How do you go from a fivehundred-page research report to a brief newspaper article? Each year, since 1986, the SOM-institute has published their findings and results in a large volume. The results are however not only meant to stay inside the academic ivory tower, the SOM-institute actively circulates them, and they appear in public debate on several occassions. You could of course read the report for a couple of days, then write a summary or a review. But it is much more convenient to have the results of the report summarized and explained for you by someone else. And along the same lines, if you want your research to circulate outside the report, to reach out to people who do not have the time or means to spend a couple of days in the library, as a researcher you need to translate the numerous words, graphs, tables and conclusions to a compressed yet credible statement.
One such moment of translation are the press conferences that SOM has held since XinsertyearX. I visited one in XinsertyearX in Stockholm, and one in Gothenburg in 2011. The latter one I recorded and analyzed with translation as a focal point.
When you enter the press conference you get a copy of the fivehundred-page report handed out to the audience of about 25 people, most of them academic researchers and reporters. The public service TV broadcaster Sveriges Television is filming the event, and on the whiteboard the Twitter hashtag #somgu has been written. As the second largest city, Gothenburg is not considered to be the epicenter of media impact, so the conference takes place in an ordinary lecture hall at Annedalsseminariet, where some of the social science departments are based.
The press conference is opened by the three editors of the report, Lennart Weibull, Sören Holmberg and Henrik Oscarsson, the two former introduced as the co-founders of the institute. At first, Weibull presents how the survey was made, while referring to the report that was handed out while people walked in the room:
Here we have everything. The SOM-institute is a scientific institute where we work extensively with methodological developments /.../ Thus, the sample is 9000 and make three emissions and [of] questionnaires. One is more [focused on] political, one more on media and culture, one a bit more on life-styles and health. Our base questions in the SOM-survey are in all three of these questionnaires. This is not something you need to know, since it is all here [in the report]. But if one is interested too look it up [more closely] /.../ from page 595 and onwards you have the three questionnaires in extenso documented in the book. (my italics)
Weibull summarizes how the survey was made, but the details are too exensive to give in a two hour seminar, so they are referred, or linked, to a particular page in the report. To get to the questionnaire, you need to go one more step.
The press conference continues with short presentations of each of the chapters in the report. The audience learns that levels of trust in political institutions are on the same high levels as during the seventies, that political interest increases close to elections, that Swedes have more positive attitudes to immigration, that women are more active in social media, and that people living in rural areas far away from the center of decision making are more skeptical towards wolves in the forests than people living in urban environments. Every now and then, especially when a number or graph is quoted, the report is once again linked with statements such as "As you can see on page X in figure Y". Some of the descriptions are general and some take a more technical turn. Sören Holmberg for example describes new techniques for measuring "job performance" for evaluating how public institutions are perceived:
We have been inspired by American research on consultants and politics when we made our measurements, our financial ratios, as [presented] on page 109. It is [called] name recognition, you have to know the ~~SAK~~FÖRETEELSE to be able to associate it with values. Secondly, evaluation; not of trust this time which we measure in other instances, not with personal satisfaction, but the evaluation of job performance, how you perceive that the job you are expected to do, how well it is done. That is called job performance in American [English]. (The words "name recognition" and "job performance" appear in English in original)
The results, on page 109 and narrated by Holmberg, are quite devastating for two of the institutions that were measured. The Försäkringskassan (Social insurances) and Arbetsförmedlingen (Swedish Public Employment Service) indicate low job performance.
Another interesting finding is presented by John Magnus Roos, a researcher at the Centre for Consumer Science at Gothenburg University. He argues against the widespread belief that so called "shopoholics" are mostly women buying purses, makeup and clothes. On the contrary his data shows that gender is not an important factor. Rather the shopoholics are people young who are dissatisfied with life in general, and that their degree of empathy is lower than average. The following day, this is reported by the local public service radio station P4 Göteborg where Roos is interviewed by the reporter Anna Olofsson, who attended the seminar. The radio station publishes an interview on their website the following day (Olofsson 2011):
- If we know more about the personality type [of the shopoholic] then we can both prevent these problems and help the person in need of support, says the researcher John Magnus Roos.
Thus, from this article, it is possible to "reverse-engineer" a widely circulated fact, back to a press conference, which in turn refers to a scientific report. But to go further, to translate from the easy reads of news media to the esoteric science, we need to go to the reports. Or, we have to open more black boxes.
To continue reading about black boxes, please go to the previous post