Rewrite the Headlines: “Dr. Dillner’s Dilemma”
Over the first few weeks of February, we’re showcasing the top entries in both the school and undergraduate categories of our Rewrite the Headlines competition (full list of winners here). We continue with another of our commended undergraduate entries, from Emmi de Vries at the University of Glasgow.
“Dr. Dillner’s Dilemma” Or: The Challenges of a Practical Application of Research Findings
There are many ways to present research findings in the media. One of them is demonstrated by Dr. Luisa Dillner, who solves supposed ‘health dilemmas’ for the Guardian. Rather sensational sounding headlines invite the readers to learn more about how current research findings are relevant to their lives. One such article recently set out to answer the question “How much sex will make me happy?”. Dr. Dillner’s solution: if you are in a relationship, having sex once a week will make you happiest. She bases this claim on a study conducted by three researchers from the University of Toronto Mississauga. The research paper “Sexual Frequency Predicts Greater Well-Being, But More is Not Always Better” by Muise, Schimmack, and Impett, published in the journal Social Psychological and Personality Science, is composed of three smaller sub-studies that analyse survey responses from over 30,000 participants in total. It overall concludes that there is a curvilinear as opposed to a linear association between sexual frequency and well-being for people in romantic relationships. Several very good summaries of the three studies and their individual findings are included in the research paper. However, the media report neglected to differentiate between the different studies and only reported on the overall findings.
The importance of being precise
Properly defining concepts is the be-all and end-all of any research or presentation of research findings to avoid bias and exaggeration and thus intentionally or unintentionally misleading the reader. Sometimes a media piece muddies the waters by way of (over)simplification and sometimes the source of the problem can already be located in the research itself. Or as in the present case: both.
While the media article pretty much stuck to the term happiness, the research paper used the terms happiness, satisfaction with life, and well-being somewhat interchangeably, but neither of them gave a proper definition of the concept employed. Only in sub-study 2, where the satisfaction with life variable was used, did the researchers give a more detailed description of what the concept included. In sub-studies 1 and 3, the surveys asked very board questions about general happiness (without further explanation) and participants were asked to rate their happiness level on a scale. This approach alone is prone to a variety of different interpretations but seen in combination with the fact that some of the survey data is from as far back as 1989, the results are unlikely to be comparable and thus lose some of their validity and reliability.
A similar problem can be found with regards to the variable sexual frequency, where it is not clear what counts as sex, and the use of the term romantic relationship, which would be more accurately described as heterosexual married couples based on the available data.
The problem of correlation vs. causation
Putting aside the inconsistency and vagueness of the definitions, the research paper does reach the conclusion that there is an association between sexual frequency and well-being for people in romantic relationships. However, as made very clear in the research paper, this does not indicate causation either way. It could not be verified whether more sex makes you happier or if you simply have more sex when you are happier. While this is mentioned clearly in the bottom half of the newspaper article, many expressions used in the piece (“having it more often won’t increase your happiness”, “it will make you happiest if you have it once a week”, and “more than once a week didn’t increase their happiness”) still blur the lines of correlation and causation and hint at more causation than can be inferred from the research findings. A better way of phrasing the findings might have been to state that couples that had more sex reported greater levels of happiness. It should also have been included that rather than a direct relationship between sexual frequency and well-being, the researchers found an indirect relationship between the two mediated by relationship satisfaction. This appears to be an explanation for the finding that the discovered association only applied to couples.
The art of putting findings into the right context
No study stands alone, so thorough research is essential to put the findings of any study into the right context. This was really well done by the authors of the research study but less so by the author of the media piece. Some of the research from the study was used in the news article but it was presented out of context and not properly tied in with the argument of the media report. Similarly, there was mention of another news article about the optimum amount of sex for pregnancy that was not relevant to the main question of this particular news article. Trying for a balanced and comprehensive account of an issue is commendable up to the point where the additional information raises more questions than it answers and does no longer contribute to the overall argument.
The verdict on Dr. Dillner’s solution to this ‘health dilemma’:
It is an interesting idea to make research assessable to readers by giving it relevance with regards to their own lives rather than simply informing them about it. Unfortunately, however, research does not always allow for the definite and clear answers or advice that a question such as the one presented in Dr. Dillner’s article lets the reader hope for. Therefore, despite a clear effort to report the study’s findings accurately and balanced, the format of the article, as well as the slightly sensational headline that wakes unrealistic expectations in the reader, keep the media report from succeeding completely.
However, as demonstrated above, some of the difficulty in this particular case was also due to a lack of clear definitions and the choice of data sources in the research study itself. Despite evidencing extensive research into the subject and great care in detailing the research procedures, the fundamental definitions underlying the research were not clear and consistent enough and thus the media report is not singularly at fault here for being potentially misleading.
Emmi de Vries, University of Glasgow
Dillner, L. (2015) How much sex will make me happy?. Available at: http://www.theguardian.com/lifeandstyle/2015/nov/30/how-much-sex-will-make-me-happy
Muise, A., Schimmack, U. and Impett, E. A. (2015) ‘Sexual Frequency Predicts Greater Well-Being, But More is Not Always Better’, Social Psychological and Personality Science. doi: 10.1177/1948550615616462
Prizes were supported by the British Academy, the University of Strathclyde, the School of Chemistry at the University of St. Andrews, the School of Social Work and Social Policy at the University of Strathclyde, the University of Dundee School of Life Sciences, and the Particle Physics Experiment Research Group at the University of Edinburgh, the Social Research Association, the Scottish History Society, and Palgrave Macmillan.
Competition details can be accessed at https://researchtheheadlines.org/rewritetheheadlines, and the full list of winners is available here.