Tidy Data

import numpy as NP # numerics
import numpy.random as RND # random numbers
import pandas as PD # data tables in python
import names as NAME # random names
import datetime as DT # date and time

RND.seed(42) # this makes that random numbers always come out the same on my computer

n_students = 24 # the number of students in the hypothetical class


averages = 1. - RND.beta(2., 5., n_students) # values between zero and one often follow a beta distribution: https://en.wikipedia.org/wiki/Beta_distribution
sex_is_male = RND.choice([0., 1.], 24) # whether a student is male (1) or female (0)

names = [NAME.get_full_name(gender = 'male' if male else 'female') for male in sex_is_male] # choose random names
# (I'm sorry that this all seem to be American/British names.)

# get random birth dates, some time in 2006 or 2007
start_year = DT.datetime.strptime('1/1/2006', '%d/%m/%Y')
birth_day = [start_year + DT.timedelta(days = int(day)) for day in RND.uniform(0, 365*2, n_students).astype(int)] # gets some random birth dates within a year
birth_day = [bd.strftime('%d/%m/%Y') for bd in birth_day]


def GetTestScores():
    # this function will generate random test scores.

    scores_base = averages + RND.normal(0., 0.1, n_students) # roll the dice
    scores_base -= 0.05 * sex_is_male
    failure = RND.uniform(0., 1., n_students) < 0.05 # sometimes, a student will fail!
    prepared = RND.uniform(0., 1., n_students) < 0.05 # sometimes, a student will be prepared!

    scores = scores_base - failure * 0.1 + prepared * 0.1 # add in failures and preparation

    # check limits: scores are between 0% and 100%
    scores[scores < 0.] = 0.
    scores[scores > 1.] = 1.

    return NP.round(scores, 2) # return rounded test results.


# combine the data
data = PD.DataFrame.from_dict({ \
                                'naam': names \
                                , 'geslacht': [('m' if male else 'v') for male in sex_is_male ] \
                                , 'jarig': birth_day \
                               })
data.index += 1
data.index.name = 'klas_nummer'

# calculate test scores
for test in ['hoofdstuk 1', 'hoofdstuk 2', 'hoofdstuk 3', 'dagelijks werk', 'examen']:
    data[test] = GetTestScores()


# save the data on the hard drive.
data.to_csv('data_klasoverzicht.csv', sep = ';', decimal = ',')
print (data.head(3).T)