Aidan Litt Ben Soderberg Chase Sonnemaker Joe Tortorello

Slides:



Advertisements
Similar presentations
THIS IS AN EXAPLE OF A DATABASE! A database is a somewhere that you record data !
Advertisements

Transfer Pricing Finding Comparables for Intangibles
SEARCHING, SORTING, TOPOLOGICAL SORTS Most real world computer applications deal with vast amounts of data. Searching for a particular data item can take.
6.1 Reading Circle, Bar and Line Graphs. A graph shows information visually. The type of graph usually depends on the kind of information being disclosed.
Access 2007 ® Use Databases How can Access help you to find and use information?
Your Imagination, Our Innovation Add Event or Presentation Title in Master Slide Errors on searching on lot numbers After tabbing out of “start” field.
Group Buying
Data from Luanda By Charlotte and Keina.
Measures of Central Tendencies Week # 2 1. Definitions Mean: or average The sum of a set of data divided by the number of data. (Do not round your answer.
Statistics Recording the results from our studies.
Statistics 2. Variables Discrete Continuous Quantitative (Numerical) (measurements and counts) Qualitative (categorical) (define groups) Ordinal (fall.
Descriptive Statistics Prepared by: Asma Qassim Al-jawarneh Ati Sardarinejad Reem Suliman Dr. Dr. Balakrishnan Muniandy PTPM-USM.
GrowingKnowing.com © Frequency distribution Given a 1000 rows of data, most people cannot see any useful information, just rows and rows of data.
Grade 8 Math Project Kate D. & Dannielle C.. Information needed to create the graph: The extremes The median Lower quartile Upper quartile Any outliers.
Carrying out a statistics investigation. A process.
Chapter 1: Exploring Data
Melissa Francis. O UTLINE Background Value Created Competition.
I wonder if,……… How to use the scientific process to find answers to what you wonder about.
Lecture 7 Data Analysis.  Developing coding scheme  Data processing  Data entry  Data cleaning & transformation  Data analysis  Interpretation of.
Welcome to Stoneleigh’s STEM Fair Parent Night!
Chapter 14 Statistics and Data Analysis. Data Analysis Chart Types Frequency Distribution.
Lately I’ve been, I’ve been doing math Thinking about the different plots and graphs An outlier’s less or greater than the rest Said no more counting numbers,
Coastal Carolina University
Data analysis is one of the first steps toward determining whether an observed pattern has validity. Data analysis also helps distinguish among multiple.
Chapter 12 Understanding Research Results: Description and Correlation
Amazon Echo Dot: An Echo Without A Decent Speaker
Honors Do Now: (10 mins. Max)
3.4 Solving Systems with 3 variables
How to set up successful graphs
Bringing data to life -statistical approaches to global issues years Session 3 Add notes about what the lesson is about or background info about.
The scatterplot shows the advertised prices (in thousands of dollars) plotted against ages (in years) for a random sample of Plymouth Voyagers on several.
Equations and Inequalities
Organizing Data: Mean, Median, Mode and Range
Co-Curricular Hours vs. Homework Hours
Hook.
Statistics: Stem-and-Leaf Plots
Is this the Asian Century?
Statistics Exam questions
Module 8 Statistical Reasoning in Everyday Life
Crystal Formations! By Quinn Santucci.
Scratch Where Are You Now?
Psychology Statistics
Determining Local Teaching and Learning Priorities
PRACTICE A normal distribution of scores has a standard deviation of 10. Find the z-scores corresponding to each of the following values: a) A score of.
During this survey, I went up to 50 random students and asked each one what time they woke up in the morning for school .These are the results I got…..
Statistics: The Interpretation of Data
Access: Database Design Participation Project
JAPI 2016 Foreign Student in Japan Survey – post-graduation careers
Notes Solving a System by Elimination
Notes Solving a System by Elimination
Mean Absolute Deviation
Top Rated English Movies of This Decade
Summary (Week 1) Categorical vs. Quantitative Variables
The Standard Deviation as a Ruler and the Normal Model
Summary (Week 1) Categorical vs. Quantitative Variables
Statistical Analysis and Unit Improvement Plan Book pgs
Scientific Method Outline
Cloudy With a Chance Of Ice Cubes
The Frequency Distribution
Week 2 Fundamental Research Approaches
Science Fair Projects Atlantis Elementary
Warmup Take out the Fossil Find worksheet from yesterday and begin answering the questions on the second page.
Analyze Data: IQR and Outliers
Research skills for developing your Big Idea
Evidence Gathering Journals
Honors Do Now: People who are vegan (they only eat fruits and vegetables) rarely get cancer. How would a scientist explore such a statement?
Unit 2: Descriptive Statistics
Chapter 1 The Science of Biology
Samples and Populations
Samples and Populations
Presentation transcript:

Aidan Litt Ben Soderberg Chase Sonnemaker Joe Tortorello

What is Kickstarter? Ben

The Anatomy of a Kickstarter Project Title Creator Blurb Pledged Amount/ Goal Number of Backers Time Remaining Category Staff Pick Location Ben The Anatomy of a Kickstarter Project

With billions of dollars invested, Kickstarter is a big deal! Ben With billions of dollars invested, Kickstarter is a big deal!

How’d we get this data? Ben

Cleaning Process… Kaggle: Data Sharing Website Creation of Variables Year Percent Goal Indicator Removal of Variables Issues… Launch Years Suspended Projects Outliers Ben

We used a Python program and the rPython package to scrape several additional data fields from Kickstarter’s advanced search form After many hours, Kickstarter temporarily banned us Some of the projects we searched for were not found, likely because the creators had changed the project names Because of these two things, we ended up with the additional data for 172,941 projects our of our original 378,661 We got additional pieces of data for 172,941, about 45% of the projects in our original data set

Demographics of the Data

English-speaking, western countries have the largest number of projects.

Number of projects rises until 2015, then falls Most projects ask for $15,000 but get less than $5,500

Most popular categories: Film/Video, Music, and Publishing

Alright, Chief. I want to make the best Kickstarter ever Alright, Chief. I want to make the best Kickstarter ever. What should I make?

When should I create my project?

You should have made your project in 2011. Oops! Failed Succeeded You should have made your project in 2011. Oops!

2013: Growth of Backers Levels Out

What Country should it be based in?

Hong Kong has the highest success rate.

What Category should it be?

Graphs for Backers and USD Pledged per Category

Median USD Pledged per Category

What words should be used in the blurb?

Our blurbs contain 124,108 unique words, after normalizing case and removing punctuation. We removed these common words: And The To Are You For In I An A With On We It This From Of By That Be Is

Which words are better? Failed Projects Successful Projects Minimum frequency for these graphs was 500, so even though a lot of the words appear in both, fewer made the cut for successful, and the big ones are generally bigger. Also note that “my” is bigger than “our” in the failed projects, but this is reversed in the successful ones. Failed Projects Successful Projects

A sentiment score of -0.1875 has the highest success rate! Failed Succeeded A sentiment score of -0.1875 has the highest success rate!

Create multiple Projects!

Failed Succeeded Ben Most creators who make multiple projects have high success rates among their projects (besides Steven Walvick).

Is it a Staff Pick?

Chicken or the Egg? Do staff picks make success or do the staff chose projects more likely to succeed?

So, what’s the Conclusion?

What kind of project should you make? Year? 2009 - 2011 Country of Origin? Countries advancing in technology Ex. US, Hong Kong, Singapore, UK Category? Dance, Theater, Comics Words? Sentiment Score of -0.1875 Cooperative words like “our” and “help” Multiple Projects? Yes! Make multiple, similar projects. Staff Pick? Yes!

What would we do if we had more time?

We got most of what we wanted to do done We got most of what we wanted to do done. Just a few more small things to add: Scrape data for remaining records Investigate backer statistics Gather all of the data for 2018 Investigate the Hong Kong Phenomenon (HKP) further Create a better model Test our model by creating a Kickstarter project of our own.

Have an Idea?

Try your project idea on your laptop or mobile device www.kickstats.org/try Scan or type @kickstats.org aidan ben chase joe } Questions? Try it out for yourselves!