I’m sure, just like me, you’ve taken part in a quiz and thought “that team seems to be doing surprisingly well…”!
With the rise of Zoom quizzes during lockdown, many of us will regularly be competing in online quizzes or have even stepped up to the role of quiz master ourselves. One of the interesting aspects of quiz-mastery is being able to run a fair quiz and making sure all teams play within the rules.
But that is easier said than done! Particularly when people who sign up to competitive events are often naturally highly competitive themselves.
The Status Defence (a sense of entitlement based on your self-worth)
The Bonding Defence (your sense of right and wrong based on the culture around you)
Leveling the Playing Field (the idea that you are at a disadvantage and being unfairly treated versus everyone else)
The list goes on and on and, although incredibly interesting, I’m not going to go into the many reasons behind cheating here – we’re going to look at how to spot and stop it in your own quizzes!
In March this year, I decided to run my very own online quiz during lockdown to keep me entertained in those long, drawn out, days. Luckily for me, it turned into a very popular Saturday night event where I would regularly have over 25 teams signing up to take part in what I called House of Quiz.
After around eight weeks of running the quiz, one team’s score began to look like an outlier. At this stage it’s worth saying that that is not uncommon: teams can have one-off incredible performances simply because the stars align and their particular knowledge set serves them well on that day. However, that outlier very quickly started to repeat itself and it raised suspicion in House of Quiz HQ. So it got me thinking about what I could do to favour the honest players and neutralise foul play.
After lots and lots of thought and planning, I came up with 5 main strategies:
#1 Set the Ground Rules – No Cheating Allowed!
It sounds simple doesn’t it? But simply creating that self-awareness at the beginning of your quiz can be all that teams need to police themselves.
Unfortunately, it is not the complete the solution. Telling people what they can’t do can often spark the opposite response you want (for those of you with teenage children you’ll be all too familiar with this) and they actively work against you to regain choice and feel like they have taken control of the situation. So further action is required!
#2 Quiz Masters are the Quiz Markers
There are some great blog posts out there that guide you on how to run your own quiz. What I was surprised to see was the number that suggested you ask the teams to mark themselves and let you know their scores at the end of the round! Maybe I’m overly pessimistic, but that’s a level of trust I simply don’t have.
When the quiz master does the marking, it avoids the issue with borderline answers. For example:
Question: The Weeknd reached Number 1 in the UK charts with which song in February 2020? Correct Answer: Blinding Lights Incorrect Answer: Blinded by the Light
Blinding Lights is the correct answer; Blinded by the Light is close but is in fact a very famous 1973 song by Bruce Springsteen. The Quiz Master can apply the same fair judgment to all teams to make sure that only those that hit the target score the points.
Credit: Casito, Cropped, CC BY-SA 3.0
Having the Quiz Master mark also means that he or she is in control of the scores. If teams are asked to tell you their scores, the competitive spirit can often get the better of them and they are able nudge their score up by a point or two with justifications like “we had the correct answer written down but changed our mind”! I’ve seen it done before and I’m ashamed to say I’ve done it once or twice myself!
#3 Keep the Rounds a Mystery
This is the very essence of grey-area cheating. To build anticipation you might want to send your round titles out in advance or even just give them all at the beginning of the quiz.
It seems like a reasonable thing to do! But, by providing that little bit of extra information, teams can read up on your rounds, memorise answers or even make notes. They might justify it as revision because they haven’t seen the questions, but I call that plain old cheating!
Keep every round a mystery until each round starts and this element of cheating is eliminated.
Credit: Haradhi, CC BY 2.5
#4 Engineer your Questions
My most effective approach to stop cheating is to design your quiz so it’s actually a disadvantage to look up the answers. How do I do it? You increase the complexity of the questions and introduce a time element.
A perfect example of this is the final round of all my quizzes which I call the House of Quiz MashUp. Each question has a picture and a general knowledge element, the answers to which need to be combined into one single answer. Also, a maximum of half the teams can score when they send in their correct answers – so it’s also a race!
In the example above, you need to identify that the picture is of some Rhubarb, know that Bridgetown is the capital of Barbados, combine the answers to give Rhubarbados and send your answer quick enough to be in the top 50% of teams! So the design of the question means that you simply need to know the answers because if you have to Google any of the elements (which is tricky with pictures) you’ll already be too late!
Not every round needs to be engineered, so you can also consider doubling the points on your anti-cheat rounds so they are weighted in favour of fair players. This can lead to some very revealing scoring which brings us nicely onto my last strategy…
#5 Analyse your Data
The first four strategies are ways for you to prevent cheating but data is how you can spot it. Make a simple spreadsheet that you use to enter scores so you can sort, filter and chart your data any way you like and see if a team’s score is out of the ordinary and difficult to explain.
Earlier on I mentioned a team that was consistently becoming an outlier in my own quiz. The figure below is a visual representation of what was happening. What it shows is the team’s score (I’ve called them Team X) and how they compare against the group mean and mean of the Top 5 teams (discounting them if they were in it).
On the far left you can see how Team X scored when there was only one player in it. They performed far below the Top 5 and also below the group average. In my quiz, teams could be up to three players in size but typically nearly all teams consisted of two people. So it’s not surprising that a team of one would not be in the Top 5 and also find themselves below the group average.
In the middle of the chart you can see how Team X performed when the team size increased to two – essentially matching most other teams. Their performance fluctuated but broadly the additional team member brought their performance to above average. This is completely in line with what I would expect and to me looks like normal scoring.
The interesting change happens when a third team member is introduced. After the first week they perform very well, in fact they are easily in the Top 5. This could simply be explained by the extra team member boosting the teams knowledge and naturally improving their score.
But, as the weeks go on, their score continues to get better and better to the point where they are consistently answering nearly every question correctly. It’s the repeatability of incredibly high performance and lack of variation that stands out – particularly when you can see that the group mean and Top 5 performance remains relatively unchanged.
What I haven’t mentioned yet is that the scoring shown above is for every round except my anti-cheat round I covered in Strategy 4. Team X were scoring more than 90% in every standard question and answer round but were the worst performing team in the MashUp!
I believed that their consistent scoring in the early rounds was evidence of front loading their score to improve their chances of winning because they knew they couldn’t score in the fair-play-weighted final round.
With this information I spoke to the team and the team member who took them to three people dropped out and their scores returned to normal again. Case closed!
So there are my 5 strategies to spot and stop cheating in your own quizzes.