anywhere and is safely stored in the cloud.
Anyway, not many students completed the form, but it was still useful as I used
it to input all the data by completing the questionnaire, myself, over and over
again, with the answers from the paper forms. Repetitive? Yes, indeed, but at the same time essential
to avoid extra mistakes in data input as most of it could be just done with a
click. And faster, of course.
It is the first time I design a
questionnaire, hand it out and transform it in a dataset, so I am starting to
see how my process is definitely faulty (in Spanish we say “hace aguas” i.e.
like a badly designed boat that allows water to seep through). I needed to
understand how I am going to test each part of the questionnaire, above all the
numerical part, before designing it, but I found it impossible to understand
which statistical analysis would be required for an inexistent dataset, so,
more or less, took the bull by the horns and got on with data collection.
Perhaps a mistake, but by typing information over and over again certain things
are becoming clear: possible labels for the analysis of qualitative data,
patterns in the numerical data, clear differences between languages and
questions that can be elicited from the dataset which I had not thought of
earlier. I am also starting to fear that I will end up having too much data in
my hands.
Did I mention there is a
qualitative part to this? I have so many volunteers (86 out of 400, oh god 0_o) that I do not know what to
do with myself, or with them, frankly. I need to elucidate a “fob off” email to
let them know that I will be in touch soon, I mean, later, as need to first
complete part 1 of the study before getting immersed in the interview part, and
then see whether I could just do the interviews, record them, and do nothing
with them until the second part of the assignment, were there to be too many.
So, does Germany rhyme with
cheese? I am not sure. Should I have just removed this questionnaire from the
lot? Not really, as it makes for a great anecdote which am sure my German
colleagues will appreciate when I share the preliminary results with all in
June. Watch this space though, as serious work needs to start soon with
understanding how to question the data in statistical terms without putting my dainty
hoof in it.