by Weronika Kajdanska
Everyday we come across an infinite amount of information: news, facts, numbers. But have you ever thought about where all this knowledge is coming from?
Magazines and newspapers draw the readers’ attention by using these sensational headlines like “Chocolate can fight depression” or “It was proved that more than 200 Polish women do their makeup while sitting in the car”. Reasonable question arises: ‘How do they know?’
Everything that captures people’s attention and curiosity is being studied right now. They are not only general questions but more often precise cases we face everyday, like “Can ice-cream be considered as a comfort food”.
There is a complicated survey and a lot of calculations behind each answer we get. And the driving engine of each research is a well-designed questionnaire.
Have you ever filled one of those? The sheet of paper you would love to get rid of as quickly as possible? Everybody did and It always seemed like it’s a waste of time. But in practice it is not. Furthermore, it’s a crucial part of the survey. Bad questionnaire can lead to drawing the Wrong Conclusions and ruining the data.
There is a number of rules regulating the form of questionnaire. The amount of research questions is limited, wording and order of questions matters, deciding on the information required etc. But what is more interesting and worth focusing on are the answers we see under each question. Those are called measurement scales. Basically it is the spectrum where we can fit all the possible answers.
Description – Order – Distance – Origin
Those are main qualities of the answers.
Scaling that is used can gain more power and meaning by adding each of those characteristics.
People are grouped together according to some characteristic. (Gender: male-female)
Answers can be sorted from low to high. (level of education: primary-secondary-bachelor-master)
There are same ‘distances’ between the answers (Attitudes: disagree-neutral-agree. There is a symmetry in such an answer)
There is a natural 0 point. The answer can’t assume negative values. (Age, incomes)
So let’s take a look at some examples.
The company’s question is: “Which brand of chocolate is more successful on the market?”. The answers are the names of the brands. There is definitely a quality of description (you can check your favorite product from the list) but what about the order? Can we put Milka higher than Ritter Sport? And how to deal with the distance and a zero-point?
Only one quality could be applied to such question which means that there is not much that can be done with the information we will receive.
Another example can be the question about the incomes of the respondents.
Description is present, we can sort the answers from low to high, distances can be assumed as same (we have units of measurement) and there is a starting point – 0. All four qualities are present which means that there are a lot of mathematical calculations we can use.
Not all the information we get from the researches is correct though. If the questions are designed with poor knowledge about it – the research will fail. If the group of target respondents is defined wrongly – again, the research will fail. Rational results require good knowledge.
This information is useful for analysts and researchers but there is one fundamental conclusion that needs to be drawn. Details matter. Everything that was written or said should be on a firm ground.
Even such common and unoriginal thing as questionnaire has a science behind it. We take for granted things that once were complicated and non-understandable.
People’s interests lead to a progress. So next time you ask a question – make sure you question the answer.