2018 state of Haskell survey results

The 2018 state of Haskell survey opened at the start of November and closed a few days ago. It was the second annual state of Haskell survey. This year the survey was co-sponsored by Haskell Weekly and Haskell.org.

Last year the 2017 state of Haskell survey received 1,335 responses. This year we received almost four times as many. However a large number of the responses appeared to be submitted by a script. Of the 5,096 responses, 3,735 were identified as problematic and removed from the result set. That left 1,361 legitimate responses, which compares favorably to last year, especially when you consider that this year’s survey was much longer. To everyone that participated in the survey: Thank you! You feedback is hugely valuable.

I am extremely disappointed that someone decided to submit thousands of bogus survey responses. Apparently they wanted to tip the scales on a few divisive issues in the community. Fortunately their attack was not very sophisticated, so it was easy to identify and ignore their responses. (This function shows you exactly how the problematic responses were identified.) Even so, it has two profoundly negative impacts: One, it casts doubt on the survey responses as a whole; and two, it means that I’ll have to devote time to countering attacks like this in future surveys. That being said, I have confidence in the filtered survey results and I hope you do too.

This post will graph the answers to the multiple-choice questions. I will not attempt to summarize the answers to free-response questions. I encourage any interested parties to download the raw results, do their own analysis, and share their results. You can download the anonymized responses as a CSV, which is licensed under the ODbL 1.0 license.

# Submission date ^

Answer	Count	Percent
2018-11-01	513	37.7%
2018-11-02	310	22.8%
2018-11-03	118	8.7%
2018-11-04	75	5.5%
2018-11-05	63	4.6%
2018-11-06	56	4.1%
2018-11-07	32	2.4%
2018-11-08	44	3.2%
2018-11-09	22	1.6%
2018-11-10	12	0.9%
2018-11-11	16	1.2%
2018-11-12	28	2.1%
2018-11-13	35	2.6%
2018-11-14	37	2.7%

# Haskell usage ^

Answer	Count	Percent
1 heart	4	0.3%
2 hearts	1	0.1%
3 hearts	4	0.3%
4 hearts	4	0.3%
5 hearts	35	2.6%
6 hearts	16	1.2%
7 hearts	40	2.9%
8 hearts	84	6.2%
9 hearts	55	4.0%
10 hearts	501	36.8%

# Do you use Haskell? (single select) ^

Answer	Count	Percent
Yes	1202	88.3%
No, but I used to	103	7.6%
No, I never have	45	3.3%

# If you stopped using Haskell, how long did you use it before you stopped? (single select) ^

Answer	Count	Percent
Less than 1 day	6	0.4%
1 day to 1 week	11	0.8%
1 week to 1 month	20	1.5%
1 month to 1 year	66	4.8%
More than 1 year	53	3.9%

# If you do not use Haskell, why not? (multiple select) ^

Answer	Count	Percent
My company doesn't use Haskell	89	6.5%
Haskell is too hard to learn	44	3.2%
Haskell's documentation is not good enough	35	2.6%
Haskell lacks critical tools	31	2.3%
Haskell lacks critical libraries	30	2.2%
Haskell does not support the platforms I need	12	0.9%
Haskell's performance is not good enough	12	0.9%
Haskell lacks critical features	8	0.6%

# How long have you been using Haskell? (single select) ^

Answer	Count	Percent
Less than 1 day	3	0.2%
1 day to 1 week	7	0.5%
1 week to 1 month	35	2.6%
1 month to 1 year	219	16.1%
1 year to 2 years	234	17.2%
2 years to 3 years	213	15.7%
3 years to 4 years	147	10.8%
4 years to 5 years	112	8.2%
More than 5 years	330	24.2%

# How frequently do you use Haskell? (single select) ^

Answer	Count	Percent
Daily	490	36.0%
Weekly	422	31.0%
Monthly	249	18.3%
Yearly	58	4.3%
Rarely	80	5.9%

# How would you rate your proficiency in Haskell? (single select) ^

Answer	Count	Percent
Beginner	247	18.1%
Intermediate	533	39.2%
Advanced	354	26.0%
Expert	167	12.3%
Master	28	2.1%

# Where do you use Haskell? (multiple select) ^

Answer	Count	Percent
Home	1077	79.1%
Work	666	48.9%
School	153	11.2%

# Do you use Haskell at work? (single select) ^

Answer	Count	Percent
Yes, most of the time	389	28.6%
Yes, some of the time	318	23.4%
No, but my company does	20	1.5%
No	558	41.0%

# If you do not use Haskell at work, why not? (multiple select) ^

Answer	Count	Percent
My company doesn't use Haskell	438	32.2%
Haskell is too hard to learn	76	5.6%
It is too hard to hire Haskell developers	72	5.3%
Haskell lacks critical libraries	48	3.5%
Haskell lacks critical tools	41	3.0%
Haskell does not support the platforms I need	33	2.4%
Haskell's documentation is not good enough	31	2.3%
Haskell's performance is not good enough	12	0.9%
Haskell lacks critical features	8	0.6%

# Which programming languages other than Haskell are you fluent in? (multiple select) ^

Answer	Count	Percent
Python	659	48.4%
JavaScript	599	44.0%
Java	550	40.4%
C	512	37.6%
C++	394	28.9%
Shell	320	23.5%
C#	203	14.9%
Scala	198	14.5%
Ruby	174	12.8%
Rust	158	11.6%
TypeScript	146	10.7%
PHP	142	10.4%
Go	110	8.1%
Clojure	93	6.8%
Assembly	90	6.6%
Ocaml	87	6.4%
Perl	83	6.1%
R	65	4.8%
Lua	64	4.7%
F#	64	4.7%
Erlang	63	4.6%
Matlab	50	3.7%
Objective-C	47	3.5%
Swift	46	3.4%
Kotlin	46	3.4%
Groovy	21	1.5%
VB.NET	20	1.5%
VBA	15	1.1%
Julia	9	0.7%
Hack	5	0.4%

# Which types of software do you develop with Haskell? (multiple select) ^

Answer	Count	Percent
Command-line programs (CLI)	684	50.3%
API services (returning non-HTML)	532	39.1%
Data processing	437	32.1%
Libraries or frameworks	433	31.8%
Web services (returning HTML)	371	27.3%
Automation or scripts	358	26.3%
Agents or daemons	219	16.1%
Desktop programs (GUI)	108	7.9%

# Which industries do you use Haskell in? (multiple select) ^

Answer	Count	Percent
Web	329	24.2%
Banking or finance	185	13.6%
Education	184	13.5%
Commerce or retail	127	9.3%
Gaming	59	4.3%
Healthcare or medical	43	3.2%
Government	35	2.6%
Embedded	33	2.4%
Mobile	32	2.4%

# Projects ^

Answer	Count	Percent
1 heart	5	0.4%
2 hearts	0	0.0%
3 hearts	4	0.3%
4 hearts	13	1.0%
5 hearts	28	2.1%
6 hearts	40	2.9%
7 hearts	78	5.7%
8 hearts	115	8.4%
9 hearts	76	5.6%
10 hearts	356	26.2%

# How many Haskell projects do you contribute to? (single select) ^

Answer	Count	Percent
0	432	31.7%
1	168	12.3%
2 to 5	484	35.6%
6 to 10	94	6.9%
11 to 20	34	2.5%
More than 20	32	2.4%

# What is the total size of all the Haskell projects you contribute to? (single select) ^

Answer	Count	Percent
Less than 1,000 lines of code	207	15.2%
1,000 lines of code to 9,999 lines of code	311	22.9%
10,000 lines of code to 99,999 lines of code	252	18.5%
100,000 or more lines of code	115	8.4%

# Which platforms do you develop Haskell on? (multiple select) ^

Answer	Count	Percent
Linux	1003	73.7%
MacOS	496	36.4%
Windows	209	15.4%
BSD	44	3.2%

# Which platforms do you target? (multiple select) ^

Answer	Count	Percent
Linux	1075	79.0%
MacOS	399	29.3%
Windows	314	23.1%
BSD	80	5.9%
Android	38	2.8%
iOS	33	2.4%

# Compilers ^

Answer	Count	Percent
1 heart	4	0.3%
2 hearts	1	0.1%
3 hearts	3	0.2%
4 hearts	9	0.7%
5 hearts	25	1.8%
6 hearts	41	3.0%
7 hearts	50	3.7%
8 hearts	114	8.4%
9 hearts	80	5.9%
10 hearts	368	27.0%

# Which Haskell compilers do you use? (multiple select) ^

Answer	Count	Percent
GHC	1296	95.2%
GHCJS	131	9.6%
Eta	37	2.7%

# Which installation methods do you use for your Haskell compiler? (multiple select) ^

Answer	Count	Percent
Stack	916	67.3%
Nix	299	22.0%
Operating system package	277	20.4%
Haskell Platform	153	11.2%
Official binaries	149	10.9%
Source	93	6.8%
Minimal installer	35	2.6%

# Has upgrading your Haskell compiler broken your code in the last year? (single select) ^

Answer	Count	Percent
No	736	54.1%
Yes	437	32.1%

# How has upgrading your Haskell compiler broken your code in the past year? (multiple select) ^

Answer	Count	Percent
Incompatible dependencies	247	18.1%
Expected changes, such as the MonadFail proposal	225	16.5%
New warnings	76	5.6%
Compiler bugs	58	4.3%
Unexpected changes	54	4.0%

# Which versions of GHC do you use? ^

Answer	Count	Percent
HEAD	72	5.3%
8.6.x	329	24.2%
8.4.x	825	60.6%
8.2.x	318	23.4%
8.0.x	235	17.3%
7.10.x	78	5.7%
7.8.x	25	1.8%
7.6.x	7	0.5%
7.4.x	6	0.4%
7.2.x	2	0.1%
7.0.x	3	0.2%

# How do you feel about the new GHC release schedule? (single select) ^

Answer	Count	Percent
I like it	533	39.2%
I am indifferent	361	26.5%
I dislike it	50	3.7%
I was not aware of it	210	15.4%

# Which GHC language extensions would you like to be enabled by default? (multiple select) ^

Answer	Count	Percent
OverloadedStrings	486	35.7%
LambdaCase	327	24.0%
DeriveGeneric	263	19.3%
DeriveFunctor	260	19.1%
BangPatterns	254	18.7%
ScopedTypeVariables	254	18.7%
GADTs	251	18.4%
FlexibleInstances	223	16.4%
FlexibleContexts	217	15.9%
DeriveFoldable	192	14.1%
RankNTypes	187	13.7%
TupleSections	183	13.4%
MultiParamTypeClasses	178	13.1%
GeneralisedNewtypeDeriving	178	13.1%
DeriveTraversable	171	12.6%
DataKinds	144	10.6%
TypeApplications	138	10.1%
TypeOperators	136	10.0%
KindSignatures	135	9.9%
TypeFamilies	127	9.3%
ApplicativeDo	122	9.0%
StandaloneDeriving	117	8.6%
ConstraintKinds	116	8.5%
DerivingVia	116	8.5%
ViewPatterns	113	8.3%
InstanceSigs	113	8.3%
MultiWayIf	111	8.2%
RecordWildCards	110	8.1%
FunctionalDependencies	101	7.4%
DeriveDataTypeable	100	7.3%
EmptyCase	92	6.8%
NamedFieldPuns	87	6.4%
EmptyDataDecls	83	6.1%
ExplicitForAll	83	6.1%
DerivingStrategies	76	5.6%
DeriveAnyClass	74	5.4%
BinaryLiterals	74	5.4%
DefaultSignatures	69	5.1%
NumericUnderscores	67	4.9%
PatternSynonyms	64	4.7%
DuplicateRecordFields	62	4.6%
ExistentialQuantification	57	4.2%
OverloadedLists	55	4.0%
PolyKinds	54	4.0%
GADTSyntax	54	4.0%
BlockArguments	53	3.9%
NoImplicitPrelude	53	3.9%
DeriveLift	47	3.5%
PatternGuards	42	3.1%
AutoDeriveTypeable	39	2.9%
UnicodeSyntax	38	2.8%
TypeSynonymInstances	37	2.7%
PartialTypeSignatures	36	2.6%
NoMonomorphismRestriction	34	2.5%
TypeInType	33	2.4%
TemplateHaskell	31	2.3%
DisambiguateRecordFields	30	2.2%
TypeFamilyDependencies	29	2.1%
QuasiQuotes	28	2.1%
RecordPuns	25	1.8%
PackageImports	24	1.8%
QuantifiedConstraints	22	1.6%
AllowAmbiguousTypes	22	1.6%
Arrows	21	1.5%
EmptyDataDeriving	21	1.5%
NamedWildCards	20	1.5%
NegativeLiterals	19	1.4%
ParallelListComp	19	1.4%
DoAndIfThenElse	19	1.4%
MonadComprehensions	18	1.3%
OverloadedLabels	18	1.3%
RecursiveDo	17	1.2%
NumDecimals	17	1.2%
NullaryTypeClasses	17	1.2%
PatternSignatures	16	1.2%
Rank2Types	16	1.2%
ForeignFunctionInterface	15	1.1%
UnboxedTuples	15	1.1%
StarIsType	14	1.0%
RoleAnnotations	14	1.0%
MagicHash	13	1.0%
LiberalTypeSynonyms	13	1.0%
CPP	12	0.9%
HexFloatLiterals	12	0.9%
MonadFailDesugaring	11	0.8%
StrictData	11	0.8%
ExplicitNamespaces	11	0.8%
UnboxedSums	11	0.8%
UndecidableInstances	10	0.7%
ImplicitParams	9	0.7%
ConstrainedClassMethods	9	0.7%
DoRec	9	0.7%
OverlappingInstances	8	0.6%
Strict	7	0.5%
DatatypeContexts	6	0.4%
CApiFFI	6	0.4%
TransformListComp	5	0.4%
TemplateHaskellQuotes	5	0.4%
PolymorphicComponents	5	0.4%
RelaxedPolyRec	5	0.4%
ExtendedDefaultRules	4	0.3%
PostfixOperators	4	0.3%
NPlusKPatterns	4	0.3%
ParallelArrays	3	0.2%
ImpredicativeTypes	3	0.2%
InterruptibleFFI	3	0.2%
AlternativeLayoutRule	2	0.1%
NoTraditionalRecordSyntax	2	0.1%
RebindableSyntax	2	0.1%
UnliftedFFITypes	2	0.1%
UndecidableSuperClasses	2	0.1%
RelaxedLayout	2	0.1%
AlternativeLayoutRuleTransitional	1	0.1%
MonoLocalBinds	1	0.1%
JavaScriptFFI	1	0.1%
GHCForeignImportPrim	1	0.1%

# How important do you feel it would be to have a new version of the Haskell standard? (single select) ^

Answer	Count	Percent
Not at all important	68	5.0%
Slightly important	135	9.9%
Moderately important	370	27.2%
Very important	302	22.2%
Extremely important	148	10.9%

# Tooling ^

Answer	Count	Percent
1 heart	4	0.3%
2 hearts	8	0.6%
3 hearts	1	0.1%
4 hearts	17	1.2%
5 hearts	28	2.1%
6 hearts	35	2.6%
7 hearts	70	5.1%
8 hearts	97	7.1%
9 hearts	81	6.0%
10 hearts	321	23.6%

# Which build tools do you use for Haskell? (multiple select) ^

Answer	Count	Percent
Stack	995	73.1%
Cabal	654	48.1%
Nix	311	22.9%
Make	106	7.8%
Shake	75	5.5%
ghc-pkg	69	5.1%
Bazel	24	1.8%
Mafia	13	1.0%

# Which editors do you use for Haskell? (multiple select) ^

Answer	Count	Percent
Emacs	523	38.4%
Vi	471	34.6%
Visual Studio Code	329	24.2%
Atom	112	8.2%
Sublime Text	83	6.1%
IntelliJ IDEA	62	4.6%
Notepad++	19	1.4%

# Which version control systems do you use for Haskell? (multiple select) ^

Answer	Count	Percent
Git	1236	90.8%
Darcs	42	3.1%
Mercurial	29	2.1%
Subversion	11	0.8%
Pijul	10	0.7%
Fossil	8	0.6%

# Where do you get Haskell packages from? (multiple select) ^

Answer	Count	Percent
Stackage	867	63.7%
Hackage	831	61.1%
Source	305	22.4%
Nix	273	20.1%

# Which libraries do you use to test Haskell code? (multiple select) ^

Answer	Count	Percent
QuickCheck	696	51.1%
Hspec	537	39.5%
HUnit	333	24.5%
Tasty	323	23.7%
Hedgehog	162	11.9%
SmallCheck	58	4.3%
Haskell Test Framework	53	3.9%
LeanCheck	4	0.3%

# Which libraries do you use to benchmark Haskell code? (multiple select) ^

Answer	Count	Percent
Criterion	446	32.8%
Bench	54	4.0%
Gauge	34	2.5%

# Infrastructure ^

Answer	Count	Percent
1 heart	6	0.4%
2 hearts	5	0.4%
3 hearts	4	0.3%
4 hearts	7	0.5%
5 hearts	26	1.9%
6 hearts	33	2.4%
7 hearts	69	5.1%
8 hearts	107	7.9%
9 hearts	62	4.6%
10 hearts	309	22.7%

# Which tools do you use to deploy Haskell applications? (multiple select) ^

Answer	Count	Percent
Static binaries	510	37.5%
Docker images	402	29.5%
Dynamic binaries	189	13.9%
Nix expressions	183	13.4%

# Where do you deploy Haskell applications? (multiple select) ^

Answer	Count	Percent
Self or company owned servers	504	37.0%
Amazon Web Services (AWS)	328	24.1%
Google Cloud	81	6.0%
Digital Ocean	81	6.0%
Heroku	53	3.9%
Microsoft Azure	15	1.1%

# Community ^

Answer	Count	Percent
1 heart	2	0.1%
2 hearts	3	0.2%
3 hearts	5	0.4%
4 hearts	8	0.6%
5 hearts	22	1.6%
6 hearts	30	2.2%
7 hearts	52	3.8%
8 hearts	93	6.8%
9 hearts	68	5.0%
10 hearts	327	24.0%

# Where do you interact with the Haskell community? (multiple select) ^

Answer	Count	Percent
Reddit	699	51.4%
GitHub	590	43.4%
Twitter	392	28.8%
IRC	348	25.6%
Stack Overflow	266	19.5%
Meetups	250	18.4%
Mailing lists	248	18.2%
Slack	225	16.5%
Conferences (commercial)	206	15.1%
Conferences (academic)	157	11.5%
Discord	110	8.1%
Gitter	65	4.8%
Lobsters	65	4.8%
Telegram	42	3.1%
Mastodon	27	2.0%
Matrix/Riot	13	1.0%

# Which of the following Haskell topics would you like to see more written about? (multiple select) ^

Answer	Count	Percent
Best practices	563	41.4%
Design patterns	375	27.6%
Application architectures	375	27.6%
Performance analysis	312	22.9%
Library walkthroughs	296	21.7%
Tooling choices	271	19.9%
Debugging how-tos	245	18.0%
Production infrastructure	233	17.1%
Case studies	213	15.7%
Web development	189	13.9%
Algorithm implementations	174	12.8%
GUIs	163	12.0%
Beginner fundamentals	156	11.5%
Machine learning	156	11.5%
Project setup	134	9.8%
Project maintenance	125	9.2%
Game development	123	9.0%
Mobile development	93	6.8%
Comparisons to other languages	60	4.4%

# Feelings ^

Answer	Count	Percent
1 heart	3	0.2%
2 hearts	1	0.1%
3 hearts	5	0.4%
4 hearts	9	0.7%
5 hearts	17	1.2%
6 hearts	36	2.6%
7 hearts	58	4.3%
8 hearts	96	7.1%
9 hearts	74	5.4%
10 hearts	342	25.1%

# I feel welcome in the Haskell community. ^

Answer	Count	Percent
Strongly disagree	11	0.8%
Disagree	42	3.1%
Neutral	226	16.6%
Agree	502	36.9%
Strongly agree	449	33.0%

# I am satisfied with Haskell as a language. ^

Answer	Count	Percent
Strongly disagree	9	0.7%
Disagree	43	3.2%
Neutral	109	8.0%
Agree	587	43.1%
Strongly agree	516	37.9%

# I am satisfied with Haskell's compilers, such as GHC. ^

Answer	Count	Percent
Strongly disagree	5	0.4%
Disagree	36	2.6%
Neutral	96	7.1%
Agree	626	46.0%
Strongly agree	471	34.6%

# I am satisfied with Haskell's build tools, such as Cabal. ^

Answer	Count	Percent
Strongly disagree	34	2.5%
Disagree	144	10.6%
Neutral	323	23.7%
Agree	539	39.6%
Strongly agree	162	11.9%

# I am satisfied with Haskell's package repositories, such as Hackage. ^

Answer	Count	Percent
Strongly disagree	9	0.7%
Disagree	67	4.9%
Neutral	204	15.0%
Agree	653	48.0%
Strongly agree	256	18.8%

# I can find Haskell libraries for the things that I need. ^

Answer	Count	Percent
Strongly disagree	12	0.9%
Disagree	92	6.8%
Neutral	256	18.8%
Agree	687	50.5%
Strongly agree	159	11.7%

# I think Haskell libraries are high quality. ^

Answer	Count	Percent
Strongly disagree	5	0.4%
Disagree	49	3.6%
Neutral	300	22.0%
Agree	588	43.2%
Strongly agree	241	17.7%

# I have a good understanding of Haskell best practices. ^

Answer	Count	Percent
Strongly disagree	53	3.9%
Disagree	269	19.8%
Neutral	376	27.6%
Agree	419	30.8%
Strongly agree	106	7.8%

# I think Haskell libraries are well documented. ^

Answer	Count	Percent
Strongly disagree	127	9.3%
Disagree	353	25.9%
Neutral	421	30.9%
Agree	265	19.5%
Strongly agree	39	2.9%

# I can easily compare competing Haskell libraries to select the best one. ^

Answer	Count	Percent
Strongly disagree	131	9.6%
Disagree	441	32.4%
Neutral	368	27.0%
Agree	217	15.9%
Strongly agree	34	2.5%

# I think that Haskell libraries are easy to use. ^

Answer	Count	Percent
Strongly disagree	43	3.2%
Disagree	217	15.9%
Neutral	546	40.1%
Agree	340	25.0%
Strongly agree	45	3.3%

# I think that Haskell libraries provide a stable API. ^

Answer	Count	Percent
Strongly disagree	11	0.8%
Disagree	93	6.8%
Neutral	405	29.8%
Agree	556	40.9%
Strongly agree	82	6.0%

# I think that Haskell libraries work well together. ^

Answer	Count	Percent
Strongly disagree	18	1.3%
Disagree	105	7.7%
Neutral	396	29.1%
Agree	522	38.4%
Strongly agree	117	8.6%

# I think that software written in Haskell is easy to maintain. ^

Answer	Count	Percent
Strongly disagree	7	0.5%
Disagree	31	2.3%
Neutral	132	9.7%
Agree	485	35.6%
Strongly agree	523	38.4%

# Once my Haskell program compiles, it generally does what I intended. ^

Answer	Count	Percent
Strongly disagree	7	0.5%
Disagree	57	4.2%
Neutral	123	9.0%
Agree	649	47.7%
Strongly agree	384	28.2%

# I think that Haskell libraries perform well. ^

Answer	Count	Percent
Strongly disagree	5	0.4%
Disagree	47	3.5%
Neutral	320	23.5%
Agree	638	46.9%
Strongly agree	142	10.4%

# Haskell's performance meets my needs. ^

Answer	Count	Percent
Strongly disagree	14	1.0%
Disagree	60	4.4%
Neutral	159	11.7%
Agree	585	43.0%
Strongly agree	367	27.0%

# I can easily reason about the performance of my Haskell code. ^

Answer	Count	Percent
Strongly disagree	114	8.4%
Disagree	374	27.5%
Neutral	418	30.7%
Agree	240	17.6%
Strongly agree	39	2.9%

# I would recommend using Haskell to others. ^

Answer	Count	Percent
Strongly disagree	5	0.4%
Disagree	32	2.4%
Neutral	95	7.0%
Agree	412	30.3%
Strongly agree	682	50.1%

# I would prefer to use Haskell for my next new project. ^

Answer	Count	Percent
Strongly disagree	15	1.1%
Disagree	42	3.1%
Neutral	88	6.5%
Agree	315	23.1%
Strongly agree	765	56.2%

# Haskell is working well for my team. ^

Answer	Count	Percent
Strongly disagree	26	1.9%
Disagree	47	3.5%
Neutral	261	19.2%
Agree	222	16.3%
Strongly agree	276	20.3%

# Haskell is critical to my company's success. ^

Answer	Count	Percent
Strongly disagree	132	9.7%
Disagree	113	8.3%
Neutral	245	18.0%
Agree	153	11.2%
Strongly agree	186	13.7%

# As a candidate, I can easily find Haskell jobs. ^

Answer	Count	Percent
Strongly disagree	219	16.1%
Disagree	305	22.4%
Neutral	283	20.8%
Agree	103	7.6%
Strongly agree	45	3.3%

# As a hiring manager, I can easily find qualified Haskell candidates. ^

Answer	Count	Percent
Strongly disagree	69	5.1%
Disagree	113	8.3%
Neutral	274	20.1%
Agree	70	5.1%
Strongly agree	33	2.4%

# Demographics ^

Answer	Count	Percent
1 heart	8	0.6%
2 hearts	0	0.0%
3 hearts	5	0.4%
4 hearts	6	0.4%
5 hearts	22	1.6%
6 hearts	29	2.1%
7 hearts	54	4.0%
8 hearts	94	6.9%
9 hearts	74	5.4%
10 hearts	310	22.8%

# Which country do you live in? (single select) ^

Answer	Count	Percent
United States	364	26.7%
United Kingdom	107	7.9%
Germany	107	7.9%
Australia	70	5.1%
France	55	4.0%
Russia	49	3.6%
Canada	44	3.2%
Sweden	36	2.6%
Netherlands	34	2.5%
India	24	1.8%
Japan	21	1.5%
Italy	20	1.5%
Poland	20	1.5%
Finland	19	1.4%
Switzerland	19	1.4%
Austria	17	1.2%
Norway	16	1.2%
Czechia	15	1.1%
Denmark	14	1.0%
Belgium	13	1.0%
Spain	13	1.0%
New Zealand	11	0.8%
Brazil	10	0.7%
China	9	0.7%
Ireland	8	0.6%
Israel	8	0.6%
Ukraine	8	0.6%
Portugal	7	0.5%
Croatia	6	0.4%
Belarus	5	0.4%
Romania	5	0.4%
Argentina	4	0.3%
Bulgaria	4	0.3%
Greece	4	0.3%
Indonesia	4	0.3%
Latvia	4	0.3%
South Africa	4	0.3%
Turkey	4	0.3%
Ecuador	3	0.2%
Estonia	3	0.2%
Hungary	3	0.2%
Kenya	3	0.2%
Singapore	3	0.2%
Mexico	2	0.1%
Thailand	2	0.1%
Algeria	1	0.1%
Armenia	1	0.1%
Azerbaijan	1	0.1%
Bolivia	1	0.1%
Cameroon	1	0.1%
Chile	1	0.1%
Colombia	1	0.1%
Egypt	1	0.1%
Guatemala	1	0.1%
Iceland	1	0.1%
Iran	1	0.1%
Kazakhstan	1	0.1%
Lebanon	1	0.1%
Malta	1	0.1%
Nepal	1	0.1%
Nigeria	1	0.1%
Pakistan	1	0.1%
Paraguay	1	0.1%
Peru	1	0.1%
Philippines	1	0.1%
Serbia	1	0.1%
Slovakia	1	0.1%
Slovenia	1	0.1%
South Korea	1	0.1%
Syria	1	0.1%
Uganda	1	0.1%
Vietnam	1	0.1%

# How old are you? (single select) ^

Answer	Count	Percent
Under 18 years old	13	1.0%
18 to 24 years old	200	14.7%
25 to 34 years old	608	44.7%
35 to 44 years old	283	20.8%
45 to 54 years old	80	5.9%
55 to 64 years old	26	1.9%
65 years or older	8	0.6%

# What is your gender? (single select) ^

Answer	Count	Percent
Male	1119	82.2%
Female	49	3.6%
Non-binary	29	2.1%

# Do you identify as transgender? (single select) ^

Answer	Count	Percent
No	1079	79.3%
Yes	35	2.6%

# Are you a student? (single select) ^

Answer	Count	Percent
No	987	72.5%
Yes, full time	158	11.6%
Yes, part time	68	5.0%

# What is the highest level of education you have completed? (single select) ^

Answer	Count	Percent
Less than high school diploma	22	1.6%
High school diploma	101	7.4%
Some college	131	9.6%
Associate degree	24	1.8%
Bachelor's degree	432	31.7%
Master's degree	338	24.8%
Professional degree	23	1.7%
Doctoral degree	145	10.7%

# What is your employment status? (single select) ^

Answer	Count	Percent
Employed full time	876	64.4%
Self employed	124	9.1%
Employed part time	89	6.5%
Not employed, but looking for work	67	4.9%
Not employed, and not looking for work	52	3.8%
Retired	4	0.3%

# How large is the company you work for? (single select) ^

Answer	Count	Percent
Fewer than 10 employees	205	15.1%
10 to 99 employees	363	26.7%
100 to 999 employees	225	16.5%
1,000 or more employees	276	20.3%

# How many years have you been coding? (single select) ^

Answer	Count	Percent
0 to 4 years	95	7.0%
5 to 9 years	315	23.1%
10 to 14 years	313	23.0%
15 to 19 years	208	15.3%
20 to 24 years	133	9.8%
25 to 29 years	69	5.1%
30 or more years	94	6.9%

# How many years have you been coding professionally? (single select) ^

Answer	Count	Percent
0 to 4 years	397	29.2%
5 to 9 years	353	25.9%
10 to 19 years	305	22.4%
20 to 29 years	102	7.5%
30 or more years	26	1.9%

# Do you code as a hobby? (single select) ^

Answer	Count	Percent
Yes	1183	86.9%
No	59	4.3%

# Have you contributed to any open source projects? (single select) ^

Answer	Count	Percent
Yes	1031	75.8%
No	206	15.1%

# Meta survey ^

Answer	Count	Percent
1 heart	5	0.4%
2 hearts	3	0.2%
3 hearts	3	0.2%
4 hearts	5	0.4%
5 hearts	27	2.0%
6 hearts	26	1.9%
7 hearts	45	3.3%
8 hearts	95	7.0%
9 hearts	67	4.9%
10 hearts	321	23.6%

# Did you take last year's survey? (single select) ^

Answer	Count	Percent
No	616	45.3%
Yes	322	23.7%
I don't remember	294	21.6%

# How did you hear about this survey? (single select) ^

Answer	Count	Percent
Reddit	394	28.9%
Twitter	296	21.7%
Mailing list	121	8.9%
Haskell Weekly	113	8.3%
Lobsters	78	5.7%
Slack	65	4.8%
Discord	49	3.6%
Other	48	3.5%
In person	30	2.2%
IRC	12	0.9%
Telegram	12	0.9%
Gitter	6	0.4%
Matrix/Riot	2	0.1%
Mastodon	2	0.1%