Earlier in 2017, the GNQ team wrote some notes on quiz statistics. Since then, GNQ have been asked numerous times regarding things like breakdown by country and other stats. So we have put together some simple question data which some readers may find interesting.
The GNQ questions are set into five different categories - Geography, History, Science, Culture and Trivia. Obviously, it might seem that all general knowledge questions would fall under trivia. However, trivia is essentially the category we use to put all the questions that don't fall into any of the other four categories. Note that these are our classifications and therefore sometimes the questions might stretch a little.
As some examples, 'What shape has ten sides ?' would fall under Science. The question 'The Corleone family features in which film ?' falls under Culture as it relates to a film.
At the time of writing, the percentage breakdown by questions was as follows :
However, the performance by category varied significantly.
The geography category had a significant reduction in performance. Where GNQ classified 15% of questions to be geographic in nature, the same set of questions only managed to get 9% of correctly answered questions. This actually represents quite a swing and means that geography and related topics are a weak spot in terms of general knowledge. Conversely, science questions performed slightly better than expected. There may be a number of reasons for this, including the relative difficulty of the questions. However, there are now thousands of questions in the database so, on average, the level of the questions will be fairly even. It can of course imply that geographical questions are, by nature, more difficult but that's very difficult to quantify.
Question response time is the amount of time that it takes for a user to answer a question. As the quiz is timed, the actual amount of time taken to answer a question is recorded. The quicker someone can answer, the higher the score. If someone answers very slowly then the score will be zero. This was originally intended to prevent people looking up the answers on the internet but it makes the game more interesting.
The data for every question is recorded and it is possible to compare the time it takes to answer a question by the type and category of question. GNQ will go further into response times by question at some point in the future but for the moment let's look at the response times in general. It's worth mentioning here that as the maximum amount of time taken to answer a question results in a zero score, the GNQ system only registers to a certain point. In other words, if someone took 35 second to answer a question then we would record zero. If they took 3 hours to answer a question, we would still record zero. This means that the recorded response times are within a certain limited range which is currently up to 30 seconds. The actual zero point is not always set at 30 seconds though.
The results are more or less to be expected. The first band are almost certainly people that have seen the question before and have an instinctive reaction. It would be difficult to read some of the questions and answers within a couple of seconds so the response time of less than 5 seconds is very fast. The peak then comes between 6 to 10 seconds. People who have read the question and selected an answer. The response times then tail off until another peak at the 26-30 range. After this point, we do not record the data. Therefore, it is possible that there's a substantial number of people that take longer than 30 seconds to answer.
The data doesn't show whether the user got the answer correct or not. GNQ will break that data down at a future date.
Additionally, there are variations in response time across different categories. Some categories tend to get a faster response than others. However, the data might be slightly inaccurate because some categories tend to have average question length longer than others. This means that it would of course take longer to read !
The quiz is taken by people all over the world. The largest percentage is from the UK which is likely to be because it's a UK website address. However, GNQ has many users which regularly return and take the test and the audience is international. The top five countries to visit GNQ during November 2017 were the UK (56%), the US (18%), Ireland (8%), India (8%) and Australia (5%). Other countries making a showing were Spain and Germany. Obviously, as this is an english language quiz, there's more likely to be visitors from english speaking countries.
In terms of performance by visits, it's not looking good for the UK.
There may be a number of reasons for this. Not least of which that to find this website from other countries is more difficult as it's on a UK domain name. Therefore, the people that find it out are perhaps more 'quiz' type people and therefore have a better go of things. The questions that are selected are international. GNQ avoid selecting questions that are specific to one single country. If the question regards a country then the answer will be one where anyone in the world might be expected to answer. For example, if the question was 'In which city would you find Central Park ?', then although the question relates to the US, many people throughout the world are likely to know the answer.