Using your data analysis plan for guidance, carryout an initial exploratory analysis (summary statistics and graphs) on your collected data.

Assignment Brief

Quantitative Methods

Data Analysis Report:

Data Collection

Go onto the Kickstarter funding website (www.kickstarter.com). Using your data analysis plan as a guide, select your sample of projects

Introduce the topic and describe your sampling approach in detail including:

• The sampling methodology you have used
• How many projects you have used in your analysis
• Explain if you have excluded any projects, if you have, explain why. d) Any limitations in the data that may affect the analysis

Initial Analysis of your data

Using your data analysis plan for guidance, carryout an initial exploratory analysis (summary statistics and graphs) on your collected data.

Regression and Correlation Analysis

• Use regression and correlation analysisw ith accompanying graphs to analyse the relationship between ‘Amount Pledged’ and your chosen independent variable (depending on your research question). Discuss your regression equation
• Test your regression equation and interpret the results.
• Explain whether your chosen variable is a good predictor for the ‘Amount Pledged’. Consider the correlation (r) and the coefficient of determination Rsquare. If your chosen variable does not account for 100% of the variation in ‘Amount Pledged’ discuss why this could be the case.

Sample Answer

Quantitative Methods: Kickstarter Data Analysis Report

Introduction and Sampling Approach

Kickstarter is a well-known crowdfunding platform where people can support creative projects. For this report, we are using quantitative methods to explore how certain project characteristics affect the amount of money pledged.

a) Sampling Methodology

To collect the data, I used purposive sampling — I focused on a specific category of projects (e.g. “Technology” or “Games”) so the data would be relevant to my research question. I chose this method to ensure consistency in the type of project and reduce variation caused by comparing different industries.

I searched Kickstarter’s website and manually selected 30 projects that were launched within the last two years and had completed funding campaigns. This gives a good-sized sample for analysis while keeping the data manageable.

b) Sample Size

• Total Projects Selected: 30

• Category: [insert category e.g. Technology]

• Time Frame: Projects funded between [insert year range, e.g. 2023–2024]

c) Excluded Projects

Some projects were excluded from the analysis:

• Projects with missing data (e.g. no pledged amount or unclear goal).

• Projects that were cancelled before the end of their campaign.

• Projects with extreme values/outliers that would skew the analysis (e.g. celebrity-backed campaigns).

d) Data Limitations

There are a few limitations in this data:

• The Kickstarter platform only shows summary project data, not in-depth financials.

• Backer motivation or promotional efforts are not captured in the dataset.

• Outliers can still affect results even after exclusions.

• The sample is limited to one category and may not reflect all projects on the site.

Regression and Correlation Analysis

For the main analysis, I chose “Goal Amount” as the independent variable (X) and “Amount Pledged” as the dependent variable (Y). The purpose is to see if the goal amount can predict how much money the project actually raised.

a) Regression Equation

The simple linear regression equation is:

Y = a + bX
Where:

• Y = Amount Pledged

• X = Goal Amount

• a = Intercept

• b = Slope

From my analysis (values will depend on your actual data):

Regression Equation Example:

Pledged = 2500 + 0.8 × Goal

This means that for every £1 increase in the goal, the amount pledged goes up by £0.80, on average.

b) Interpretation of Results

• Slope (b): Positive, showing a direct relationship.

• R² (coefficient of determination): [e.g. 0.64] – This means that 64% of the variation in pledged amounts is explained by the goal amount.

• Correlation Coefficient (r): [e.g. 0.80] – A strong positive correlation.

c) Is the Independent Variable a Good Predictor?

The goal amount appears to be a good predictor, but it does not explain 100% of the variation. This is likely because:

• Projects may receive more attention due to marketing or media exposure.

• Quality of project description, rewards, or creator reputation affects success.

• Timing, trends, and social sharing can also impact the amount pledged.

Continued...