Whilst the variables of each research project are carefully considered and calculated to ensure the smallest margin of error possible, the one variable that remains a struggle to control is the respondent (as much as we might like to). The fundamental homogeneous groups that, we can define within our society, are changing.
Impacts of immigration…
The fact is that sections of society are simply not as clean cut as they used to be, and that can be a big problem for Researchers. Over the last 50 years it has become more and more difficult to ‘define’ a person’s cultural origin (for the purposes of a research study) as both cultural lines; and definitions of heritage, become blurred. This of course is representative of the way the world is changing; immigration has become a worldwide phenomenon, and changed the face of Britain. Obviously none of this is news, immigration didn’t just happen over night. However, overtime the widespread effects of this cultural paradigm shift have become clear, and for Researchers it has fundamentally changed the way we conduct research.
One particular challenge that has arisen is that it is becoming increasingly difficult to define a group culturally for research purposes. What a respondent may consider to be their heritage, or cultural class might just be their personal interpretation. A confusing concept, but simply put it’s exceptionally difficult to define someone’s culture, for any purposes, if the person themselves has no clear idea of what cultural group they actually belong to. Therefore meaning that for research grouping purposes it is impossible to classify a person’s heritage definitively and as a result to establish the variables for any project.
Unfortunately you only contribute further to the wide range of errors that must be accounted for when you begin a study. Aside from any cultural confusions that my arise, the biggest pitfall of any study is the issue of human error. In an article for Fast Company Douglas Van Praet acknowledges that “The elephant in the room [is] that the vast majority of decisions are made unconsciously”, through which he is implying that it is not possible to rely on a person’s response when answering survey questions. This ultimately means that for the average research study we cannot determine whether a person is telling the truth at any given point in time, therefore rendering the data accumulated based s great deal upon good faith in respondents and potentially inaccurate.
It seems, therefore, that the ‘changing face’ of our society is one factor that is making it harder for Researchers to conduct their jobs, as it makes data increasingly inaccurate. These problems are, however, a drop in an ocean of uncontrollable variables and as Van Praet points out, the issues within the Market Research industry are far larger than this, “The conscious mind is simply not running the show, but we’ve created an entire industry pretending that it does”.
So what can we do about it?
Fundamentally we just have to accept that good faith is part of the territory, and its all we can do as Researchers to trust that the respondent is telling the truth. Instead we should take more steps towards informing clients that all studies do come with a margin of error, and the development of homogeneous groups is just one factor which could potentially have an impact of their data. We should start to help clients to write questionnaires or studies which have a direct aim of tackling these cultural anomalies, and ensuring that these are kept to a minimum if possible. Steps could also be taken to inform respondents about cultural classifications so that they are as well informed as they could possibly be. Ultimately however researchers must be prepared to go back into the field and explore the culturally associated anomalies which initial research unearths, in order to ensure that such anomalies can be avoided in the future.
How do you think Homogeneous Groups are changing the way we do research? Do you think they will hold a significant impact on future studies? Let us know your thoughts here: