Movie Survey Results
The Oscars are tomorrow and so before Hollywood annouces who their winners are, I would like to annouce the results the the movie survey that so many of you participated in. The purpose behind this survey was to help me formulate my own Top Ten Movies from 2004, and I will post that list tomorrow. But it was more than just that, it was a chance for me to see how people rated not only the movies under my consideration, but also their take on movies or aspects of movies released in 2004. And so I will show the results to all of the questions asked and explain how the data was synthesized. Also note that my results are included in this data, though not biased in any way.
The first set of data I would like to show is the 'Percent Seen' chart, which shows what percentage of people saw each movie on the list. Eighteen participants answered this segment and revealed that the most seen movie of 2004 was in fact "Napoleon Dynamite", whereas the least seen was "Hotel Rwanda". The remaining results can be seen in this figure.
Moving on to the Top 12 rankings of the movies under my consideration for my Top 10, I received rankings from 15 people and found three ways to analyze the data.
Method #1: I took the rankings as they were posted on the survey site from each voter. These results are unchanged and not manipulated. The average ranking of each movie was taken and from this average the movies were ranked accordingly. This ranking is referred to as the Rated ranking.
Method #2: After several tests had been made, I was receiving complaints from people who didn't find it fair to rate a movie they had never seen or heard anything about before. Therefore, another analysis was done taking the movies seen for each person and their top 12 ranking and creating for each person a new ranking. I simply looked at how they rated the movies they had seen with each other, assuming that the movies they did see were rated comparatively with one another regardless of those they hadn't seen. So new ranks were assigned to the movies for each person, with zeros in place of the movies they had not seen. Then the averages of each poll was taken for each movie and a new ranking was assigned to the movies. This new ranking is now referred to as the Pollster's ranking.
Method #3: One more method of analyzing the data was performed to even out the discrepencies between the other methods. This method took the original rankings of each individual and determined how many 1st place votes each movie received, how many 2nd place votes, and so on up through the 12th place votes. Once these values were tallied for each movie, a numerical model was applied such that when the sum of these votes occurred, the 1st place votes received more influence on the ranking than the 12th place votes. I will spare you the technical details, but this model was applied to each movie and a third ranking was taken; this one referred to as the Numerical ranking.
All three rankings are then averaged together and a final ranking is generated, giving the overall results to the Top 12 ranking section. In the result of a tie in any of these rankings (i.e. two tied for 3rd), they are proscribed the same ranking and the following movie receives the ranking it would normally receive if there were no tie (MovieA - 3rd, MovieB - 3rd, MovieC - 5th, etc.). I want to make this clear before you see the results and complain to me about them. So the overall results can now be seen in the table below.
As you can tell, "Garden State" receives a unanimous number one ranking, which leads me to say that it is the best movie of 2004 of these 12 films, as dictated by this survey.
The rest of the survey got a little more specific and allowed more freedom in choosing favorite films, as well as favorite aspects of these films. The results to these questions are on their own separate webpage found here. The most unanimous results can be seen in the Best Character chart (Napoleon Dynamite), Best Soundtrack chart (Garden State), and Best Special Effects chart (Harry Potter). Overall it was wonderful to receive so much feedback on these films and to see not only what people liked but what particular things they liked about it. This survey is completely anonymous so I don't know who you all are (some of you I can figure out), but thank-you for your input and check back tomorrow for my own Top Ten list!
The first set of data I would like to show is the 'Percent Seen' chart, which shows what percentage of people saw each movie on the list. Eighteen participants answered this segment and revealed that the most seen movie of 2004 was in fact "Napoleon Dynamite", whereas the least seen was "Hotel Rwanda". The remaining results can be seen in this figure.
Moving on to the Top 12 rankings of the movies under my consideration for my Top 10, I received rankings from 15 people and found three ways to analyze the data.
Method #1: I took the rankings as they were posted on the survey site from each voter. These results are unchanged and not manipulated. The average ranking of each movie was taken and from this average the movies were ranked accordingly. This ranking is referred to as the Rated ranking.
Method #2: After several tests had been made, I was receiving complaints from people who didn't find it fair to rate a movie they had never seen or heard anything about before. Therefore, another analysis was done taking the movies seen for each person and their top 12 ranking and creating for each person a new ranking. I simply looked at how they rated the movies they had seen with each other, assuming that the movies they did see were rated comparatively with one another regardless of those they hadn't seen. So new ranks were assigned to the movies for each person, with zeros in place of the movies they had not seen. Then the averages of each poll was taken for each movie and a new ranking was assigned to the movies. This new ranking is now referred to as the Pollster's ranking.
Method #3: One more method of analyzing the data was performed to even out the discrepencies between the other methods. This method took the original rankings of each individual and determined how many 1st place votes each movie received, how many 2nd place votes, and so on up through the 12th place votes. Once these values were tallied for each movie, a numerical model was applied such that when the sum of these votes occurred, the 1st place votes received more influence on the ranking than the 12th place votes. I will spare you the technical details, but this model was applied to each movie and a third ranking was taken; this one referred to as the Numerical ranking.
All three rankings are then averaged together and a final ranking is generated, giving the overall results to the Top 12 ranking section. In the result of a tie in any of these rankings (i.e. two tied for 3rd), they are proscribed the same ranking and the following movie receives the ranking it would normally receive if there were no tie (MovieA - 3rd, MovieB - 3rd, MovieC - 5th, etc.). I want to make this clear before you see the results and complain to me about them. So the overall results can now be seen in the table below.
As you can tell, "Garden State" receives a unanimous number one ranking, which leads me to say that it is the best movie of 2004 of these 12 films, as dictated by this survey.
The rest of the survey got a little more specific and allowed more freedom in choosing favorite films, as well as favorite aspects of these films. The results to these questions are on their own separate webpage found here. The most unanimous results can be seen in the Best Character chart (Napoleon Dynamite), Best Soundtrack chart (Garden State), and Best Special Effects chart (Harry Potter). Overall it was wonderful to receive so much feedback on these films and to see not only what people liked but what particular things they liked about it. This survey is completely anonymous so I don't know who you all are (some of you I can figure out), but thank-you for your input and check back tomorrow for my own Top Ten list!
posted by Chris Burzlaff @ 2/26/2005
<< Home