Aiden Jung, Matthew Gao, Trung Le
2/22/22
Kickstarter is a public benefit company that encourages crowdfunding for creative projects.
We will be looking at important statistics for each Kickstarter project: a goal, the number of backers (donators), and the total amount pledged to the project.
We hypothesize that there is direct correlation between the number of backers to a project and the amount pledged to that project.
To perform our tests, we cleaned the data, tidied it, performed a linear regression test, and used data vis (a scatterplot).
Given a linear regression equation of y_hat = B_0 + B_1 * x…
H_o : B_1 = 0
H_a : B_1 > 0
Call:
lm(formula = pledged ~ ln, data = kickstarter_cleaned)
Residuals:
Min 1Q Median 3Q Max
-118703 -39486 -12516 21719 6038202
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -47992.4 1607.0 -29.86 <2e-16 ***
ln 23969.3 437.1 54.84 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 120000 on 17694 degrees of freedom
(2936 observations deleted due to missingness)
Multiple R-squared: 0.1453, Adjusted R-squared: 0.1452
F-statistic: 3007 on 1 and 17694 DF, p-value: < 2.2e-16
With a p-value of 2.2e-16 < 0.05, we reject H_o that B_1 = 0. There is a clear relationship between the number of backers to a Kickstarter project and the amount of money pledged to that project.
We encountered some limitations:
The dataset was pulled from Kaggle: https://www.kaggle.com/yamqwe/kickstarter-campaign-successe
R Documentation