Marketing Research Project: Detailed data analysis on a Movie Questionnaire
Overview
In our business analytics course, we had a core subject for Marketing Research. One of the projects in this subject involved a detailed data analysis project. In this project, we were given a marketing research survey questionnaire and a dataset with the respondents’ answers. The questionnaire and dataset contained information and data from several users regarding their frequency to watch the movies at the theatres. The target audience for this survey were students and the main objective of the research was to conduct a market segmentation of students based on their attendance to the movie theatres. The task was to understand all this information and find patterns and insights that would help us answer a series of questions that contribute to customer segmentation.
The Questionnaire
The questionnaire comprised of 14 questions that extracted information which was sufficient enough to understand the attendance of students at the theatres. The first question of the questionnaire asked if the student had seen any one movie in a theatre in the past year. This was the main segmentation question as the survey was designed to create students in two groups, students who watched more movies in theatres and students who rarely went to the theatres.
The students who answered Yes, were asked to move ahead to other questions that asked for information like:
● how important for them was to go to the movies?
● how much they would spend when they visit the theatres?
● What movie theatre items like clean restrooms, size of screens, soft drink and food options etc. are most and least important to them?
The students who answered No, were redirected to the end of the questionnaire where all students were asked how likely they think they are socially and physically active.
The questionnaire ended with a set of demographic questions asking the ethnic background of the student, gender and education classification like junior, senior or graduate student.
The Dataset
An Excel datasheet contained the students’ answers for the questionnaire. The sheet contained the dataset with the responses, a description of the question known as variable label and each value of the label.
Cleaning the Dataset
Here are the steps that were followed to ensure that all inconsistencies are removed from the dataset so that we can start using it for analysis.
● Cleared any inappropriate formatting in the dataset.
● Converted the entire dataset in tabular form which makes it easier for analysis.
● Removed duplicate records.
● Removed any outliers that could result in faulty information or extreme skewness in the data.
● Removed all missing values.
Questions for the Marketing Research Analysis
Some of the important questions that helped us in gaining insight about understanding the student segmentation were as follows:
● Is there a statistically significant difference between people who have gone to the movies at least one time in the past year to the people who have not been to a theatre?
● Is there a statistically significant difference between the male and female who have seen at least one movie in the past year?
● Is there any relationship between the education level and ethnicity of the respondent?
● Which movie item can be predicted to be the most important factor for people when they go to movie theatres?
● How important is it for people to go to the movie theatres?
Final Approach
In order to answer the above questions, the dataset had to be studied and analysed thoroughly.
To compute the statistical difference between the responses, we made use of popular statistical tests like t test and z test which will be ideal to determine the contrast between the groups. In our case study, 449 respondents had watched at least one movie in the past year and 49 respondents had not watched any movie in the theatre in the past year. The mean value for both these groups is 3.2204 and 3.224, respectively, which is almost similar. The conclusion from the z test and t test show that the p values were greater than 0.05 in all the tests.
To find the statistical difference between male and female who have watched at least one movie in the past year, we computed that there were a total of 448 respondents and 201 were male and 247 were female. The mean response for male was 3.4 and for female it was 3.04. The p values for these groups were found to be less than 0.05, which meant that there is a statistical difference between male and female who watched at least one movie at the theatre in the past year.
When we cross tabulated the ethnicity and gender of the respondents, we found that 76% Caucasians, 15% African Americans, 2% Hispanic and 7% other communities.
For Caucasian respondents, 38% were senior, 28% were junior, 18% sophomore, 14% grad student and 2% freshmen.
For African American respondents, 32% were senior, 39 % were junior, 22% sophomore,2 % grad student and 5% freshmen.
Our analysis also showed that women tend to participate in the survey more than men in case of both education level and ethnicity.
To determine which movie items are important to respondents when they go to the theatres to watch the movies, we calculated simple linear regression.
The above table shows that, selection of food and drinks, auditorium type seating, comfortable chairs and plentiful restrooms are the important movie items that determine their visit to a particular theatre for a movie.
The below figure explains the distribution of how important it is for the respondents to for a movie. 4% respondents said it is very important, 32% respondents said that it is somewhat important, 19% respondents said that it is very unimportant and 45% said that it is somewhat unimportant.
Conclusion
Using various statistical and data manipulation approaches, we were able to identify and answer all the questions that were put forth on the dataset. This data analysis project in Marketing Research not only gave us a chance to learn about how to gather data for a customer segmentation market, but it also helped us apply and learn several steps of data analysis and deriving some key insights from the data. We understood the relevance of using z test and t test, simple and multiple linear regression and how they can be used to obtain valuable information that can help perform better business decisions.
