Most Used Languages for Data Related Jobs¶

We are going to check the most used programming languages for data related jobs based on Stack Overflow Developer Survey 2021 results.

Import Libraries¶

In [1]:

import pandas as pd
import numpy as np

Load Dataset¶

In [2]:

result_df = pd.read_csv("C:/Users/Marselo/Downloads/Stackoverflow survey results/2021 survey_results_public.csv")

Overview of the Dataset¶

In [3]:

result_df.head(5)

Out[3]:

	ResponseId	MainBranch	Employment	Country	US_State	UK_Country	EdLevel	Age1stCode	LearnCode	YearsCode	...	Age	Gender	Trans	Sexuality	Ethnicity	Accessibility	MentalHealth	SurveyLength	SurveyEase	ConvertedCompYearly
0	1	I am a developer by profession	Independent contractor, freelancer, or self-em...	Slovakia	NaN	NaN	Secondary school (e.g. American high school, G...	18 - 24 years	Coding Bootcamp;Other online resources (ex: vi...	NaN	...	25-34 years old	Man	No	Straight / Heterosexual	White or of European descent	None of the above	None of the above	Appropriate in length	Easy	62268.0
1	2	I am a student who is learning to code	Student, full-time	Netherlands	NaN	NaN	Bachelor’s degree (B.A., B.S., B.Eng., etc.)	11 - 17 years	Other online resources (ex: videos, blogs, etc...	7	...	18-24 years old	Man	No	Straight / Heterosexual	White or of European descent	None of the above	None of the above	Appropriate in length	Easy	NaN
2	3	I am not primarily a developer, but I write co...	Student, full-time	Russian Federation	NaN	NaN	Bachelor’s degree (B.A., B.S., B.Eng., etc.)	11 - 17 years	Other online resources (ex: videos, blogs, etc...	NaN	...	18-24 years old	Man	No	Prefer not to say	Prefer not to say	None of the above	None of the above	Appropriate in length	Easy	NaN
3	4	I am a developer by profession	Employed full-time	Austria	NaN	NaN	Master’s degree (M.A., M.S., M.Eng., MBA, etc.)	11 - 17 years	NaN	NaN	...	35-44 years old	Man	No	Straight / Heterosexual	White or of European descent	I am deaf / hard of hearing	NaN	Appropriate in length	Neither easy nor difficult	NaN
4	5	I am a developer by profession	Independent contractor, freelancer, or self-em...	United Kingdom of Great Britain and Northern I...	NaN	England	Master’s degree (M.A., M.S., M.Eng., MBA, etc.)	5 - 10 years	Friend or family member	17	...	25-34 years old	Man	No	NaN	White or of European descent	None of the above	NaN	Appropriate in length	Easy	NaN

5 rows × 48 columns

In [4]:

result_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 83439 entries, 0 to 83438
Data columns (total 48 columns):
 #   Column                        Non-Null Count  Dtype  
---  ------                        --------------  -----  
 0   ResponseId                    83439 non-null  int64  
 1   MainBranch                    83439 non-null  object 
 2   Employment                    83323 non-null  object 
 3   Country                       83439 non-null  object 
 4   US_State                      14920 non-null  object 
 5   UK_Country                    4418 non-null   object 
 6   EdLevel                       83126 non-null  object 
 7   Age1stCode                    83243 non-null  object 
 8   LearnCode                     82963 non-null  object 
 9   YearsCode                     81641 non-null  object 
 10  YearsCodePro                  61216 non-null  object 
 11  DevType                       66484 non-null  object 
 12  OrgSize                       60726 non-null  object 
 13  Currency                      61080 non-null  object 
 14  CompTotal                     47183 non-null  float64
 15  CompFreq                      52150 non-null  object 
 16  LanguageHaveWorkedWith        82357 non-null  object 
 17  LanguageWantToWorkWith        76821 non-null  object 
 18  DatabaseHaveWorkedWith        69546 non-null  object 
 19  DatabaseWantToWorkWith        58299 non-null  object 
 20  PlatformHaveWorkedWith        52135 non-null  object 
 21  PlatformWantToWorkWith        41619 non-null  object 
 22  WebframeHaveWorkedWith        61707 non-null  object 
 23  WebframeWantToWorkWith        52095 non-null  object 
 24  MiscTechHaveWorkedWith        47055 non-null  object 
 25  MiscTechWantToWorkWith        38021 non-null  object 
 26  ToolsTechHaveWorkedWith       72537 non-null  object 
 27  ToolsTechWantToWorkWith       65480 non-null  object 
 28  NEWCollabToolsHaveWorkedWith  81234 non-null  object 
 29  NEWCollabToolsWantToWorkWith  73022 non-null  object 
 30  OpSys                         83294 non-null  object 
 31  NEWStuck                      83052 non-null  object 
 32  NEWSOSites                    83171 non-null  object 
 33  SOVisitFreq                   82413 non-null  object 
 34  SOAccount                     82525 non-null  object 
 35  SOPartFreq                    67553 non-null  object 
 36  SOComm                        82319 non-null  object 
 37  NEWOtherComms                 82828 non-null  object 
 38  Age                           82407 non-null  object 
 39  Gender                        82286 non-null  object 
 40  Trans                         80678 non-null  object 
 41  Sexuality                     73366 non-null  object 
 42  Ethnicity                     79464 non-null  object 
 43  Accessibility                 77603 non-null  object 
 44  MentalHealth                  76920 non-null  object 
 45  SurveyLength                  81711 non-null  object 
 46  SurveyEase                    81948 non-null  object 
 47  ConvertedCompYearly           46844 non-null  float64
dtypes: float64(2), int64(1), object(45)
memory usage: 30.6+ MB

Filtering the Dataset¶

In [5]:

result_df["MainBranch"].value_counts()

Out[5]:

I am a developer by profession                                                   58153
I am a student who is learning to code                                           12029
I am not primarily a developer, but I write code sometimes as part of my work     6578
I code primarily as a hobby                                                       4929
I used to be a developer by profession, but no longer am                          1237
None of these                                                                      513
Name: MainBranch, dtype: int64

We are primarily interested with those who code for the purpose of work. So we will isolate these two MainBranch: "I am a developer by profession" and "I am not primarily a developer, but I write code sometimes as part of my work".

In [6]:

result2_df: pd.DataFrame = result_df.loc[(result_df["MainBranch"]=="I am a developer by profession") | (result_df["MainBranch"]=="I am not primarily a developer, but I write code sometimes as part of my work")]
result2_df.head(5)

Out[6]:

	ResponseId	MainBranch	Employment	Country	US_State	UK_Country	EdLevel	Age1stCode	LearnCode	YearsCode	...	Age	Gender	Trans	Sexuality	Ethnicity	Accessibility	MentalHealth	SurveyLength	SurveyEase	ConvertedCompYearly
0	1	I am a developer by profession	Independent contractor, freelancer, or self-em...	Slovakia	NaN	NaN	Secondary school (e.g. American high school, G...	18 - 24 years	Coding Bootcamp;Other online resources (ex: vi...	NaN	...	25-34 years old	Man	No	Straight / Heterosexual	White or of European descent	None of the above	None of the above	Appropriate in length	Easy	62268.0
2	3	I am not primarily a developer, but I write co...	Student, full-time	Russian Federation	NaN	NaN	Bachelor’s degree (B.A., B.S., B.Eng., etc.)	11 - 17 years	Other online resources (ex: videos, blogs, etc...	NaN	...	18-24 years old	Man	No	Prefer not to say	Prefer not to say	None of the above	None of the above	Appropriate in length	Easy	NaN
3	4	I am a developer by profession	Employed full-time	Austria	NaN	NaN	Master’s degree (M.A., M.S., M.Eng., MBA, etc.)	11 - 17 years	NaN	NaN	...	35-44 years old	Man	No	Straight / Heterosexual	White or of European descent	I am deaf / hard of hearing	NaN	Appropriate in length	Neither easy nor difficult	NaN
4	5	I am a developer by profession	Independent contractor, freelancer, or self-em...	United Kingdom of Great Britain and Northern I...	NaN	England	Master’s degree (M.A., M.S., M.Eng., MBA, etc.)	5 - 10 years	Friend or family member	17	...	25-34 years old	Man	No	NaN	White or of European descent	None of the above	NaN	Appropriate in length	Easy	NaN
8	9	I am a developer by profession	Employed part-time	India	NaN	NaN	Bachelor’s degree (B.A., B.S., B.Eng., etc.)	18 - 24 years	Coding Bootcamp	6	...	25-34 years old	Man	No	NaN	South Asian	NaN	I have a concentration and/or memory disorder ...	Appropriate in length	Easy	NaN

5 rows × 48 columns

Drop the rows if DevType column is null.

In [7]:

print(result2_df.shape)
result2_df = result2_df.dropna(subset="DevType")
print(result2_df.shape)

(64731, 48)
(61602, 48)

In [8]:

result2_df["DevType"].value_counts()

Out[8]:

Developer, full-stack                                                                                                                                                                                                                                                                                                        8415
Developer, back-end                                                                                                                                                                                                                                                                                                          5378
Developer, front-end                                                                                                                                                                                                                                                                                                         2304
Developer, front-end;Developer, full-stack;Developer, back-end                                                                                                                                                                                                                                                               2108
Developer, full-stack;Developer, back-end                                                                                                                                                                                                                                                                                    1759
                                                                                                                                                                                                                                                                                                                             ... 
Developer, desktop or enterprise applications;Developer, full-stack;Developer, back-end;Database administrator;Developer, QA or test;DevOps specialist;Engineer, site reliability;System administrator;Educator                                                                                                                 1
Developer, mobile;Developer, front-end;Developer, desktop or enterprise applications;Developer, full-stack;Other (please specify):;Developer, back-end;Academic researcher;Database administrator;Developer, game or graphics;Developer, embedded applications or devices;DevOps specialist;Designer;System administrator       1
Developer, mobile;Developer, front-end;Developer, desktop or enterprise applications;Developer, full-stack;Developer, back-end;Database administrator;Developer, embedded applications or devices;Designer;System administrator;Educator                                                                                        1
Developer, front-end;Developer, full-stack;Developer, back-end;Academic researcher;Database administrator;DevOps specialist                                                                                                                                                                                                     1
Developer, mobile;Developer, desktop or enterprise applications;Data scientist or machine learning specialist;Developer, back-end;Engineering manager                                                                                                                                                                           1
Name: DevType, Length: 8159, dtype: int64

We can see that one person is allowed to choose more than one option for DevType column (separated by ;). Now, we will check options available for DevType column.

In [9]:

result2_df.loc[~result2_df["DevType"].str.contains(";"), "DevType"].unique()

Out[9]:

array(['Developer, mobile', 'Developer, front-end',
       'Data scientist or machine learning specialist',
       'Developer, back-end', 'Developer, full-stack',
       'Developer, game or graphics',
       'Developer, embedded applications or devices',
       'Developer, desktop or enterprise applications',
       'Data or business analyst', 'Engineer, data',
       'Academic researcher', 'Other (please specify):',
       'Engineering manager', 'DevOps specialist',
       'Senior Executive (C-Suite, VP, etc.)', 'Product manager',
       'Developer, QA or test', 'Designer', 'Scientist',
       'System administrator', 'Database administrator', 'Student',
       'Engineer, site reliability', 'Educator',
       'Marketing or sales professional'], dtype=object)

These are the available data related jobs for DevType options:

Data scientist or machine learning specialist
Data or business analyst
Engineer, data
Database administrator

All data related jobs contain the word "data".

In [10]:

data_job_df = result2_df.loc[result2_df["DevType"].str.contains("data", case=False)]
data_job_df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 12316 entries, 4 to 83436
Data columns (total 48 columns):
 #   Column                        Non-Null Count  Dtype  
---  ------                        --------------  -----  
 0   ResponseId                    12316 non-null  int64  
 1   MainBranch                    12316 non-null  object 
 2   Employment                    12316 non-null  object 
 3   Country                       12316 non-null  object 
 4   US_State                      2614 non-null   object 
 5   UK_Country                    758 non-null    object 
 6   EdLevel                       12304 non-null  object 
 7   Age1stCode                    12308 non-null  object 
 8   LearnCode                     12281 non-null  object 
 9   YearsCode                     12179 non-null  object 
 10  YearsCodePro                  11976 non-null  object 
 11  DevType                       12316 non-null  object 
 12  OrgSize                       12105 non-null  object 
 13  Currency                      12143 non-null  object 
 14  CompTotal                     9346 non-null   float64
 15  CompFreq                      10334 non-null  object 
 16  LanguageHaveWorkedWith        12257 non-null  object 
 17  LanguageWantToWorkWith        11452 non-null  object 
 18  DatabaseHaveWorkedWith        11231 non-null  object 
 19  DatabaseWantToWorkWith        9688 non-null   object 
 20  PlatformHaveWorkedWith        8771 non-null   object 
 21  PlatformWantToWorkWith        7194 non-null   object 
 22  WebframeHaveWorkedWith        9167 non-null   object 
 23  WebframeWantToWorkWith        7693 non-null   object 
 24  MiscTechHaveWorkedWith        9026 non-null   object 
 25  MiscTechWantToWorkWith        7749 non-null   object 
 26  ToolsTechHaveWorkedWith       10871 non-null  object 
 27  ToolsTechWantToWorkWith       9898 non-null   object 
 28  NEWCollabToolsHaveWorkedWith  12052 non-null  object 
 29  NEWCollabToolsWantToWorkWith  10930 non-null  object 
 30  OpSys                         12305 non-null  object 
 31  NEWStuck                      12287 non-null  object 
 32  NEWSOSites                    12299 non-null  object 
 33  SOVisitFreq                   12249 non-null  object 
 34  SOAccount                     12261 non-null  object 
 35  SOPartFreq                    10294 non-null  object 
 36  SOComm                        12244 non-null  object 
 37  NEWOtherComms                 12249 non-null  object 
 38  Age                           12212 non-null  object 
 39  Gender                        12189 non-null  object 
 40  Trans                         11984 non-null  object 
 41  Sexuality                     11025 non-null  object 
 42  Ethnicity                     11812 non-null  object 
 43  Accessibility                 11514 non-null  object 
 44  MentalHealth                  11475 non-null  object 
 45  SurveyLength                  12093 non-null  object 
 46  SurveyEase                    12125 non-null  object 
 47  ConvertedCompYearly           9272 non-null   float64
dtypes: float64(2), int64(1), object(45)
memory usage: 4.6+ MB

Drop the rows if the LanguageHaveWorkedWith column is not available.

In [11]:

print(data_job_df.shape)
data_job_df = data_job_df.dropna(subset="LanguageHaveWorkedWith")
print(data_job_df.shape)

(12316, 48)
(12257, 48)

In [12]:

data_job_df["DevType"].value_counts()

Out[12]:

Data scientist or machine learning specialist                                                                                                                                                                                               632
Engineer, data                                                                                                                                                                                                                              358
Engineer, data;Developer, back-end                                                                                                                                                                                                          316
Data or business analyst                                                                                                                                                                                                                    269
Data scientist or machine learning specialist;Data or business analyst                                                                                                                                                                      197
                                                                                                                                                                                                                                           ... 
Developer, mobile;Developer, full-stack;Database administrator;Data or business analyst;Senior Executive (C-Suite, VP, etc.)                                                                                                                  1
Developer, desktop or enterprise applications;Data scientist or machine learning specialist;Database administrator;Designer                                                                                                                   1
Developer, front-end;Developer, full-stack;Developer, back-end;Academic researcher;Database administrator;DevOps specialist                                                                                                                   1
Developer, mobile;Developer, front-end;Developer, desktop or enterprise applications;Developer, full-stack;Developer, back-end;Database administrator;Developer, embedded applications or devices;Designer;System administrator;Educator      1
Developer, mobile;Developer, desktop or enterprise applications;Data scientist or machine learning specialist;Developer, back-end;Engineering manager                                                                                         1
Name: DevType, Length: 4830, dtype: int64

Since the rows with more than one DevType might skew the language count results, we are going to include only the options below: "Data scientist or machine learning specialist", "Data or business analyst", "Engineer, data", "Database administrator"

In [13]:

data_job_df = data_job_df[(data_job_df["DevType"] == "Data scientist or machine learning specialist") | (data_job_df["DevType"] == "Data or business analyst") | (data_job_df["DevType"] == "Engineer, data") | (data_job_df["DevType"] == "Database administrator")]
data_job_df["DevType"].value_counts()

Out[13]:

Data scientist or machine learning specialist    632
Engineer, data                                   358
Data or business analyst                         269
Database administrator                            44
Name: DevType, dtype: int64

Data Transformation¶

Now we want to see the programming languages that are used by data related professions.

In [14]:

language_df: pd.DataFrame = data_job_df.loc[:,["DevType","LanguageHaveWorkedWith"]]
language_df

Out[14]:

	DevType	LanguageHaveWorkedWith
9	Data scientist or machine learning specialist	C++;Python
61	Data or business analyst	Python;R;VBA
77	Data scientist or machine learning specialist	HTML/CSS;Python;R
137	Data or business analyst	HTML/CSS;JavaScript;R;SQL
160	Engineer, data	Assembly;Python
...	...	...
83112	Data scientist or machine learning specialist	Bash/Shell;Perl;Python;R
83120	Data scientist or machine learning specialist	C;C++;Java;Kotlin
83203	Engineer, data	Bash/Shell;C#;Ruby;Rust
83216	Data or business analyst	HTML/CSS;JavaScript;SQL;VBA
83335	Engineer, data	Bash/Shell;Python;R

1303 rows × 2 columns

Similar to DevType, one person is allowed to choose more than one option for LanguageHaveWorkedWith column (separated by ;). Now, we will check the options available for DevType column.

In [15]:

languages = language_df.loc[~language_df["LanguageHaveWorkedWith"].str.contains(";"), "LanguageHaveWorkedWith"].unique()
for language in languages:
    print(language)

Matlab
Python
SQL
VBA
C#
R
C
Scala
Delphi
Perl
PHP
Java
C++
Clojure
Julia
APL
Swift
HTML/CSS

Now we will make a dataframe for each DevType ("Data scientist or machine learning specialist", "Data or business analyst", "Engineer, data", and "Database administrator").

In [16]:

ds_df = language_df[language_df["DevType"].str.contains("Data scientist or machine learning specialist")]
ds_df

Out[16]:

	DevType	LanguageHaveWorkedWith
9	Data scientist or machine learning specialist	C++;Python
77	Data scientist or machine learning specialist	HTML/CSS;Python;R
161	Data scientist or machine learning specialist	Matlab
209	Data scientist or machine learning specialist	C#;HTML/CSS;Matlab;Python;R;SQL
224	Data scientist or machine learning specialist	Python
...	...	...
82900	Data scientist or machine learning specialist	HTML/CSS;Python
82906	Data scientist or machine learning specialist	HTML/CSS;Python;R;SQL
82921	Data scientist or machine learning specialist	Python;R
83112	Data scientist or machine learning specialist	Bash/Shell;Perl;Python;R
83120	Data scientist or machine learning specialist	C;C++;Java;Kotlin

632 rows × 2 columns

In [17]:

da_df = language_df[language_df["DevType"].str.contains("Data or business analyst")]
de_df = language_df[language_df["DevType"].str.contains("Engineer, data")]
dba_df = language_df[language_df["DevType"].str.contains("Database administrator")]

First we create a new dataframe to store the languages and their count based on each DevType. We calculate the total occurrences for each DevType and the occurrences of each language using a for loop. C and C++ are not included in the for loop because str.count is based on Regex pattern, and they would cause error/miscount.

In [18]:

lang_count_df = pd.DataFrame()
languages = np.delete(languages,(languages == "C") | (languages == "C++"))

def add_lang_count(df: pd.DataFrame, job_type: str):
    lang_count_df.loc["total", job_type] = df.shape[0]
    for language in languages:
        lang_count_df.loc[language,job_type] = df["LanguageHaveWorkedWith"].str.count(language).sum()
    lang_count_df.loc["C", job_type] = df["LanguageHaveWorkedWith"].str.count("C[^\w+#]").sum()
    lang_count_df.loc["C++", job_type] = df["LanguageHaveWorkedWith"].str.count("C\+\+").sum()

In [19]:

add_lang_count(ds_df,"Data Scientist")
add_lang_count(da_df, "Data Analyst")
add_lang_count(de_df, "Data Engineer")
add_lang_count(dba_df, "Database Administrator")
lang_count_df

Out[19]:

	Data Scientist	Data Analyst	Data Engineer	Database Administrator
total	632.0	269.0	358.0	44.0
Matlab	55.0	13.0	29.0	1.0
Python	595.0	179.0	294.0	18.0
SQL	315.0	168.0	234.0	36.0
VBA	25.0	77.0	24.0	7.0
C#	42.0	34.0	46.0	11.0
R	219.0	81.0	58.0	9.0
Scala	40.0	2.0	89.0	1.0
Delphi	0.0	3.0	2.0	1.0
Perl	10.0	5.0	8.0	2.0
PHP	14.0	19.0	23.0	5.0
Java	218.0	115.0	230.0	15.0
Clojure	9.0	6.0	5.0	1.0
Julia	33.0	5.0	9.0	1.0
APL	0.0	5.0	3.0	3.0
Swift	8.0	4.0	4.0	4.0
HTML/CSS	113.0	78.0	86.0	11.0
C	89.0	20.0	49.0	8.0
C++	136.0	23.0	58.0	5.0

Data Analysis and Visualization¶

Import visualization libraries

In [20]:

import plotly.graph_objs as go
import plotly.express as px
import plotly.offline as pyo
pyo.init_notebook_mode()

We will add a new column to calculate the most used programming languages across all data related professions. Note that the sample size of data scientist is larger compared to other groups, and the sample of database administrator is smaller compared to the other groups.

In [21]:

total_ser = lang_count_df.sum(axis = 1)
lang_count_df["Total"] = total_ser

Normalize the data to percentage, and add a color code to each language.

In [22]:

normalize_df = lang_count_df.iloc[1:]/lang_count_df.loc["total"]
normalize_df["Color"] = "NA"

In [23]:

for n in range(normalize_df.shape[0]):
    normalize_df.iloc[n, -1] = px.colors.qualitative.Light24[n]
normalize_df

Out[23]:

	Data Scientist	Data Analyst	Data Engineer	Database Administrator	Total	Color
Matlab	0.087025	0.048327	0.081006	0.022727	0.075211	#FD3216
Python	0.941456	0.665428	0.821229	0.409091	0.833461	#00FE35
SQL	0.498418	0.624535	0.653631	0.818182	0.577897	#6A76FC
VBA	0.039557	0.286245	0.067039	0.159091	0.102072	#FED4C4
C#	0.066456	0.126394	0.128492	0.250000	0.102072	#FE00CE
R	0.346519	0.301115	0.162011	0.204545	0.281658	#0DF9FF
Scala	0.063291	0.007435	0.248603	0.022727	0.101305	#F6F926
Delphi	0.000000	0.011152	0.005587	0.022727	0.004605	#FF9616
Perl	0.015823	0.018587	0.022346	0.045455	0.019186	#479B55
PHP	0.022152	0.070632	0.064246	0.113636	0.046815	#EEA6FB
Java	0.344937	0.427509	0.642458	0.340909	0.443592	#DC587D
Clojure	0.014241	0.022305	0.013966	0.022727	0.016117	#D626FF
Julia	0.052215	0.018587	0.025140	0.022727	0.036838	#6E899C
APL	0.000000	0.018587	0.008380	0.068182	0.008442	#00B5F7
Swift	0.012658	0.014870	0.011173	0.090909	0.015349	#B68E00
HTML/CSS	0.178797	0.289963	0.240223	0.250000	0.221028	#C9FBE5
C	0.140823	0.074349	0.136872	0.181818	0.127398	#FF0092
C++	0.215190	0.085502	0.162011	0.113636	0.170376	#22FFA7

Extract the top 5 languages from each job type. sort_index() is used to give a better visualization result.

In [24]:

top_5_total = normalize_df[["Total","Color"]].sort_values("Total", ascending=False)[:5].sort_index()
top_5_ds = normalize_df[["Data Scientist", "Color"]].sort_values("Data Scientist", ascending=False)[:5].sort_index()
top_5_da = normalize_df[["Data Analyst", "Color"]].sort_values("Data Analyst", ascending=False)[:5].sort_index()
top_5_de = normalize_df[["Data Engineer", "Color"]].sort_values("Data Engineer", ascending=False)[:5].sort_index()
top_5_dba = normalize_df[["Database Administrator", "Color"]].sort_values("Database Administrator", ascending=False)[:5].sort_index()

top_5_list = [top_5_total, top_5_ds, top_5_da, top_5_de, top_5_dba]
for df in top_5_list:
    df.columns = ["Total", "Color"]

top_5_total

Out[24]:

	Total	Color
HTML/CSS	0.221028	#C9FBE5
Java	0.443592	#DC587D
Python	0.833461	#00FE35
R	0.281658	#0DF9FF
SQL	0.577897	#6A76FC

In [25]:

fig = go.Figure(go.Barpolar(
    r= top_5_total["Total"],
    theta= top_5_total.index,
    width=0.45,
    marker_color=top_5_total["Color"],
    marker_line_color="black",
    marker_line_width=2,
    opacity=0.7
))

fig.update_layout(
    template="ggplot2",
    polar = dict(
        radialaxis = dict(range=[0, 1.0], showticklabels=False, ticks=''),
        angularaxis = dict(showticklabels=True, ticks='')
    )
)

fig.show()

In [26]:

from plotly.subplots import make_subplots

fig = make_subplots(rows=2, cols=2, specs=[[{'type': 'polar'}]*2]*2, subplot_titles=("<b>Data Scientist</b>", "<b>Data Analyst</b>", "<b>Data Engineer</b>", "<b>Database Administrator</b>"))

fig.add_trace(go.Barpolar(
    r= top_5_ds["Total"],
    theta= top_5_ds.index,
    width=0.45,
    marker_color=top_5_ds["Color"],
    marker_line_color="black",
    marker_line_width=2,
    opacity=0.7
), 1, 1)
fig.add_trace(go.Barpolar(
    r= top_5_da["Total"],
    theta= top_5_da.index,
    width=0.45,
    marker_color=top_5_da["Color"],
    marker_line_color="black",
    marker_line_width=2,
    opacity=0.7
), 1, 2)
fig.add_trace(go.Barpolar(
    r= top_5_de["Total"],
    theta= top_5_de.index,
    width=0.45,
    marker_color=top_5_de["Color"],
    marker_line_color="black",
    marker_line_width=2,
    opacity=0.7
), 2, 1)
fig.add_trace(go.Barpolar(
    r= top_5_dba["Total"],
    theta= top_5_dba.index,
    width=0.45,
    marker_color=top_5_dba["Color"],
    marker_line_color="black",
    marker_line_width=2,
    opacity=0.7
), 2, 2)
fig.update_traces(hovertemplate='%{theta}: %{r:.0%}<extra></extra>')
fig.update_layout(
    font_color="darkblue",
    font_family="Helvetica",
    title_font_color="darkblue",
    title_font_family="Helvetica",
    title_font_size=24,
    margin = dict(t = 150),
    paper_bgcolor='#DEDCC6',
    height = 800,
    width = 1000,
    showlegend = False,
    title=dict(text="<b>Popular Languages for Data Professionals</b>"),
    template="ggplot2",
    polar = dict(
        radialaxis = dict(range=[0, 1.0], showline = False, showticklabels=False, ticks=''),
        angularaxis = dict(showticklabels=True, ticks='', linecolor='black', linewidth=1.5)
    ),
    polar2 = dict(
        radialaxis = dict(range=[0, 1.0], showline = False, showticklabels=False, ticks=''),
        angularaxis = dict(showticklabels=True, ticks='', linecolor='black', linewidth=1.5)
    ),
    polar3 = dict(
        radialaxis = dict(range=[0, 1.0], showline = False, showticklabels=False, ticks=''),
        angularaxis = dict(showticklabels=True, ticks='', linecolor='black', linewidth=1.5)
    ),
    polar4 = dict(
        radialaxis = dict(range=[0, 1.0], showline = False, showticklabels=False, ticks=''),
        angularaxis = dict(showticklabels=True, ticks='', linecolor='black', linewidth=1.5)
    )

)
fig.update_annotations(yshift=20)
fig.add_annotation({
    "x": 1.05,
    "y": -0.1,
    "font": {"size": 14},
    "text": "tmtsmrsl.github.io",
    "showarrow": False
})

fig.show()