Transparency

The principle of transparency in research refers to the practice of being open and honest about all aspects of the research process.

In the TV show, Great British Bake Off, contestants compete in a UK-wide baking competition. In some tasks bakers are assessed on their creativity and flavour combinations, but in the show’s ‘Technical challenge’ they all have to follow the same recipe and make the same thing. Simple, right? Well, if it’s so simple, then why are there so many soggy bottoms?

Technical challenge instructions:

  1. Make the cake batter
  2. Bake the cake
  3. Ice the cake

Even when a recipe is provided, there can be a lot missing. In the technical challenge above, the instructions do not specify which ingredients should go into the cake batter, in what order they should be added, or how long the cake should be baked. A recipe for disaster!

A similar issue occurs in many academic research papers. Researchers describe what they did, but often not in enough detail for someone else to clearly understand what was done, or to repeat the study themselves. This means that another researcher hoping to reproduce the study may not able to reproduce it accurately. This is referred to as a study’s replicability.

Similarly, why are there so many amazing looking cakes on Pinterest, which become ‘Pinterest fails’ when people try to make them at home?

The image has two panels each of which shows a cake in the shape of a cat. The cat cake in the left panel has a solid shape, and the cat has a well defined face with blue eyes, a pink nose and an upturned mouth. The icing has been shaped to resemble fur. The cat cake on the right shows less technical skill. The two eyes are looking in opposite directions, there is no nose, and the mouth is downturned. The cake looks like it has melted and the overall appearance is scary rather than cute.
Pinterest cat cakes: when the reality doesn’t quite match up to the expectation.

Sometimes it might be because the person sharing the beautiful picture of their cake tried thirty times before one turned out right, but fails to mention this, and instead tells readers how easy it was to make!

Similarly, researchers can fail to disclose versions of studies that they have tried, but that in their view didn’t ‘work’, choosing instead to share results from those studies that had the results that they desired.

What are all the things you could include in a recipe to give the baker the best chance of success? Make a note in the box below. Try to think of at least five before revealing the answers below.

To use this interactive functionality a free OU account is required. Sign in or register.
Interactive feature not available in single page view (see it in standard view).

When you are ready, press 'reveal' to see our comments.

Discussion

The details will differ from person to person. In your answer, you might have included things like: pictures of each step, temperature of the oven, ingredients, length of baking time, size of tin, order of adding ingredients, how long to mix, and possible substitutions. It is likely that the details you included are those that will help determine whether your cake looks like the cat cake on the left, or the cat cake on the right. In the same way, researchers are more likely to be able to get the same result as other researchers if they can follow clear instructions on how to conduct the study.

Spotlight: the Pelagios Network

Integrity