Methodology
Movie Trend
Learn about the statistical process behind the history, and download the data sets for continued research
Research Questions
Did the types of films shown in Little Rock change over time?
Did these changes correlate with major historic and cultural events?
How did these trends and the films shown in Little Rock compare to national trends?
The Process: Part 1
Data Set Creation
Because of the large quantity of available movie data, I decided to collect a representative sample. I choose the following years: 1933, 1939, 1943, 1949, 1954, 1959, and 1963.
These years are evenly spaced for each decade (with the exception of 1954). They also correlate with significant changes in the United States and cinema history. For both the Little Rock and National Movie Data sets, I used the same years.
I used the following genres to tag the movies: adventure, animation, biography, comedy, crime, drama, family, fantasy, film-noir, history, horror, musical, mystery, romance, sci-fi, sport, thriller, war, and western.
For my research, I focused on war, film noir, horror, and family movies because they were either created during the period between the 1930s and the 1960s or I suspected they might have been popular then.
Little Rock Movie Data
1.
Gathered data using the theater managers' weekly reports from the Arkansas Amusement Corporation Papers (UALR.MS.0113)
2.
Recorded the theater name, week ending date, movie title, and distributor for each theater
3.
Added theaters' addresses from the Little Rock city directories
4.
Added genres and U.S. box office release dates from the Internet Movie Database (IMDb)
National Movie Data
1.
Gathered data using the rankings from Ultimate Movie Rankings (UMR)
2.
Recorded the top 50 movies according to actual domestic box office revenue
3.
Collected the movie title, year, and actual and adjusted domestic box office grosses (in millions)
4.
Added genres from the Internet Movie Database (IMDb)
The Process: Part 2
Statistical Analysis
To answer my research questions, I ran a statistical analysis on the Little Rock and National Movie Data Sets using chi-square tests. Through the chi-square tests, I showed there were statistically significant patterns in the use of these genres in Little Rock and the nation over time.
​
A chi-square test examines the frequency and distribution of a variable in a group of objects. It produces chi-square (χ2) and p-values to indicate if there are statistically significant patterns in the data set.
​
Click through the images below to learn more about chi-square tests.
Pattern Present
P-value less than 0.10
and
Chi-square value greater than 14
=
significant differences in distribution
Here, the variable is the genre type (horror, film-noir, war, and family) and the group of objects is time (1933, 1963, 1943, ect.)
​
If a genre is unevenly distributed between 1933-1963, then it was probably more popular in certain years and less popular in others. The chi-square test will indicate there is a pattern or trend present. The chi-square value will be greater than 14 and the p-value will be less than 0.10. This trend is likely the result of historical, cultural, and economic factors of certain years.
​
I ran chi-square tests on both the Little Rock and National Movie data sets. Then, I compared their results. Click on the buttons below to view the full test results or to explore the movie trends of the 1930s-1960s.