According to the [data dictionary of the Census Income Data Set ](https://archive.ics.uci.edu/ml/datasets/census+income), one column name in the repository is: `fnlwgt` In the [data set Udacity provides](https://raw.githubusercontent.com/udacity/nd0821-c3-starter-code/master/starter/data/census.csv), this column is misspelled as `fnlgt` Ergo, the "w" is missing here.