Skip to 0 minutes and 6 secondsNICK BARTER: Good evening. And welcome to Research Matters News. I'm Nick Barter. Our top story tonight is the price of sugary drinks. Researchers have confirmed what many doctors had long suspected, few of us would buy fizzy drinks if prices were raised. The report's lead author hopes these results will add weight to calls from health professionals to put a tax on sugary drinks. Chris Stevenson, our reporter in the field, has the story.

Skip to 0 minutes and 36 secondsCHRIS STEVENSON: It's the latest salvo in the war on obesity. In Australia, we buy almost enough fizzy drink to make one can per day per person. Sugar sweetened beverages make a big contribution to obesity in Australia. The question for governments and health professionals is, how can we encourage people to make healthier choices? Researchers at Deakin led a study that showed if you put up the price of sugary drinks, it does lead to people making healthier purchases.

Skip to 1 minute and 10 secondsMIRANDA BLAKE: The price has increased by 20% on the sugary unhealthy drinks for 17 weeks. And we watched to say what happened to sales of those drinks and healthy alternatives over time. Basically, what we saw is that sales of those unhealthy drinks fell by about 28%. And sales of healthier alternatives has increased at the same time.

Skip to 1 minute and 30 secondsCHRIS STEVENSON: The research comes as calls for the introduction of a health levee in Australia, similar to the one in the United Kingdom, come from leading health bodies like the Australian Medical Association, Diabetes Australia, and the Cancer Council. However, these calls have been rejected by the federal government. And industry bodies have responded by saying that a tax will never work. The Australian Beverages Council says, "It's disappointing that in 2018 with both government and opposition reject any idea, that the AMA continues to promulgate a tax as a solution to the nation's expanding waistline. This type of measure lacks any evidence anywhere in the world that it has any discernible impact on public health."

Skip to 2 minutes and 14 secondsSenior research fellow, Kathryn Backholer, has heard all this before.

Skip to 2 minutes and 19 secondsKATHRYN BACKHOLER: Industry, in the area of food and nutrition-- industry is a really strong player. Unless you have evidence to support policies, they will do everything they can to try and resist public health efforts. So we need that evidence to counteract the industry's resistance to public health measures.

Skip to 2 minutes and 37 secondsCHRIS STEVENSON: However, there are some changes in the pipeline. Retailers in public health and sporting organisations are leading the charge themselves by voluntarily increasing the price of sugary drinks or eliminating them altogether. This research reassure them that while this may cost a little in anticipated growth, they shouldn't see any drop in overall revenue.

Skip to 3 minutes and 0 secondsKATHRYN BACKHOLER: I guess one of the other important points to make is that total sales remain the same. So the retailer profits, essentially, didn't change, which is good from a business perspective.

Skip to 3 minutes and 11 secondsCHRIS STEVENSON: The research also showed that hardly any of their customers, even noticed.

Skip to 3 minutes and 17 secondsMIRANDA BLAKE: So amazingly, the retailer estimated that 15% of customers had no risk intervention. And 15% of customer had that risk intervention, according to the surveys.

Skip to 3 minutes and 25 secondsCHRIS STEVENSON: Ultimately, Miranda and Kathryn hope that their research will add weight to the growing calls for a levy on sugary drinks.

Skip to 3 minutes and 31 secondsKATHRYN BACKHOLER: We certainly support these measures, because there is clear evidence now that if you increase the price of a sugary drink or an unhealthy product, you will reduce the purchase of that product at a population level.

Skip to 3 minutes and 42 secondsCHRIS STEVENSON: The federal government continues to resist the call. But with the mounting evidence, it's anyone's guess how long they can hold out. Deakin and Griffith University Research News.

Building the picture

Let the data speak for itself.

This is a very common maxim among quantitative researchers. It means that you should:

  • choose the right test for your data
  • arrange your data in to an accurate picture
  • make sure your presuppositions don’t cloud your interpretation of the data.

Once you’ve collected your data, you will need to arrange it so that it makes sense. It’s at this point that you’ll need to decide exactly how you’ll test your variables and present the results.

In quantitative research, the simplest explanation is the best one. You should try to build an accurate picture with as simple an explanation as possible without compromising the integrity of your analysis. This well-established scientific principle is sometimes referred to as ‘parsimony.’

In practice, parsimony usually means:

  • going with the simplest possible explanation when describing your results
  • undertaking only the most relevant statistical tests to answer your question.

The more tests you do, the more likely you are to find a result that confirms what you’d been hoping to find. If you’re an untrained statistician, you risk falling victim to cognitive bias – presenting a picture based on the test that gives you the results you expected to find rather than the most appropriate one – without even realising that this is what you’re doing.

To avoid this, you should find the most appropriate statistical test(s) to demonstrate what’s happening with your data and stick with them. We’ll look at how to do this in greater detail during Week 2.

What happened in the convenience store?

So, what did Miranda’s data tell us?

results graph

Image credit: (Blake et al. 2017). Select this image to access as a PDF

The graph shows us an immediate drop in sales of unhealthy red drinks, eventually falling by 24% by the end of the trial. The sale of amber drinks also fell by around 18%. Meanwhile, the sale of healthy green drinks rose by around 35%.

While those results are strong enough on their own, the researchers knew something was missing. It’s one thing to say that sales dropped, but what if they would have dropped anyway? Miranda and the team went to consult a statistical expert to help them think about how they could have predicted sales in that period. That expert was none other than your lead educator, Chris Stevenson.

After this consultation, Miranda and the team were able to produce counterfactuals to predict what would have happened if the trial had not gone ahead, based on previous sales from the past two years. When you take these into account, the findings are even stronger:

  • Sales of red (unhealthy) drinks decreased by 27.6% from expectations, based on sales trends.
  • Sales of amber drinks decreased by 26.7%
  • Sales of green (healthy) drinks increased by 26.9%

Your task

Letting data speak for itself is just the first step of evidence-based practice. We also need to do something with what we’ve discovered.

  • Think about one or two brief recommendations you could give to decision makers based on the evidence from Miranda’s research findings.

  • What recommendations might you be able to make based on the research project you’ve been fleshing out over the previous few steps?

Share your description of the proposals and reflections to the group.

Reference:
Blake, M. R., Peeters, A., Lancsar, E., Boelsen-Robinson, T., Corben, K., Stevenson, C. E., Palermo, C. & Backholer, K. (2017). Retailer-Led Sugar-Sweetened Beverage Price Increase Reduces Purchases in a Hospital Convenience Store in Melbourne, Australia: A Mixed Methods Evaluation. Journal of the Academy of Nutrition and Dietetics. Figure 7 from this article was reproduced above with the permission of the Academy of Nutrition and Dietetics.

Share this video:

This video is from the free online course:

Why Numbers Matter: Quantitative Research

Deakin University