Summary and Conclusions

Summary and Conclusions

In part 1 the tips dataset was described using statistics and plots using the pandas and seaborn libraries. The type of variable determined the type of statistic and visualisation to use.

In part 2 I did some investigation into simple linear regression and looked at the relationship between total bill and tip amount and calculated some regression and correlation statistics using the numpy library. The main focus though was on using seaborn plotting library. To look at performing an actual regression analysis another package such as statsmodels or the scikit-learn package sklearn would be required but this is outside the scope of this particular project.

Part 2 looked at some background on regression and applied this to the total bill and tip amounts. There is a correlation between total bill amount and the tip amount but the plots and statitics show that there are other factors at play. Party size of the party was also positively correlated with the tip size but negatively correlated with percentage tip. There were very few parties of small and large sizes in this dataset though. There is definitely a relationship between the total bill amount and the tip. The plots suggested a linear relationship between total bill and tip amounts for lower and medium sized total bills but the relationship looked less linear as the total bill size increases.The bill amount appears to influence the tip amount in general but in particular for lower bill amounts than for larger total bill amounts. The relationship seems to weaken as the bill amount grows. There are many bills for which you would expect higher tips in the lower right side of the plot and for which a waiter might be disappointe

Part 3 looked more closely at the sex, smoker and size variables in the dataset. The statements about any of these variables only apply to this dataset and cannot be used to draw conclusions outside of this dataset!

This dataset was a good dataset to explore and learn how to use the Seaborn library but not for drawing any conclusions about tipping patterns of diners in general!.