Does anyone have access to the original paper? From the look at this article, I am extremely curious as to how they quantified what they defined as "community smells". From the article:
----
For the study, the authors looked at four types of community smells:
1. Organizational Silos. Groups in the community don’t communicate with each other, except through one or two of their respective members.
2. Black Clouds. Community members are overloaded with information because communication isn’t structured well.
3. Lone Wolves. Defiant, disrespectful contributors won’t listen to others.
4. Bottlenecks. One team member wants to be the only way information moves between sub-communities.
Previous studies have shown that women are fundamental to reducing community smells and avoiding accumulating social debt for teams in other industries. But these studies show that women increase team efficiency and organizational quality on software engineering teams specifically. Teams with no women have 7 mean community smells compared to 3 for teams with at least one woman. These results are statistically significant (p-value < 0.05) and have a large effect size (d=0.68).
Has this been proof-read? There's a grammatical error in the second sentence on the first page.
> we found [...] a strictly connection between gender diversity and communication patterns
Obviously this doesn't necessarily reflect the quality of the research but it is jarring to the reader and makes one question what else about the paper has not been checked over before publication
I'm not sure. I found it by Googling. I think, given there are several authors listed that, yes, it has been proof-read.
Generally I wouldn't consider the occasional typo to be indicative of the quality of the underlying research. Zooming in on a minor detail in an attempt to paint the paper negatively seems needlessly pedantic, but that's just my opinion.
Possibly, but being charitable, this is a preprint written by someone how specializes in software engineering, and it is written in their second language.
A typo seems like a fair mistake in these circumstances.
"As an important output of our survey, participants highlighted that besides technical expertise good communication skills are fundamental for developers within a software team ."
Absolutely true in my experience. A high-performing team of majority women coached me to have our teams write "ground rules" for team communication. The improvement was a big leap forward. Many example ground rules are in my git repo:
> good communication skills are fundamental for developers within a software team
This is absolutely true. Fundamentally, a great engineer is doing a TON of communicating -- writing technical designs, whiteboarding different approaches, explaining reasoning in code reviews, breaking down tasks for more junior engineers, coaching and mentoring others, interviewing... and about a million other things. The role of communication only increases as the problem complexity and scope increases.
This is dangerous junk science and is only having a counterproductive effect on gender relations by cheapening the achievements of women.
The same goes for other minorities who are treated like lesser people who need handouts.
This is the softest of sciences, minimally quantitative, non-experimentally verifiable and therefore subject to massive institutional bias - the same way that proponents of forced equity believe that the institutions they are fighting with have developed their own biases.
Equality of opportunity does not imply equality of outcome. But it doesn't matter - this post will likely be flagged because our own institution is increasingly biased toward certain unacceptable viewpoints - to the degree that we forbid discussion on nearly every popular forum. But you're only emboldening and recruiting extremists when you silence an entire alternative point of view.
Edit: Let me explain further, and justify my use of the word "dangerous." Let's forget for a moment the specific topic at hand. Imagine a field of study which is strongly influencing corporate and, local, state, and federal policy, which is simultaneously immune from criticism because of cultural norms.
That is where all of these pro-diversity and pro-inclusion studies are coming from. I'm not necessarily suggesting that they're wrong - only that systematically, institutionally, this is poor science and possibly leading society down a suboptimal path.
One side of the diversity and inclusion discussion is forbidden. We are forced, top down, to take it as a given that this is how things should be done. I cannot support alternative viewpoints with sources because researchers who would publish anything against the grain would be, in modern parlance, cancelled.
Here's the study [1]. If you want to discuss it, address specifics from it. Right now, you are simply coming in and spouting your personal political opinions in a detached and needlessly argumentative way.
Thank you for linking. Now I can confirm that the study is, frankly, bullshit. Ignoring the ridiculous sample size of 34, the "smells" are totally subjective. In fact one could argue in favor of each of them - this reminds me of open concept cargo culting:
Organizational Silo. Siloed
groups in the community that do
not communicate with each other,
except through one or two of their
respective members;
>1. Organizational Silo. Siloed
groups in the community that do
not communicate with each other,
except through one or two of their
respective members;
If I'm working on team A I don't need to know what literally every person on every other team is doing.
>2. Black Cloud. There is an
excessive information overload due
to the lack of structured
communication. This might lead to
a huge increase of data exchanges
across a community;
What? Lack of communication leads to excessive information? That doesn't even make sense.
>3. Lone Wolf. Unsanctioned or
defiant contributors who work in an
irrespective manner or regardless
of their team;
Where I'm from we call this initiative. I've been rewarded for showing it my entire life. This is not black and white.
>4. Radio Silence. One team member
interposes themselves into every
formal communication across two
or more sub-communities with little
or no flexibility to introduce other
parallel channels.
I don't even understand what this means.
None of the first three points are black and white, as they make them out to be. This is pure ideologically driven data mining playing off the stereotype of women being better communicators. Ironic.
Do you know of any studies on this topic that aren't "junk science", or do you just not like the idea put forward here?
In general it doesn't seem far-fetched to me that women might be better communicators than men in general, but I don't have any hard evidence for or against this.
It's likely going to get flagged because you discard the entire article as junk science without even vaguely addressing the points of the article or study and reverting to the same tired, outdated, and in this case irrelevant to the article, talking points.
I think it is you and not the article who is showing immense bias.
Here's an interesting summary of how usage of woke & SJW terminology dramatically increased in NYT publications since 2010 (alongside its increase in other forms of media).
It's a "divide and conquer" social engineering strategy to get regular people fighting each other based on in-borne traits (via tribalist psychological manipulation -- pervasive in our entertainment & political media currently) rather than teaming up to fight corporate control of politics, media, & the military industrial complex.
On one hand women have innately different traits than men (but only good ones, except being physically stronger).
On the other, women and MTF trans are the same because it’s all about how you feel and any differences are because of the patriarchy.
Which is it? You can’t have it both ways.
Hopefully such junk science stops being published. If there is a way of communicating that’s better for both guys and girls, and some of both are capable of doing it, then it can be taught to both with most learning it.
Unless you’re going to claim most guys are simply incapable of reading social queues, or female social queues (excluding sexual attraction, where intent is often hidden).
Women are more people-oriented while men are more thing-oriented, on average (1, 2).
I don't think this is grounds for "All men should communicate like women" or
"Women's style of communication is more productive in [name some male-oriented group]", or "We should prioritize women's communication style in [highly technical groups which rarely include women due to biologically factual statistical differences in occupational orientation ^]", which is a dubious proposition.
(^ = differences between women and women in interest levels and at the tails of achievement in "thing"/"systems" oriented technology occupations -- see citations 1 & 2)
Notice the pattern.
Look at how much usage of words related to "patriarchy", "misogyny", "institutional racism", "diversity and inclusion", "social justice" have increased in NYT and other mainstream media sources from 2010-2020 (3, 4).
We see this again and again in media. Here is some hard data on this mentality and its pervasion of politics, media, entertainment, -- an obvious social engineering effort to anyone paying attention over the last 10-15 years.
Interestingly enough OP's particular study would not be here on HN if it had any other result.
I highly doubt a sample size of 34 for something as ridiculous sounding as "community smell" is going to be in anyway groundbreaking, at least when compared to actual substantial research on the topic of male vs female occupational interests.
The 2nd study here has 500,000 participants, for example. Not 34.
Bottom line: Identity politics is a worthless, and came into vogue in order to dissolve the Occupy Wallstreet movement which sprang up in the aftermath of the 2008-early 2010's Recession. (5).
[3] The NYTimes is Woke "Many trends develop over decades but I’ve never seen change so rapid as the breathtaking success of what one might call social justice concerns. Beginning around 2010-2014 there appears to have been a inflection point. Here from Zach Goldberg on twitter are various words drawn from Lexis-Nexis."
https://marginalrevolution.com/marginalrevolution/2019/06/th...
[4] Zach Goldberg on twitter "1/n Spent some time on LexisNexis over the weekend. Depending on your political orientation, what follows will either disturb or encourage you. But regardless of political orientation, I'm sure we can all say 'holy fucking shit'" https://twitter.com/ZachG932/status/1133440945201061888
[5] Discussion on how identity politics conveniently sprang up as a counter-movement to deliberately disrupt up the Occupy Wallstreet movement, which is corroborated by the timelines of sources 3 and 4 above.
https://www.reddit.com/r/conspiracy/comments/1wecyn/identity...
Maybe they are, maybe they aren't, but this study doesn't say anything about that. It only compares all male teams to mixed gender teams. Without comparing all male to all female teams we can't draw that conclusion. It could be that simply making a team more heterogeneous along any significant axis would achieve the same effect and gender happens to be a proxy for one such axis.
Which is exactly what the conclusions of the study imply...
In which case, if men and women do not on average perform equally at given tasks, there's no reason to expect gender parity or even pay parity in meritocratic systems.
That's an unspoken contradiction inmplied by many of these studies. But I haven't seen any researcher mention it recently, for increasingly obvious reasons.
Well, they seem to be better at communicating, so maybe they should be paid MORE.
But regardless, even if there’s a difference, we have to consider nature vs. nurture. I think men could be brought up to be more communicative, but that’s just me.
Furthermore, it’s absurd to say the pay gap somehow reflects different skillsets. We’re talking about unequal pay for the same work. Of course women should be paid the same as men doing the same work as they are.
There are various styles of communication (such as Direct and Indirect) and in various contexts (from highly technical, objective, to highly emotional, subjective) and with various focuses (mission oriented vs. exploratory brainstorming vs. goal-less sharing, etc.)
This idea that one group "seems" to be better than another seems rather vague and unsubstantiated, as does the basis for being paid more.
Being paid is a consequence of skills, value, productivity, sales & marketing, profitability efficiency, alongside supply and demand. If you're not marketing your skills (or don't have skills to market), that's no one's fault but yours-- no one owes you anything, regardless of 'communication abilities'.
The original pay gap came from a trash paper which compared average pay of women and men across disciplines. So you had doctors and engineers grouped with nurses and daycare workers. That was the origin of ".75¢ on the dollar". Within most industries there is no paygap that cannot be explained by lack of negotiation and shorter work hours, among other factors. We also take it as a given that average performance of men and women on any task is identical, and that's nonsense.
The paygap myth is pure propaganda.
Edit: imagine my surprise when even CBS is admitting as much!
Why do you think differences in men and women would be limited solely to communication? You don't think millions of years of sexually dimorphic specialization might have lead to different behavioral tendencies? What about different hormone serum levels and metabolism?
This utopian Western notion of the mind being a perfect clean slate from birth has no basis. Do you really think that the mountains of evidence, in the form of gender discrepancies which have been almost universal across time and culture, can be so trivially handwaved away as "social constructs"?
What about different behaviors in other sexually dimorphic mammals? You know, like, literally all of them?
Sample size: 34, for determining "community smell" ahh yes how quantitative their metrics for that must be.
Nooo, this isn't politically biased at all ;)
They didn't SEEK this outcome of their study to conform with currently in-vogue political trends, not at all-- it's in no way biased ;)
No, no, you see-- We must get men to think, act and communicate like women! And women to think, act and communicate like men.
Yes, that's the answer-- Confuse people into thinking they're the same while demonizing men and elevating women with fake, clearly ideologically biased, nonsensical studies which use terms like "community smell".
Neomarxists/diversity-outraged: Women and men are equal.
Neomarxists/diversity-outraged: Women are better than men. We need more women in positions of power! (Search: "Entryism")
Noo... this couldn't be biased by current fad politics... not at all ;)
/sarcasm
In case anyone is wondering why Woke/SJW/identity-politics neomarxist messaging suddenly increased in the media in the 2010-2020 timeframe--
Here's an interesting summary of how usage of woke & SJW terminology dramatically increased in NYT publications since 2010 (alongside its increase in other forms of media).
I read the link. It’s one paragraph and a bunch of graphs.
I don’t dispute it’s some form of social engineering. I don’t think it has anything to do with occupy. There is a larger socialist movement going on in America. Look at Bernie. Maybe started as economic during occupy but turned to equality and acceptance for all. Gay marriage and a lot of other things were going on during this time frame.
----
For the study, the authors looked at four types of community smells:
1. Organizational Silos. Groups in the community don’t communicate with each other, except through one or two of their respective members.
2. Black Clouds. Community members are overloaded with information because communication isn’t structured well.
3. Lone Wolves. Defiant, disrespectful contributors won’t listen to others.
4. Bottlenecks. One team member wants to be the only way information moves between sub-communities.
Previous studies have shown that women are fundamental to reducing community smells and avoiding accumulating social debt for teams in other industries. But these studies show that women increase team efficiency and organizational quality on software engineering teams specifically. Teams with no women have 7 mean community smells compared to 3 for teams with at least one woman. These results are statistically significant (p-value < 0.05) and have a large effect size (d=0.68).
----