Vfrdtyky

list1 one is a dict of dfs. data1 is a lone df.

I want to create another dict of dfs(d) based off a crosstab which will categorize each column in each df in 'list1' on specific questions in 'data1'. This is what I've tried:

d={}

for df in list1:

    for col in list1[df]:

        d[df]= pd.DataFrame(pd.concat([pd.crosstab([df][col],data1['column1']),

                                        pd.crosstab([df][col],data1['column2'])],

                                        axis=1)) 

I then get 'TypeError: list indices must be integers or slices, not str'.

I appreciate that there have been many questions on this before on stackoverflow, however I have struggled to apply any of the solutions to my problem.

df iterates over the keys of list1 when you do for df in list1. So it is a string. Later, when you pass [df][col], it creates a list of one element which is a key, from which you try to get a string column. I suppose you meant d[df] instead.

For clarity purpose I would also suggest renaming list1 and changing the for loop in for df in <new_name>.keys()