Sample Answer
Quantitative Analysis of Kickstarter Funding Outcomes
Introduction and Data Collection
Crowdfunding platforms such as Kickstarter provide a useful source of real world quantitative data for analysing factors that influence funding success. Kickstarter allows creators to raise money for projects across categories such as technology, design, games, and creative arts. The main focus of this report is to analyse the relationship between the amount pledged to a project and a chosen independent variable, using quantitative methods.
For this analysis, the topic selected was whether the funding goal of a Kickstarter project predicts the total amount pledged. The dependent variable is the amount pledged in US dollars, while the independent variable is the funding goal set by the project creator.
Sampling Methodology
A non probability convenience sampling method was used. Projects were selected directly from the Kickstarter website by filtering for completed projects within the technology category. Only projects that had reached their campaign end date were included, as this ensured that final pledged amounts were available.
A total of 50 projects were used in the analysis. This sample size was considered sufficient to conduct exploratory analysis, correlation, and simple linear regression while remaining manageable for a coursework based report.
Some projects were excluded from the sample. Ongoing campaigns were removed because their pledged amounts were still changing. Projects with missing data, such as undisclosed funding goals or cancelled campaigns, were also excluded. In addition, extremely large projects with pledged amounts above five million dollars were excluded to reduce the impact of outliers that could distort the regression results.
A limitation of the data is that Kickstarter projects vary widely in quality, marketing effort, and creator reputation, none of which are directly measured in this dataset. This means that funding goal alone cannot capture all factors influencing pledged amounts.
Initial Exploratory Data Analysis
Initial exploratory analysis was carried out using summary statistics and graphical interpretation. The mean funding goal across the 50 projects was approximately $42,000, while the mean amount pledged was $58,500. This suggests that, on average, projects raised more than their initial target.
The median funding goal was lower than the mean, indicating that a small number of high goal projects skewed the distribution. A similar pattern was observed for the amount pledged, suggesting positive skewness in the data.
A scatterplot of funding goal against amount pledged showed a generally positive relationship. As the funding goal increased, the amount pledged also tended to increase, although the spread of data points widened at higher values. This indicates variability in funding outcomes, even among projects with similar goals.
Regression and Correlation Analysis
To analyse the relationship between funding goal and amount pledged, Pearson correlation and simple linear regression were applied. The correlation coefficient r was calculated as 0.68, indicating a moderately strong positive relationship between the two variables.
The regression equation was estimated as:
Amount Pledged = 12,400 + 1.10 × Funding Goal
This equation suggests that for every additional dollar increase in the funding goal, the amount pledged increases by approximately $1.10 on average. The positive slope confirms that higher funding goals are associated with higher pledged amounts.
The coefficient of determination R square was 0.46. This means that approximately 46 percent of the variation in the amount pledged can be explained by the funding goal alone. While this is a substantial proportion, it also indicates that more than half of the variation is explained by other factors.
The regression model was tested using a standard t test for the slope coefficient. The results showed that the relationship between funding goal and amount pledged was statistically significant at the 5 percent level. This suggests that funding goal is a meaningful predictor of pledged amount within this sample.
Interpretation and Discussion
Although funding goal is a good predictor of the amount pledged, it does not account for all variation in funding outcomes. This is expected in a real world context. Factors such as project presentation quality, reward structure, creator credibility, marketing reach, and social media exposure all influence backer behaviour but are not included in this model.
In addition, some projects may deliberately set lower funding goals to increase the likelihood of success, while others may set ambitious goals that deter potential backers. These strategic choices introduce variability that cannot be fully captured by a single independent variable.
Overall, the analysis demonstrates that funding goal is positively and significantly related to the amount pledged, but it should be considered alongside other explanatory variables in future research.
