"Research is not truth, but an ongoing conversation”

Research should not be seen as individual truths, but as an ongoing conversation and a piece in the puzzle that contributes to the whole. The value lies in the complicated, completed picture, says Martin Hällsten, Professor of sociology.

Martin Hällsten.
Martin Hällsten. Photo: Elin Sahlin/Stockholm University

In his research, Martin Hällsten has studied everything from social networks, unemployment, hard rockers, integration of immigrants, educational choice, educational inequality and lifelong learning, but most of his research is about mobility and the importance of family background when it comes to inequality.

But it was not a matter of course that he one day would become a professor of sociology. In high school, he had studied science but was encouraged by the social studies teacher to study sociology because he was "excessively interested in society".

“I had not heard the word sociology before, but I borrowed Anthony Gidden's book "Sociology" and took it all in, it was impossible to get enough. Then I read the three basic courses in one go here at Stockholm University,” says Martin.

What was it about sociology that attracted you?

“Trying to understand. I don't know if I succeeded. Maybe marginally. As a young person, I wanted to try to understand the world around me, why things are the way they are. It is actually a big disappointment I guess – that there is so little that we understand. But you can always add pieces to the jigsaw puzzle.”

 

Tracing inequality back to past generations

Today, his research focus is primarily within historically-oriented mobility studies and on socioeconomic persistence. The focus is on tracing inequality back to previous generations, to see if living conditions and life chances tend to run in families in Sweden. He says that in the past there was not enough data to be able to connect more than two generations, and many believed that there was no point in trying to trace inequality further back in time.

“After a while, when there were enough data series and it became possible to link grandparents, people realized that maybe there was something there after all. Then you get curious, let’s say we trace this all back what happens then? How much inequality will we find if we go back very far? Martin Kolk from Stockholm University’s Demography Unit and I have been able to find persistence of socioeconomic conditions seven generations back using data from church registers.”

In the project “Shadow of a peasant past: the impact of past generations on living conditions in Sweden” (also including Per Engzell from University College London, UCL), they use historical census data from 1880 to 1959 and link them to modern registers from 1960 to 2020 to obtain data for the entire population.

“We look at the importance of family background in a lot of different dimensions and class is one of those dimensions. We also want to see whether occupation, education, wealth and such things are inherited.”

 

Counted research money

After studying sociology, a series of coincidences led him to employment at age 21 as research assistant at SISTER, the Swedish Institute for Studies in Education and Research, a research institute that had funded much research. There, his job was to count research money distributed to various areas and to study differences in research careers between men and women, for example, who receives money from the research councils. In 2003, he found himself at the Swedish Research Council's analysis department, where he continued to count research money. It had been shown that research money in medicine did not go to women to a particularly high degree in the 90s. Even more than ten years later, it was a very sensitive topic, says Martin.

“The medics at that time were very uninterested in this being brought up so we were forbidden by the General Secretary to work with this. Today, there is a political commissioner in each preparation group who is supposed to see if there are any gender dynamics in how you talk about applications, we have come a long way since then.”

At the same time that he was working at the Swedish Research Council, he studied the three basic courses in statistics, something he describes as the perfect complementary subject to sociology. He started to write his master's thesis on overeducation, which was never finished because he got a job as a research assistant and then as a PhD student at SOFI, The Institute of Social Research at Stockholm University.

“I've had some luck along the way,” he says.

The draft of the master's thesis is still in the drawer.

“The idea for the thesis might not have been the best, although it was good enough for a master's thesis. I still have a thought that I will tease people and put it forward eventually, maybe sometime when I retire or something…”

 

More self-censorship is needed in research

Good research is good craftsmanship, says Martin. It is important to be able to recognize when you have to stop a study, perhaps because the research design is flawed, or because it is not possible to answer the research question based on the available data.

“It is important to think about whether you can answer your research question. If you realize that you can't, you have to do something else and not just keep pretending. I can sit and get an idea somewhere, maybe I see something on TV, hear something on the radio or sit in a seminar and think ‘I should do that, I could analyze that’ and then I realize that it won't work because we lack data on this, or this process is too complicated. Then you are a little surprised when someone else manages to answer that question that seemed impossible to answer. Some people do not have the same self-censorship,” Martin says, referring to the fact that there is a wide range of quality in research.

Martin Hällsten.
Martin Hällsten. Photo: Elin Sahlin

A dilemma that many researchers face is that the research career to a large extent is about counting the number of publications in scientific journals, which in turn can sometimes create the wrong incentives for research, says Martin. Another challenge is when research is to be communicated to the public, as there is a risk that the research will be distorted when journalists pick out individual parts into a different context than was originally intended. When individual studies are presented as “truths," the results can easily be misinterpreted, he believes.

“Research is not truth, but an ongoing conversation. At its best, a lively debate. If you think that you will find a fundamental truth in every single research work you do, then you cannot work with this. Because everything we do is temporary, it's something you have to learn how to live with.”

At the same time, Martin thinks research communication is important and is himself involved in Center for Business and Policy Studies (Studieförbundet för Näringsliv och Samhälle).

“It's a bridge between academia, public decision-makers and business to spread knowledge to them, and vice versa, and it's a model I think works. You write a short report, they get an easier-to-read text and you have a seminar and a discussion around it. This way, you are able to avoid the simplification  that is otherwise easy to end up with.”

 

Pieces of a puzzle 

The important thing is to have a holistic perspective and to be aware that all studies have their shortcomings, while at the same time they can contribute to greater knowledge.

“To understand the whole picture, you need to see each study with its pros and cons, that piece of the puzzle fits into a larger context and contributes to this ongoing discussion. It requires not reading too literally and seeing research as temporary – and that is a very big challenge, I would say. It’s essential to see the complicated, completed picture,” says Martin.

In the same way, it is important to make sociology students aware of the complexity of theories and research results, he believes.

“We present a lot of things to the students that are not true. We teach them to analyze results based on different theories. But that doesn't mean it's true, it just means it's something you have to learn to do. In this way, we train the students to become like us, that is, to both be able to question but also to see the whole puzzle.”

The complexity of the research also makes it difficult to describe to outsiders.

“One problem with what you do is that you can't talk about it at a party – it's either too trivial or too complicated to work that way. But people often want to talk about it at a party, so you have to refrain from going to so many parties,” he says with a laugh.

Text: Elin Sahlin