One of our greatest goals here at splatterjunkie.com is to quantify gore in order to make comparisons between movies. One way we have done this is with our Gore Quotient, which is a measure of the proportion of screen-area that is occupied by blood throughout a given movie. With the Gore Quotient, we are able to quantitatively tell if one movie is bloodier than another movie and by how much.
What we can’t tell at the moment is if a given movie is bloodier than average or not. We don’t know what the average amount of bloodiness is for a horror movie and we can’t calculate this average without calculating many more Gore Quotients. How many more Gore Quotients? Well, using IMDB’s advanced search feature, we looked up the total amount of feature film and direct-to-video horror releases between the years 1890 and 2014. The number we came up with was 18,480. That’s 18,480 total horror movies made worldwide since 1890.
Normally, to calculate an average, you would add up all your numbers and then divide by the amount of numbers, but 18,480 is way too much to sort through. So instead of dealing with all those movies, we simply take a representative sample from which we can determine what is true of the larger population of horror movies. Our total population is 18,480 and using the wonderful sample-size calculator at http://www.macorr.com, we were able to determine that a representative sample with a 95% confidence level and a 5% confidence interval would consist of 376 movies.
That means that we would need to calculate Gore Quotients for 376 movies in order to calculate the overall average amount of bloodiness amongst all horror movies. With a sample size of 376, we could be 95% sure that the average we get from the sample is true of the whole population of 18,480. Once we have this representative sample, calculating the average bloodiness would only be the beginning; we could perform all manner of statistical tests with our data and we could be 95% sure that our findings would be true of all horror movies.
For example, if we wanted to see how many horror movies feature most of their kills in the last 30 minutes, we could look at our Kill Graph data, determine how many movies meet that criteria, and then calculate the percentage. Let’s say that we find that 40% of our 376-movie sample (with confidence level 95% and confidence interval 5%) features movies where most of the kills occur in the last 30 minutes, then that means we can be 95% sure that between 35% and 45% of all horror movies have most of their kills in the last 30 minutes. Granted, there’s a lot of room for error between 35% and 45%, but our accuracy will steadily improve as our sample of movies grows. Long-term, we’re aiming for a confidence interval of 2%, which would decrease our error substantially. For that level of accuracy, we would need a sample size of 2,125 movies.
We’re not there yet, but one day we’ll reach our first goal of 376 movies reviewed and then we can get really get our statistics on. We are working toward a total understanding of gore, both to satisfy the curiosity of the average horror-junkie and to advance the study and knowledge of horror movies as a legitimate research area. That’s something you won’t find at any other horror review site. We are splatterjunkie.com, spread the word–Horror Through Math!
splatterjunkie 