Script to re-process Stack Overflow Developer Survey 2019 data to get tech popularity stats by self-reported developer type

lang_by_dev_type.py

import csv
from collections import Counter


have_tally = {}
want_tally = {}
# Get the dataset at insights.stackoverflow.com/survey
path_to_datum = '/path/to/downloaded/developer_survey_2019/survey_results_public.csv'

# This *should* work without too much modification with SO survey datasets from other
# years, but might require changing the names of columns.
with open(, 'r') as datum:
    reader = csv.DictReader(datum)
    for row in reader:
        dev_types = row['DevType'].split(';')
        have_worked_language = row['LanguageWorkedWith'].split(';')
        have_worked_web_frame = row['WebFrameWorkedWith'].split(';')
        have_worked_database = row['DatabaseWorkedWith'].split(';')
        have_worked_platform = row['PlatformWorkedWith'].split(';')
        have_worked_misc_tech = row['MiscTechWorkedWith'].split(';')
        want_work_language = row['LanguageDesireNextYear'].split(';')
        want_work_web_frame = row['WebFrameDesireNextYear'].split(';')
        want_work_database = row['DatabaseDesireNextYear'].split(';')
        want_work_platform = row['PlatformDesireNextYear'].split(';')
        want_work_misc_tech = row['MiscTechDesireNextYear'].split(';')
        have_techs = have_worked_language + have_worked_web_frame + have_worked_database + have_worked_platform + have_worked_misc_tech
        want_techs = want_work_language + want_work_web_frame + want_work_database + want_work_platform + want_work_misc_tech
        for dev_type in dev_types:
            if dev_type == 'NA':
                continue
            if dev_type not in have_tally:
                have_tally[dev_type] = Counter()
            if dev_type not in want_tally:
                want_tally[dev_type] = Counter()
            for tech in have_techs:
                if tech == 'NA':
                    continue
                if tech not in have_tally[dev_type]:
                    have_tally[dev_type][tech] = 1
                else:
                    have_tally[dev_type][tech] += 1
            for tech in want_techs:
                if tech == 'NA':
                    continue
                if tech not in want_tally[dev_type]:
                    want_tally[dev_type][tech] = 1
                else:
                    want_tally[dev_type][tech] += 1

print('====== Haves ======\n')

for dev_type in have_tally:
    print(dev_type + '\n')
    for tech, count in have_tally[dev_type].most_common(15):
        print(tech + ' | ' + str(count))
    print('\n')

print('====== Wants ======\n')

for dev_type in want_tally:
    print(dev_type + '\n')
    for tech, count in want_tally[dev_type].most_common(15):
        print(tech + ' | ' + str(count))
    print('\n')