GEA1000 QUANTITATIVE REASONING WITH DATA

TUTORIAL 1
Please work on the problems in Sections 1 and 2 before coming to class.

Note to students: *Think-type questions are questions that offer some depth of discussion. It is
difficult to capture the rich perspectives provided in a group discussion in a short paragraph. While
we provide a ‘model’ response, reach out to your tutors to engage in deeper conversations.

Section 1: Working with data

The data we collect may sometimes be incomplete or contain errors. Before performing
analysis on the data set, we should first understand the variables in the data set and check if
the entries contain any ‘dirty’ data.

Background Information

The data set [Link] contains information on 6338 movies, with variables such as genre,
worldwide gross revenue, popularity, and so on. It was taken on 14 th August 2023 from the
following website:

[Link]

Below is a brief description of the variables in the data set:

Variable Description

Title The title of the movie.

ProductionBudget The cost involved in making the movie.

WorldwideGross The revenue generated from screening the movie

in public theatres around the world.

DomesticGross The revenue generated from screening the movie

in the United States.
DomesticGrossOpening The revenue generated from screening the movie
in the United States during the opening weekend.
DateOfRelease The release date of the movie in DD/MM/YY
format.
Duration The length of the movie in minutes.
MPAARating The Motion Picture Association of America film
rating system is used in the United States and its
territories to rate a motion picture’s suitability for
 certain audiences based on its content. If a film has
not been submitted for a rating or is an uncut
version of a film that was submitted, the label Not
Rated (NR) is used.

IMDbRating IMDb is an abbreviation for Internet Movie

Database, which is an online database of
information related to films. The IMDb offers a
rating scale that allows users to rate films on a scale
of one to ten based on how they feel about the film,
with a low score being regarded as an unfavourable
view and a high score being regarded as a
favourable one.

The submitted ratings are filtered and weighted in

various ways (partly depending on the stature of
the person rating the movie) to produce a
“weighted mean” that is displayed for each film and
IMDb keeps the formula of the weighted mean as
confidential.

VoterNumbers The number of voters who rated the movie on

IMDb which eventually resulted in the IMDb rating.

Genre The style or category of the movie.

Part I: Types of Variables

We begin by looking at the data set as a whole. After verifying the source of data (i.e., the data
set comes from a reliable source), we can describe the characteristics of the data set, such as its
size, and types of variables.

a) For all the variables, except IMDbRating, determine which are numerical and which are
categorical. Were there instances where it was unclear whether to classify the variable as
numerical or categorical?

*Think: Is IMDbRating a numerical or categorical variable?

Part II: Data Cleaning

Dirty data describes data that is inaccurate or incomplete. Examples include missing or wrong
inputs, inconsistent coding, incorrect formatting, etc. Before diving into deeper analysis, we need
to clean the data, that is, to account for and rectify the dirty data, wherever possible. In this part,
you will explore one way to identify and clean dirty data.

Example: Values labelled “Unknown”

Occasionally, information could not be collected for some data points, and they could be left as
empty cells, or could be filled with placeholder terms such as “Unknown”.

b) Identify the variables (columns) with “Unknown” values and state the number of
“Unknown” values for these variables.

*Think: How would you treat this type of dirty data? Are all the identified data
points necessarily inaccurate or incomplete?
Part III: Summary statistics
Using summary statistics is a great way to understand characteristics of a data set quickly. It can
also help to identify interesting or anomalous data points.

Common summary statistics reported for numerical data include the mean, standard deviation,
and the five-number summary, which consists of:
1. Minimum
2. Q1
3. Median
4. Q3
5. Maximum

c) Provide the five-number summary, mean and standard deviation for the variable
WorldwideGross. Are there any anomalous summary statistics? How will you treat them?

*Think: Suppose we wish to use this data set to generate insights on trends and
patterns in the entire movie industry. To what extent can the findings obtained
from the analysis of this data set be generalised to the larger target population of
all movies?

Part IV: Random Assignment

Random assignment is a powerful tool used in experimental studies to form groups that have
similar characteristics. In this part, you will observe how random assignment can achieve this
effect. The data set movies_random_assign.xlsx, which has undergone significant data-
cleaning and contains pre-assigned random numbers, serves as a dummy data set to perform
random assignment on in this part of the tutorial.

d) We will first study the characteristics of the data set before conducting random
assignment. For the data set movies in movies_random_assign.xlsx:
i. State the number of movies.
ii. Calculate the proportion of movies with “Action” as Genre.
iii. Calculate the proportion of movies with “R” as MPAA-Rating.

You may find a PivotTable (see Appendix 1) helpful.

Random assignment was conducted (see Appendix 2) and the movies were randomly assigned
into two groups.

e) For each of the groups (stored as separate Worksheets):

i. State the number of movies.
ii. Calculate the proportion of movies with “Action” as Genre.
iii. Calculate the proportion of movies with “R” as MPAA-Rating.

How do the proportions compare, and what does it tell you about the effectiveness of
random assignment in assigning groups with similar characteristics?
Section 2: Study design

In this section, we explore the importance of a good study design and the practical limitations in
conducting a study.

Scenario
Popcorn is so deeply embedded in the entertainment culture that is virtually inseparable from
the movie-going experience. Is this simply a stereotype, or does popcorn really increase the level
of enjoyment of the movie?

Propose a simple study to explore the effects of consuming popcorn on the enjoyment of the
experience for the target population of NUS undergraduate students. You may choose to enlist
the help of AI or other tools to draft your plans.

Share your findings with your group. As a group, critique your individual study plans, and
develop and present a refined study plan.

In developing and critiquing your study plan, consider:

a) Would you conduct an experimental or observational study? Why?
b) How do you obtain your sampling frame?
c) What are your variables? How will you measure them?
d) If you plan to do an experiment, will you perform any blinding? Why?
e) Are there any ethical considerations?

Additionally, in your group discussion, discuss the practical challenges in recruitment and conduct
of the study (for example, for how long should the study be conducted?).

Remarks: Attempt to formulate the study plan at your comfort level. We do not expect
comprehensive and expert strategies. Instead, we invite you to deliberate on realistic
considerations should you intend to run similar studies.

Share your findings with the class.

Appendix 1: How to Create a PivotTable and PivotChart

Step 1: Go to “Insert”, Select “PivotChart”, and “PivotChart & PivotTable”.

Step 2: Ensure that the relevant data points have been selected and click OK.

Step 3: Suppose that you wish to have a breakdown on genres, you may then left-click “Genres”
from the Right Panel, and drag it twice: once under “Rows” and once under “Values”. One should
be able to observe the PivotTable and PivotChart showing the breakdown accordingly.
Appendix 2: Using Excel to Perform Random Assignment

Step 1: Create a new column. We titled this column “random assignment” in Column L. In the
first cell corresponding to an entry, input the function =RANDBETWEEN(1,2). This Excel function
randomly selects a whole number between 1 and 2 — in this case, it will be either the number 1
or the number 2.
Note: This is a dynamic function; that is, every time a new operation is performed on the
Worksheet, a new random number may be selected.

Step 2: Move your cursor to the bottom right of the cell and observe that the cursor takes the
form of a black ‘plus’. Double click to Auto Fill the column. Check that every entry has been
assigned.
Step 3: Select and copy the entire column. Then choose ‘Paste As Values’ to fix the values in
their places. This step is to turn off the dynamic feature of =RANDBETWEEN().
Note: From Step 2, the keyboard shortcut is simply “Ctrl + C” followed by “Ctrl + Shift + V”.

Step 4: Sort the column from smallest to largest. You may do so by right-clicking a cell from that
column or using the Filter function, amongst other methods. This separates the movies
assigned to the number 1 from the movies assigned to number 2, forming randomly assigned
groups.

Step 5: Copy and paste the two groups each into a new Worksheet.
Examplify Questions using Spreadsheet

The password to access these questions on Examplify is “Gea1000”.

In the case where you are prompted to enter a resume code, please enter “79A589”.
Examplify is configured such that your device will be locked down (unable to access other
programs and communications) for the final exam, but this restriction is disabled for Tutorial 1
Examplify practice. In addition, you will not receive the grade of or feedback on your answers.
(Instead, answers to the Examplify practice are provided below.)
Tutorial 1 Question on Summary Statistics [Fill-in-the-blank]

1. Minimum of WorldwideGross is 268314513 (using “=MIN” function)

2. Mean of WorldwideGross is 479052380 (using “=AVERAGE” function)

3. Maximum of WorldwideGross is 1120210896 (using “=MAX” function)

List of Examplify spreadsheet functions to be tested in 2520_GEA1000/N_FINAL EXAM

=AVERAGE() =MEDIAN()

=MODE() =STDEV.S()

=MAX() =MIN()

=CORREL() =SLOPE()

=CONFIDENCE.T() =COUNT()

=INTERCEPT() =COUNTIFS()