<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Margaret Tomaszczuk on Medium]]></title>
        <description><![CDATA[Stories by Margaret Tomaszczuk on Medium]]></description>
        <link>https://medium.com/@margaret_12746?source=rss-9293263e7ee0------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*VL44riiLJGyckNbu0fN6lA.jpeg</url>
            <title>Stories by Margaret Tomaszczuk on Medium</title>
            <link>https://medium.com/@margaret_12746?source=rss-9293263e7ee0------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 01:10:43 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@margaret_12746/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Human-Centered A.I. is the Future of Talent Management]]></title>
            <link>https://medium.com/cultivate-ai/human-centered-a-i-is-the-future-of-talent-management-bade900841e4?source=rss-9293263e7ee0------2</link>
            <guid isPermaLink="false">https://medium.com/p/bade900841e4</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[hrtech]]></category>
            <category><![CDATA[talent-management]]></category>
            <category><![CDATA[hr]]></category>
            <dc:creator><![CDATA[Margaret Tomaszczuk]]></dc:creator>
            <pubDate>Thu, 14 Jun 2018 21:48:40 GMT</pubDate>
            <atom:updated>2018-07-03T19:58:25.466Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*Cst6y-9aTrVZsFG97Ilouw.jpeg" /></figure><p>Will A.I. eliminate my job?</p><p>It’s a clickbait title most of us are now familiar with.</p><p>In recent years we’ve been met with a wave of articles and soundbites — ranging from the realistic to apocalyptic — speculating as to whether A.I. will replace human jobs, take over the world, or otherwise render Us insignificant.</p><p>Tesla CEO Elon Musk has even gone so far as to suggest that the volume of jobs that will be lost due to automation will <a href="https://www.cnbc.com/2016/11/04/elon-musk-robots-will-take-your-jobs-government-will-have-to-pay-your-wage.html">create the need for a universal basic income</a>.</p><p>A fear of new technology, and of the impact that that technology will have upon the job market is not new.</p><p>Technological developments that arose during the Industrial Revolution created public fear of mass unemployment (a fear that ultimately <a href="http://theconversation.com/what-the-industrial-revolution-really-tells-us-about-the-future-of-automation-and-work-82051">proved to be unfounded</a> given the large number of new jobs these technologies created).</p><p>Yet the narratives have never felt quite so existential before this moment.</p><p>So what is different about A.I. that has so captured the public interest, and it seems, fear?</p><p>It seems to lie in the idea that intelligent machines will not seek to supplement aspects of our existence, but rather, replace us entirely.</p><p>Computer Scientist Subhash Kak advocates for this idea with respect to the job market in his <a href="https://www.nbcnews.com/think/opinion/will-robots-take-your-job-humans-ignore-coming-ai-revolution-ncna845366">think piece for NBC News</a> (a piece, it is worth noting, entitled “Will robots take your job?”). The reason A.I presents a greater threat to society as we know it, he argues, is “today’s A.I. technology aims to <em>replace</em> the human mind,” not simply to make industries more efficient (my emphasis).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/724/1*RM3s_0Ad1Vsq_MIV8TyjzQ.jpeg" /></figure><p>It would be naive to ignore the reality of Kak’s argument with respect to tasks requiring learning and judgement. A.I. is already replacing human decision-making in industries such as transportation and manufacturing.</p><p>But are all applications of A.I. really aiming to <em>replace</em> the human mind in the workplace? And should they?</p><p>There are other views — and other technological frameworks — to be had here.</p><h3><strong>“Human-Centered A.I.”</strong></h3><p>In opposition to A.I.’s “takeover” rhetoric exists a school of thought that explicitly acknowledges the benefit of <em>partnership</em> between humans and intelligent machines.</p><p>Fei-Fei Li, director of the Stanford Artificial Intelligence Lab, <a href="https://mobile.nytimes.com/2018/03/07/opinion/artificial-intelligence-human.html">calls this approach “human-centered A.I.”</a> — a framework for guiding the development of intelligent machines by human concerns.</p><p>At a high level, the goals of human-centered A.I. are as follows:</p><ol><li>A.I. should aim to enhance human thought rather than replace it</li><li>A.I. should encompass the more nuanced and contextual aspects of human intellect, aided by outside fields such as psychology and sociology</li><li>The development of A.I. technology should be guided by a concern for its effect on humans</li></ol><p>There are a number of cross-industry applications of A.I. that can be viewed within this partnership framework.</p><p>Take, for example, the <a href="https://www.techemergence.com/machine-learning-in-surgical-robotics-4-applications/">development of robots</a> used to reduce costs, time, and human-error during surgery, allowing doctors to focus on the more nuanced aspects of the surgical process. Or, <a href="https://www.cbsnews.com/news/artificial-intelligence-could-revolutionize-farming-industry/">developments of A.I. in agriculture</a>, such as Blue River Technology’s “see and spray” technique for applying herbicide only where needed, saving farmers money on herbicide and delivering a more sustainable product to consumers.</p><p>But perhaps even more in contrast to the fear of a robot taking one’s job, is the increasing extent to which A.I. is being applied the field of talent management.</p><p>That is to say, A.I. is being used to actually <em>improve</em> the workplace and the worker experience, rather than replace the worker.</p><h3><strong>A.I. as a Tool for Improving the Workplace</strong></h3><p>In the past several years, we have seen an emergence of companies applying A.I. to problems in talent management. From <a href="https://paradox.ai/">Paradox.AI’s Olivia</a>, to <a href="https://beamery.com/">Beamery</a> and <a href="https://textio.com/">Textio</a>, its fair to say that A.I. is on HR’s radar in a way that it wasn’t 5 years ago.</p><p>What’s interesting about this trend is that unlike other industries with a stronghold in A.I., talent management has until recently been viewed almost exclusively as a “fuzzier” aspect of the business. It is an industry built on relationships, human connections, and emotional intelligence, and yet, it is being improved with A.I.</p><p>To be fair, up until now a majority of A.I. solutions for talent management have focused on the more tedious and error-prone tasks around candidate sourcing and evaluation (tedious + error-prone = a perfect opportunity for automation).</p><p>But there are also opportunities for A.I. to improve the post-hire aspects of the employee experience, and human-centric A.I. is the key.</p><p>As the marketing world has known for years, A.I. provides a unique opportunity for scaling a personalized experience. Why would you show me the same thing as everyone else, when I’m more likely to convert if you show me exactly what I want?</p><p>The same principles can be applied to the post-hire employee experience.</p><p>Employees have different skills sets and motivators. If my employer places me in an environment that is optimized for my skills and motivators, I’ll stay. If not, I’ll move on.</p><p>As the progression towards a digital workplace continues, companies also have more data about their human capital than ever before — who they are talking to, what they eat, when they’re online every day. WeWork is <a href="https://www.wired.com/story/this-is-why-wework-thinks-its-worth-20-billion/">basing their business model</a> around this data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/665/1*epPZ7FNhhYfXAqiMIiwXGw.jpeg" /></figure><p>Human-centered A.I. can unleash this data to help talent leaders create a more personalized employee experience. It is in “fuzzier” domains like talent management where human-centered A.I. shines, not just for ethical reasons, but because it provides the best user experience.</p><p>At <a href="https://cultivateai.com/">Cultivate</a>, for example, we apply human-centered A.I. to personalize the leadership development experience for managers. Using digital communication data as a proxy for leadership behavior, we analyze and predict how managers’ actions are affecting their team, and offer suggestions for how to improve.</p><p>At no point do we attempt to stand in as a replacement for a manager, or a talent leader. Rather, like a real-life leadership coach, Cultivate offers tips and suggestions that a manager can choose to take, or not.</p><p>This is the kind of personal experience employees expect from their talent leaders, scaled with A.I. And it doesn’t need to stop at learning and development. A.I. also has high-potential to impact other aspects of the employee experience, from interviewing and on-boarding to performance reviews and off-boarding.</p><h3>Looking Forward</h3><p>There is no doubt that A.I. is changing the world — and the job market — as we know it.</p><p>Industries will be disrupted. Jobs will be lost, <a href="https://economictimes.indiatimes.com/jobs/by-2020-artificial-intelligence-will-create-more-jobs-than-it-eliminates-gartner/articleshow/62053363.cms?utm_source=contentofinterest&amp;utm_medium=text&amp;utm_campaign=cppst">new jobs will be created</a>, some jobs will never be replaced.</p><p>Ethical dilemmas will be raised. They already are.</p><p>The degree of difference between aspects of human intellect and intelligent machines will become smaller.</p><p>However, with careful consideration for A.I. design that creates a sense of partnership between humans and intelligent machines, A.I. isn’t a force to be feared in the workplace, but embraced.</p><h3>About Cultivate</h3><p>Cultivate helps companies leverage their digital communication data with A.I. to extract important organizational learning and unleash leadership potential.</p><p>For more information on what we are doing at Cultivate, check out <a href="https://cultivateai.com/">our website</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dnllJfc-5N9iq4fBXSl5sw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bade900841e4" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cultivate-ai/human-centered-a-i-is-the-future-of-talent-management-bade900841e4">Human-Centered A.I. is the Future of Talent Management</a> was originally published in <a href="https://medium.com/cultivate-ai">Cultivate AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Upcoming D&I Trends in People Operations Technology]]></title>
            <link>https://medium.com/cultivate-ai/upcoming-d-i-trends-in-people-operations-technology-e39561f47fa0?source=rss-9293263e7ee0------2</link>
            <guid isPermaLink="false">https://medium.com/p/e39561f47fa0</guid>
            <category><![CDATA[hr-technology]]></category>
            <category><![CDATA[people-management]]></category>
            <category><![CDATA[people-operations]]></category>
            <category><![CDATA[diversity]]></category>
            <category><![CDATA[inclusion]]></category>
            <dc:creator><![CDATA[Margaret Tomaszczuk]]></dc:creator>
            <pubDate>Thu, 14 Dec 2017 17:58:56 GMT</pubDate>
            <atom:updated>2017-12-14T17:58:56.054Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ybrcPK_ponHcv6GmiSx81w.jpeg" /></figure><p>Last week, Cultivate had the opportunity to participate in<a href="http://www.kaporcapital.com/2017-kapor-capital-pot-competition/"> Kapor Capital’s 2017 People Operations Tech (POT) pitch competition</a> alongside 9 other pre- and post-seed companies. Companies selected to participate offer innovative solutions in the people operations space while promoting more inclusive and equitable workplaces with their businesses.</p><p>The competition provided an interesting opportunity to visualize upcoming trends in HR technology, geared towards one of the most highly talked about issues in the people operations space at this time: diversity and inclusion in the workplace. The companies that presented are tackling people operations (and D&amp;I) issues from all stages of the employee lifecycle — from candidate sourcing, to employee engagement and retention, and everything in between.</p><p>From that pool of companies, alongside active conversations in the diversity and inclusion space, we see two emerging trends:</p><ol><li>Innovative and inclusive solutions to candidate sourcing</li><li>Automation for productivity and inclusion</li></ol><h3><strong>Innovative Candidate Sourcing</strong></h3><p>With the job market the<a href="https://www.nytimes.com/2017/12/08/business/economy/jobs-report.html?_r=0"> strongest it’s been since the year 2000</a>, it is not surprising that employee recruitment is a top priority for HR departments going into 2018. <a href="https://www.bna.com/hr-department-benchmarks-p73014461273/">Bloomberg’s analysis of HR departments in 2017</a> showed that 84 percent of the surveyed departments identified recruitment and retention of qualified employees as an important priority.</p><p>A competitive recruiting market has implications for diversity and inclusion as well. To maintain and promote a diverse and inclusive workplace culture at a time of high competition for employees, employers must <em>actively</em> seek to include diverse candidates in their hiring process.</p><p>Proponents of diversity and inclusion in the workplace have long run up against the so-called<a href="http://fortune.com/2016/10/19/linkedins-new-diversity-report-pipeline-myth/"> pipeline myth</a> — a claim that hiring diverse talent isn’t possible because there isn’t the necessary talent coming up the ranks. This myth is even more pervasive in a time of high competition for employees.</p><p>Yet there are a record<a href="https://www.npr.org/2017/08/31/547646709/u-s-employers-struggle-to-match-workers-with-open-jobs"> 6 million</a> unfilled jobs in the United States as of 2017, and employers are struggling to connect with qualified candidates. Experts cite several reasons for this disconnect, including poor employer hiring policies such as unnecessarily limiting the candidate pool to those with 4 year degrees, and a hesitation to train entry-level employees.</p><p>Working to improve these shortcomings might well align with improving diversity and inclusion in companies’ hiring practices. A recent<a href="https://www.forbes.com/sites/maynardwebb/2017/10/29/how-to-alter-your-hiring-practices-to-increase-diversity/#5c949d732029"> Forbes article</a>, for example, recommends eliminating educational requirements that might create bias and seeking talent via unconventional channels as ways to attract diverse candidates.</p><p>Several candidate sourcing solutions have entered the HR tech market in the past several years, and there were companies in Kapor’s POT pitch competition tackling the same problem with a D&amp;I angle.</p><p>The team at <a href="https://www.harmoniapp.com/">Harmoni</a>, for example, is seeking to provide refugees access to IT employment (and employer access to qualified refugees) via their microwork platform. Using the Harmoni app, refugees can get paid for completing microwork technology tasks, like Image Tagging. In addition, the most qualified refugees can be vetted for acceptance into a coding bootcamp that will help them find a full-time role as a software engineer.</p><p>The team at <a href="https://include.io/">include.io</a> is tackling candidate sourcing from another employee touchpoint: direct referrals. By providing underrepresented technologists access to mini-mentoring session with industry professionals, the platform allows new candidates to grow and learn through mentorship. In addition, underrepresented candidates can receive referrals from these brief mentoring sessions, a recruiting tactic heavily relied upon by companies but also one of the main sources of bias in candidate hiring.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*b7IlXHbTRrKnlymhJky-FA.jpeg" /></figure><p>In each case, these platforms are tapping the job market for groups typically invisible to companies due to their lack of connection to the current workforce or non-traditional educational background. An interesting bonus is their offloading some of the educational burden on companies who would hire such employees: Harmoni by giving access to coding bootcamps, and include.io by giving candidates access to mentorship by industry professionals they would not otherwise have access to.</p><p>Thus, they propose a two-fold solution for companies experiencing a candidate pipeline problem: surfacing qualified candidates while helping companies to adjust their non-inclusive hiring practices.</p><h3><strong>Automation for Productivity and Inclusion on Diverse Teams</strong></h3><p>In his analysis of <a href="https://www.forbes.com/sites/joshbersin/2017/11/02/hr-technology-for-2018-ten-disruptions-ahead/#37a9501743b7">HR technology disruptions to-come in 2018</a>, HR thought-leader <a href="https://medium.com/@josh_bersin">Josh Bersin</a> predicts “a massive shift from automation to productivity.” He means that automation, as it relates to payroll, record-keeping, assessment, interview and hiring, etc. is now commonplace in HR departments (though the market for these products continues to grow). The next challenge will be to build software that targets employee engagement, improves productivity, and helps teams work better together.</p><p>From a workplace diversity and inclusion perspective, Bersin’s prediction of a trend towards software targeting team dynamics actually somewhat aligns with what D&amp;I practitioners have been emphasizing in recent years. Namely, the idea that inclusion is a concept distinct from, and crucial to diversity.</p><p>Though often conflated in the past, diversity and inclusion mean different things, and fostering each can target different parts of an employee lifecycle.</p><p><strong>Diversity</strong> = employees of different backgrounds and identities are represented in a workplace</p><p><strong>Inclusion</strong> = employees have a sense of belonging, and feel respected and valued for who they are as an individual or group</p><p>While diversity can certainly be fostered via candidate sourcing and vetting (using the solutions described above, for example), inclusion is something that must be embedded in company culture before, during, and after the hiring process is complete in order to be executed effectively.</p><p>It is easy to track movement on diversity initiatives — to know if I’m moving the needle on gender diversity I would merely need to track my number of female, male, and gender-not-identifying employees quarter over quarter. However, as noted by talent innovation experts <a href="https://hbr.org/search?term=laura+sherbin">Laura Sherbin</a> and <a href="https://hbr.org/search?term=ripa+rashid">Ripa Rashid</a>,<a href="https://hbr.org/2017/02/diversity-doesnt-stick-without-inclusion"> quantifying feelings of inclusion can be dicey</a>. Here is where upcoming automation solutions can help.</p><p>Harnessing automation for shift-scheduling, POT pitch competition second place winner <a href="http://adayjobs.com/">Aday Technologies</a> is positioning themselves as a position-based labor market for brick and mortar stores. Using the app, employees can pick shifts at brick and mortar businesses according to their schedule, freeing them from the scheduling limitations of shift-work that will negatively affect individuals such as mothers and those working multiple jobs.</p><p>Here at <a href="https://cultivateai.com/">Cultivate</a>, we are building an AI-assistant that provides managers with real time awareness on their own communication bias and their team’s engagement. By connecting with your workplace written communication platforms like Gmail and Slack, we can provide managers with analytics about how they are communicating with their direct reports, as well as actionable insights for how to improve their relationships to foster an inclusive workplace.</p><p>The team at <a href="http://www.welovework.com/">We Love Work</a> is tackling inclusion by quantifying the sticky subject of company culture. By having candidates take a company culture assessment before joining a team, the platform helps companies to avoid the high cost of a bad hire and address the concept of “culture fit” with metrics rather than within the bias-prone interview context.</p><p>Harnessing automation, each of the companies above solves a demonstrated desire to use HR technology to improve team dynamics and efficiency, while fostering diversity and inclusion goals at the same time. Companies like Cultivate and We Love Work are moving towards quantifying company culture and inclusion for the first time. Aday Technologies demonstrates how automation can be used to make an entrenched industry practice, like shift scheduling, more efficient while alleviating the lifestyle demands of that practice for individuals who need it.</p><p>Employees and companies win when teams invest in inclusion-automation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6xF5lCHinkKWogg396DEoQ.jpeg" /></figure><h3><strong>Looking Forward</strong></h3><p>The advancement of diversity and inclusion minded offerings in the HR technology space is exciting both for HR teams looking to improve their candidate and employee experience, and diversity and inclusion practitioners looking to move the needle on diverse and inclusive workspaces. Even more exciting is the idea that one may invest in a single technology to solve both an existing HR need and foster D&amp;I at the same time.</p><p>Although the business value of diversity and inclusion has become clear, having impact on <a href="https://dupress.deloitte.com/dup-us-en/focus/human-capital-trends/2017/diversity-and-inclusion-at-the-workplace.html">brand, corporate purpose, employee performance, and recruiting</a>, budgets allocated to diversity and inclusion programs are still significantly lower than those going towards other HR initiatives such as recruiting. Simply put, in order to ensure that D&amp;I initiatives receive a fair share of time and resources, HR departments and recruiting teams must do<a href="https://www.shrm.org/resourcesandtools/hr-topics/talent-acquisition/pages/recruiters-hire-more-same-budgets-2017.aspx"> more with less</a>.</p><p>Fortunately, teams like those listed above, and others who participated in Kapor Capital’s 2017 People Operations Tech pitch competition, are building innovative solutions to help companies do just that.</p><h3>About Cultivate</h3><p>Cultivate’s AI-powered platform enables engagement and inclusion by helping you understand and improve your workplace communications.</p><p>For more information on what we are doing at <a href="http://cultivateai.com/">Cultivate</a>, check out <a href="https://www.youtube.com/watch?v=nMaw1bAW14w">our interview at TechCrunch Disrupt!</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dnllJfc-5N9iq4fBXSl5sw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e39561f47fa0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cultivate-ai/upcoming-d-i-trends-in-people-operations-technology-e39561f47fa0">Upcoming D&amp;I Trends in People Operations Technology</a> was originally published in <a href="https://medium.com/cultivate-ai">Cultivate AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Yes, a statistical model can detect workplace communication bias (and why it’s ok if that makes you…]]></title>
            <link>https://medium.com/cultivate-ai/yes-a-statistical-model-can-detect-workplace-communication-bias-and-why-its-ok-if-that-makes-you-db56ee25cc99?source=rss-9293263e7ee0------2</link>
            <guid isPermaLink="false">https://medium.com/p/db56ee25cc99</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[diversity-and-inclusion]]></category>
            <category><![CDATA[people-analytics]]></category>
            <category><![CDATA[naturallanguageprocessing]]></category>
            <dc:creator><![CDATA[Margaret Tomaszczuk]]></dc:creator>
            <pubDate>Mon, 06 Nov 2017 20:36:15 GMT</pubDate>
            <atom:updated>2017-11-06T22:57:11.375Z</atom:updated>
            <content:encoded><![CDATA[<h3>Yes, a statistical model can detect workplace communication bias (and why it’s ok if that makes you uncomfortable)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*vsgHvDOz-tGA-m6PtOSMew.jpeg" /></figure><p>What if we told you that an AI could help you to better understand how you’re communicating at work?</p><p>What if we told you it could do so by reading your written workplace communication?</p><p>What if we told you that it could even provide you quantitative metrics about your communication biases and the engagement levels of your team members?</p><p>At this point in our conversation, you might start to ask questions:</p><blockquote>“But, how do you measure bias?”</blockquote><blockquote>“How do you account for different personalities when you’re measuring engagement?”</blockquote><blockquote>“Does an algorithm <em>really</em> understand my team dynamics?”</blockquote><p>In fact, we hear questions like this all the time about our work here at Cultivate.</p><p>In this post we want to address two underlying themes that we perceive in questions like these: (1) the doubt that workplace communication can be accurately analyzed and quantified with computation, and (2) the discomfort in quantifying something as subjective as a human relationship.</p><h3>Recognizing bias and engagement signals in the workplace</h3><p>As we will be discussing the analysis of workplace bias and engagement in this post, let’s first loosely define these two terms as they pertain to our research:</p><ul><li>Bias = An individual’s variation in communicative behavior towards their colleagues</li><li>Engagement = An individual’s level of involvement in, enthusiasm towards, and commitment to their work and workplace</li></ul><p>Before we discuss how a computer can score someone’s engagement or bias levels, let’s talk about how you might go about recognizing engaged or biased behaviors in your own workplace.</p><p>Recognizing your own bias is tricky, since a lot of these behaviors happen unconsciously (a difficulty unconscious bias trainings, like <a href="https://rework.withgoogle.com/guides/unbiasing-raise-awareness/steps/watch-unconscious-bias-at-work/">Google’s reWork program</a>, are intended to alleviate). Thinking about engagement from one’s own perspective is a little easier. If you were asked to score the engagement level of one of your colleagues, you’d probably think about the many instances in which you interacted with this individual to come up with an answer.</p><p>You might think about her mood, her performance, or the amount of time she’s been spending at the office recently. You might also think about how this individual’s current behavior compares to that of her colleagues, and herself throughout your working relationship.</p><p>From the outset, it is clear that bias and engagement levels are identifiable via observed (or not always observed, in the case of bias) behavior over time. Believe it or not, a similarly holistic analysis can also be achieved via computation.</p><p>By decomposing the high-level concepts of engagement and bias down into algorithmically defined sub-parts, a model can learn to not only recognize the signals of engagement and bias, but also quantify them.</p><p>It is in defining these sub-parts where the quantitative and the subjective blur. One must decide which calculable metrics are actually indicators of workplace engagement or bias.</p><p>For exemplary purposes, these are the extractable metrics we have associated with bias and engagement, respectively, at Cultivate:</p><p>Metrics related to engagement:</p><ul><li>Mood — Positivity, politeness, and job satisfaction</li><li>Attention — Speed and thoroughness of response in both 1-on-1 and group conversations</li><li>Productivity — Work initiative and output</li><li>Motivation — Job commitment and effort level</li><li>Teamwork — Willingness to collaborate in a group setting</li></ul><p>Metrics related to bias:</p><ul><li>Tone — Politeness and sentiment</li><li>Attention- Initiative and responsivity</li><li>Style — Similarity of vocabulary and sentence structure</li></ul><p>Algorithms can actually be built around each of these metrics, drawing from a combination of well-established Natural Language Processing techniques and cutting-edge sociolinguistic research. We are going to cover some of these algorithms here and describe example use-cases, but before we proceed, let’s review a brief definition of Natural Language Processing (NLP) and the related process Sentiment Analysis.</p><h3>Natural Language Processing for Dummies</h3><p>Wikipedia defines Natural Language Processing as “<a href="https://en.wikipedia.org/wiki/Natural_language_processing">a field of computer science, artificial intelligence and computational linguistics concerned with the interactions between computers and human (natural) languages</a>.” This is a pretty broad definition, and indeed, NLP techniques are used to conduct language-related computational research across a wide range of disciplines (see Stanford Natural Language Processing Group’s <a href="https://nlp.stanford.edu/pubs/">publication list</a> for some current examples).</p><p>An easy way to start thinking about the types of problems NLP can solve, and relevant to the computation of bias and engagement metrics such as mood, tone, and motivation, is a process called Sentiment Analysis.</p><p>Sentiment analysis is used to define the attitude of a string of text. For example, a very simple sentiment analysis algorithm might be responsible for reading a sentence and outputting whether the sentence is positive, negative, or neutral.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*bV7XSovWAMR6fZZ2-5w7Nw@2x.png" /></figure><p>If presented with the sentence: “I had a good day today,” how might our algorithm correctly identify the sentence as positive?</p><p>Probably the easiest way would be extracting the word “good,” which has positive connotations in the English language.</p><p>Let’s go one step further. How might the same algorithm handle the sentence: “I did not have a good day today”?</p><p>In this case, we could not only rely upon the positivity of the word “good” to define sentiment. The algorithm would also need to handle negation of the positive word “good” with the word “not” in order to correctly identify the sentence as negative.</p><p>Building upon this problem, we could continue to add more complex logic to our algorithm, for example, logic to quantify a sentiment’s direction (ie. “I don’t<strong> </strong>like the color <strong>red</strong>”) or intensity ( ie. “I had a <strong>really </strong>good day today”).</p><h3>Natural Language Processing in Practice (in the Workplace)</h3><p>From the example above, it is fairly evident how Sentiment Analysis can be used to help calculate metrics related to positivity and negativity, such as mood, tone, and motivation.</p><p>What might be less obvious is that NLP models can also be trained to recognize and measure other affective signals.</p><p>For example, you can actually train a model to score the politeness of a string of text, along with its directionality and intensity as you can in Sentiment Analysis. The model will do so by identifying signals of politeness (cultural norms will obviously come into play here), such as saying “please” or asking someone for a favor rather than telling them to perform a task.</p><p>And NLP methods go far beyond affective signal recognition. Here are some additional methods that can be used (in part) to calculate the bias and engagement metrics listed above:</p><ul><li>Style Detection — Used to analyze the similarity of text by extracting and analyzing vocabulary and structure</li><li>Dialog Act Classifiers — Used to assign a “high-level categorization of pragmatic meaning” to spoken or written language. For example, an utterance might be defined as a “Request for Information,” or “Request for Action” (see <a href="http://www.aclweb.org/anthology/N13-1099">Omuya et al.</a>).</li></ul><p>Critical to the use of these NLP processes to reliably identify signals of bias and engagement is their ability to extract reproducible, quantifiable patterns of behavior. It is even more important that these patterns be distinct enough to allow accurate prediction of social context from text alone.</p><p>Cultivate Chief Data Scientist Andy Horng put the replicability and predictive ability of the models we built to the test in his recent <a href="https://cultivateai.com/resources/Enron-Case-Study.pdf">white paper</a>. The white paper details a case study of the gender-identified Enron email corpus, analyzing variations in language use conditioned by social dynamics. The goal of the study was to reproduce findings from 3 sociolinguistic research papers using Cultivate’s existing conversational analysis pipeline.</p><p>The results? Cultivate’s models were able to reaffirm the results of each of the research studies.</p><p>So, yes, a statistical model can, in fact, quantify bias and engagement in the workplace. By identifying metrics associated with bias and engagement such as mood and politeness, and quantifying them using established NLP techniques, we can ascribe quantitative metrics to traditionally qualitative aspects of a workplace relationship.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UL4eVmNQqGm3G_3U-ofDjg.jpeg" /></figure><h3>Accepting the Qualitative with the Quantitative</h3><p>Convinced though you may be by the techniques and scholarly support for the explanation above, many individuals will still be uncomfortable with quantifying someone’s engagement or communication biases.</p><p>And that’s a reasonable discomfort; quantification is so often associated with the notion of scoring or grading. Slapping a score on something as subjective (and indictable) as workplace bias or engagement gets tricky fast.</p><p>But it is possible to frame quantitative information about bias and engagement in a way that provokes conversation and introspection rather than imposing a stark judgement of quality.</p><p>One way to go about this is to use relational visuals rather than ascribing numbers or simple yes/no judgements to these metrics. Rather than informing your user that they are “polite” or “not polite” based on arbitrary cutoffs you have set, you can instead make it clear to them that they are <strong>more</strong> polite to certain individuals than to others.</p><p>Another approach is to acknowledge that every individual is different and will exhibit different “normal” behavioral signals. You can do so quantitatively by baselining behavioral signals not only on a group level, but also on an individual level.</p><p>For example, one of your team members might generally communicate less positively than other members of your team. If you baseline this person’s mood metrics against the average of your team, you’ll probably see low mood signals for them.</p><p>This might be interesting information about the team, but it won’t actually tell you much about your colleague’s current mood. Baselining their mood against their own average will be much more informative in this respect.</p><p>But presenting metrics in a reflexive way won’t satisfy everyone, or scrub the data of all edge cases. There will be situations where someone disagrees, or is uncomfortable with a metric that is calculated based on their communication.</p><p>However, we are of the mindset that these limitations do not undermine the value of quantitatively analyzing your workplace communication.</p><p>Why? Even if they are at times debatable or discomfiting, these metrics still provoke conversation and behavioral awareness.</p><p>Maybe you’re a manager, and you receive metrics about your email and chat communication indicating that you are least responsive, and least positive to the female employees on your team.</p><p>Might there be a valid, situational reason for that? Of course, and that is for you to decide.</p><p>Will these metrics make you think more actively about how you’re communicating, and to whom, regardless? We think so.</p><h3>About Cultivate</h3><p>Cultivate’s AI-powered platform enables engagement and inclusion by helping you understand and improve your workplace communications.</p><p>For more information on what we are doing at <a href="http://cultivateai.com/">Cultivate</a>, check out <a href="https://www.youtube.com/watch?v=nMaw1bAW14w">our interview at TechCrunch Disrupt!</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dnllJfc-5N9iq4fBXSl5sw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=db56ee25cc99" width="1" height="1" alt=""><hr><p><a href="https://medium.com/cultivate-ai/yes-a-statistical-model-can-detect-workplace-communication-bias-and-why-its-ok-if-that-makes-you-db56ee25cc99">Yes, a statistical model can detect workplace communication bias (and why it’s ok if that makes you…</a> was originally published in <a href="https://medium.com/cultivate-ai">Cultivate AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>