and our next speaker is joseph bellerin a quiet and reserved researcher and practitioner joseph is considered by cognizanti to be one of the pioneers and top experts on agile methods today he will talk about data not opinions the psychometrics of team and organizational dynamics josef the stage is yours happy to be back at stretch this year although it wasn't planned i've spoken here in a craft a number of times most of the times it was planned sometimes it was spontaneous and improvised but last week i got a an email from the feller asking if i could jump in on short notice to substitute for someone who had to cancel so i said of course and asked if there was any special topic that he would like me to speak about so he asked me if there was anything i was working on currently and said oh i saw this and i thought it was funny that was this quote down here from deming without data you're just another person with an opinion and then he wrote that maybe something about this topic could be interesting about how leaders could be better at being data driven on soft topics on people topics without them having to have a postgraduate degree in psychology oops so this meant i couldn't talk the way i normally do but i thought it would be a great topic for the talk and a great challenge for me i mean essentially this is what evidence-based practice is trying to be right you try to justify things with statistics to people who hate statistics when you don't really understand them yourself right as you know studies show that 90 of all people will believe anything you say as long as your first three words are studies show that so in any case this talk is really improvised it's so new that the ink on my slides is still drying so please be gentle with me in case uh i forget something if the talk might be too short the talk might be too long uh but i'll try to get it done in time okay so the current situation the one i deal with right as peter drucker once said if you can't measure something you don't know whether your actions your investments any interventions you make or improving it or not because it's easy to measure things like revenue profit roi but it's extremely difficult to measure anything that deals with people how do you know that that training that you invested in or that consultant who wanted to improve your team and or your company's psychological safety or their performance how do you know if they achieved that goal and can you quantify by how much they did it this is a challenge we have so normally what you do is an assessment all right the problem with assessments kazimski once said the map is not the territory the same goes for an assessment an assessment says more about the person who designed it than about the person who takes it if you remember the map is not the territory the map is a subjective representation of the territory that the map maker makes to emphasize what he or she thinks is important to you the same goes for assessments i remember once in a psychometrics class where our professor came in handed out a bunch of business assessments said take a look at these and tell me what these people are trying to sell this is the problem with assessments and this is where psychometrics can help okay what is psychometrics it's one of these big words that scares people psychometrics is a branch of psychology that deals with design of instruments to measure what we call a latent construct a latent construct so if i was there with you i could probably directly measure your height or your weight if i have the proper instruments what i couldn't directly measure would be your age your iq your level of depression things like that so these properties these attributes aren't directly observable and not directly measurable okay which means that any measurement is only an estimate okay so psychometrics can help us improve our estimates in two different ways it teaches us how to design questions and how to ask people the better questions and how to run better statistics on these questions remembering that people are a bit special when you ask them questions to be honest psychometrics is what separates the real psychologists from the amateurs this is how we flex our muscles i can ask better questions than you and i'll prove it with advanced statistics so let me show you one thing that i'm working on and how i'm using psychometrics to help that this is something called the team dynamics inventory okay the model behind this is about high-performing teams okay and this is based on serious quantitative research that comes from mit and from carnegie mellon what their research is they went from the hypothesis that if you get a group of really smart people together you have a really smart team so individual intelligence will help with collective intelligence and that will help team performance but if you see those numbers there don't worry too much about what they are it's just they're between zero and one and they're quite low okay but what they discovered what would really help make teams intelligent and high performance ready to be surprised psychological safety the buzzword of the year right allow me to go on a rant about this psychological safety has become an empty buzzword and it's quite interesting thinking about discussing it with a couple colleagues it seems that the only people talking about psychological safety are not psychologists and funny we psychologists have more interesting things to talk about but we don't have to prove that we're psychologists psychological safety is cool you talk about it and as one of my friends says right psychology has to deal with people we're all people so of course we're psychologists even if our degree is from google university all right but we've all heard about this psychological safety yes it is necessary it's a climate where people feel comfortable being themselves that's the official definition from the book i prefer to have another definition from a psychologist friend of mine it's a psychologically safe space it's like a dance floor where you can be yourself and you think about that going out on the dance floor if nobody's there being the first one trying to express yourself how comfortable are you especially for somebody like me who has two left feet two left legs and one is shorter than the other right so that's a challenge so as we all know psychological safety is important especially since google found that out but we maintain psychological safety is a necessary but not sufficient prerequisite for high performing teams and organizations well what else do we need what the researchers at mit discovered is we also need cognitive diversity we need people who think differently now there this is another buzzword that's come up uh strongly recently diversity and inclusion and we tend to believe that all types of diversity be it gender diversity be it ethnic diversity be religious diversity be it sexual orientation diversity all leads to one aspect is that people think differently and what we need is this diversity of thought in order to be creative to be innovative and to be able to confront the challenges that we're currently facing as patton once said if everyone's thinking alike then somebody isn't thinking the third thing we need another buzz word empathy as simon baron cohn it says empathy is the ability to identify what someone else is thinking or feeling and to respond to that person's thoughts and feelings with the appropriate emotion this is also a buzzword once again my linkedin feed this morning i saw somebody posting this article on forbes saying that empathy is the most important thing that leaders nowaday need yes but whereas things like psychological safety cognitive diversity the more the better empathy is not like that empathy is not a trait where more is better empathy is a bell curve if you have no empathy less empathy you have the more people tend to drift into autism spectrum disorder asperger's things like that or psychopathy the opposite is also true the more empathy you have the higher risk you have of getting higher hyperempathy syndrome possibly borderline disorder and burnout so remember you have to balance that so what i'm looking at is mapping teams mapping organizations in a dry dimension in a three-dimensional space of safety diversity and empathy now based on other research from reynolds and lewis they say that diversity and safety together they lead to a higher quality interaction between groups of people and they enable higher quality innovation agility responsiveness and then higher performance i maintain that if we add empathy to this that will ease the interaction because of the increased understanding between people it will ease the interaction between these individuals the teams in the organizations as well as it will ease the acceptance of novel innovative ideas because for example of the lack of jealousy now the interesting thing is that you need all three of these only having psychological safety only having any two is not sufficient let's take a look if we only have empathy and psychological safety but no diversity of thinking we're getting into user groups group things we all think the same way we all understand each other but we don't accept another group's opinions and way of thinking about things if we have empathy and diversity but no safety in speaking out what we think things become frozen ideas are there but they're never brought in and acted upon last but not least if we have psychological safety and diversity if we feel comfortable speaking and we have different ways of speaking from each other but we have no empathy for each other we end up where we are now with covet debates okay so if we go on and build on this and jim dader said any useful idea about the future should appear to be ridiculous of course the opposite isn't true and all ridiculous ideas are not going to be useful okay but any useful idea about the future should appear be ridiculous and this is where we need all of these three have you ever experienced being in a group where we're looking for a solution to a problem looking for ideas somebody brings in an innovative a creative idea and someone else says shut up that's stupid that's happened right but the lack of psychological safety will damage any of these ideas and the fear of sanction here it's not about being disciplined for say saying something that might be considered stupid but it's about social sanctions it's about the loss of esteem and respect in the group it's about being embarrassed in front of others these are all things that we as people as humans we're attuned to and designed to avoid if at all possible how much social backlash would you suffer in your team and in your organization if you contributed a ridiculous idea so how can we go about assessing these dynamics there are two different ways we can ask people or we can observe them now since this talk is very short i'm only going to go a bit into the part about asking observation running tests and games with people that's part of another talk unfortunately but if we ask what we're going to look for and this is a big challenge it's something that is quantifiable and something that is actionable right so the best approach and what i use is to combine the two of these into a mixed method multi-model approach right now let's just take a look at asking what are we asking about also what are we looking for our behaviors reynolds and lewis define certain generative and non-generative behaviors they say that high performance teams high performance organizations innovative creative organizations have behaviors you notice people being curious encouraging resourceful flexible experimental and appreciative in organizations that aren't that way you'll find behaviors of being hierarchical people being conforming cautious resistant to change a very controlling power and being very directive so what reynolds and lewis did was map safety and diversity into four quadrants i personally think this is a good idea i wish it wasn't stated as one of these two by two management matrices where lower left is bad and upper right is good but in any case they consider four different environments right depending on the amount of diversity the amount of safety and they say that the generative environment is the one where teams and organizations score high on both aspects now using this i can track the results of an intervention that i've done where i say i've taken a team that's doesn't feel psychologically safe but has cognitive diversity and starts moving them into a different space and we can actually track this we can quantify this there is data behind this that we can actually see that this intervention over time has been successful what we can also see from the same intervention before the intervention the dominant behaviors that were reported were the non-generative ones being hierarchical being controlling being cautious etc moving over after the interventions being encouraging being flexible as you see there are still some non-generative behaviors there afterwards we can't get rid of all of them nor do we want to that culture is part of what made the organization the way it is and you don't want to remove everything of it okay so what i'd like to do at the end is show you a couple case studies they're ongoing case studies i'm not done with either of these but give you a feeling for some of the work that i'm doing and how i can do with other companies company a team dynamics inventory position what i've done is i've taken reynolds and lewis's graph and expanded it to also show empathy the larger the circle more the more empathy people see the smaller the circle the less empathy that they do what we see here is that this organization feels relatively safe have a good degree of cognitive diversity but they're not quite there yet so if we look at their behaviors what we see is that it's a very encouraging environment people are curious they're appreciative they're experimental but on the other hand they're cautious now this i found these results interesting i thought i'd go a bit deeper in them and so as a result of the assessment the testing i was doing i was also testing people's uh dispositional profiles their personality traits using the neo ffi actually the neo pi r personality inventory and what i came up with here were some interesting correlations between oh sorry this is the big ugly table just to prove that i actually did the the statistics on the top here we have the 12 different uh behaviors and on the left hand side we have the five different uh personality traits and this is a point by serial correlation coefficient and i promised myself i will not say any of these big terms without explaining to you what this stuff is okay a point by serial correlation coefficient measures how good two attributes relate to each other when one of them is a continuous variable and the other one only has a specific set of values for example does a person's iq in the united states have a direct relation to whether they vote democratic or republican we have one the iq is a continuous variable in a two-party system we have dichotomous variables but don't worry about this big stuff some people use these big terms just to bluff and just to show off and if you're worried about that do what my wife always does if my wife sees a paper where i have some big words on this she'll normally write something like harry potter and the port by serial correlation coefficient don't worry about it go on so let's not go back to that table because that's too much but what we can find here some really interesting correlations what i found is that people in this organization who ranked high in agreeableness they wanted to be liked they're friendly they're helpful to other people that they saw the behavior as being very encouraging and for any of you who's ever done psychology and statistics that r value of 0.816 is extremely high and what's even more impressing is the p value of 0.004 yes got it what we also see is that people who score higher on degree openness they tend to be encouraging they also tend to be resourceful what's interesting is that people who are neurotic score high on neuroticism they find the environment to be conforming but we also have an interesting negative correlation people who are very conscientious they don't find it to be conforming interesting so what i thought i would do is start comparing how similar people were okay so what we see here is an interla inter-class correlation coefficient don't worry about that that's harry potter and the inter-class correlation coefficient that's just a number that says how much these people agree with each other more interesting is this big graph on the right where the red or something is the more two of the people agree with each other if it's very light color they're not very much in agreement if it's very dark then they're in total agreement and what we see the diagonal is that everyone's in total agreement with the themselves which is always a good thing but what i noticed here was this big block that's interesting we have a chunk here we have a cluster so i thought might there be a number of clusters can i take a look and see if there are any clicks forming here so what i use is something that's called a scree plot and the scree plot helps us figure out how many clusters there are okay and it shows eigenvalues oh that's another big word harry potter and the eigenvalues harry potter and the eigenvalues that sounds like some rock group from the 1980s sorry about that don't worry about that scree plot's going to tell me that essentially we probably have two different clicks here so if i do a factor analysis a factor analysis is trying to crunch all these different ways of looking at things down and see are there any commonalities here what we discovered is that there are essentially two different groups of people now knowing this company it's a client of mine and knowing the people there you have to trust me this makes total sense to me these are really two different groups of people uh what i did afterwards was then i went here went back if you see i changed the heat map change the order of people in the heat map to match the factor analysis and what we see is that we do have two relatively big clusters that are in agreement with each other the way they're viewing things but not in agreement with the others now that's very interesting we could use this information to say would it make sense to put these people together in the team because they work better together so company b another one of my clients now look at this one here we have people who feel very safe but standard deviation the spread over psychological safety is a lot bigger here than it is over cognitive diversity you have a number of people who are very empathic but we have that one little black dot in there of a person who doesn't really have much empathy so if we look at their behaviors we see that they're very flexible and resourceful what's missing here what i'm working on is appreciative and experimental so essentially the people here feel they're not being appreciated for their work it's okay your work's done good let's do the next thing and that's something that i've now fed back to management is there a way that you can show them your appreciation and experimental let them try out new ways of working now their correlation heat map looks like this okay also their intraclass correlation coefficient is relatively low but there are some clicks in here right we see one here and we see one here now this happens to be two separate teams uh but also what we did i ran the scree plot did the factor analysis came up with two clusters that will cover most of the variants here but also this raider number five stuck out like a sore thumb didn't fit in any place so if we take a look here what we see is that raider number five doesn't really seem to fit in with anyone else and that's quite interesting but i went on and said okay who do you think this raider number five is here raider number five is that person what does this tell me about how i should structure my teams how i should put them together in order to have them work very efficiently how can i see that when i put them together in different ways and run these tests again that i can actually quantify the change that's been done this is important if you hear anybody talking about team performance or high performance and stuff ask them for the data ask them to quantify it and ask them to prove it okay i'm almost done i have one final quiz for you two companies company a and company b which company do you think is a software company which company you think is a consulting company think about it so just finishing up once again you have a right as a leader to demand the data demand the proof demand quantifiable evidence all right i'll leave you with this quote from deming in god we trust all others bring data okay so i just want to point out there is also a lot of research done a lot of support you know real serious scientific research here are some papers on it some more papers some more papers even more papers more more more more here's a couple more and even some more without that there's a lot of work that's been done on this really seriously okay if you want to find out more about this about the work that i do and how i do it with uh organizations and teams feel free to look me up on linkedin if you look me up on linkedin please just write me a little note saying hey i heard your talk i'd like to connect with you or go to my website and you can connect with me then there okay thanks a lot i hope you enjoyed it thank you so much i certainly enjoyed it and thank you for showing us all your references um i'm sure a lot of us can start digging in the topic um i can see that our audience is still pondering some questions so i would like to first start by asking which metric was the consulting company consulting company was uh was company a okay and the software company was company b so the uh consulting company had the narrower bandwidth narrower standard deviation of psychological safety but the larger bandwidth of diversity the software company had the higher bandwidth of psychological safety there were people who felt safe there were people who didn't feel so safe all right all right thank you very much as our audience is uh preparing up here uh is a question already from anna can you share with us some examples on what do you do in these interventions that you've mentioned previously um once again right the problem with assessments is that pardon me they are normally designed by people or companies who have something to sell an agile assessment or a scrum assessment is sold by a company that's want to sell their agile consulting or their scrum consulting or things like that uh i consider myself as a researcher to be a doctor if you go to your doctor right you you know the quote about maslow's hammer if all you have is a hammer every problem is a nail there are a lot of interventions that you can do here and what i'm actually looking for and what i find to be quite useful here is that this theory that i have tends to be more inclusive if we didn't even get to talk about the other analysis that that we do with the observations and stuff but we can do that at a very micro level um but yeah you can do work on increasing psychological safety you can do work on increasing diversity actually company a found out that their way of increasing diversity would be to start looking for more diverse clients looking for more diverse clients would indirectly cause bring them to attract more diverse workforce or also training in empathy and emotional intelligence there are a lot of interventions that that are possible there and i am not selling one particular intervention and when i work with clients if i find out that there's an intervention that's necessary that i don't do i will bring in other specialists just the way your doctor does this is the only way of being a professional is to know the bounds of your professional competency thank you thank you um here is on the next question that or so has gotten some upwards from chaba how could we get a sample of those magic questions you use for gathering the data okay this is my research this is my proprietary ip right if you are a psychologist and you've studied this stuff if you study cognitive psychology and psychometrics you should be able to figure this out for yourself um but part of this is part of my consulting offering yeah so get in touch with josef and you will get the sample uh another another suggestion if you're really interested in this i'm doing a half day workshop at the end of january with the psychologist colleague uh at ut eckstein's conference the oop in munich where we're going to go into detail and show maybe some of the questions we're going to show the exercises we do for the other part of the assessment and you can learn how to do at least a good bit of it there thank you thank you what an awesome opportunity thank you for for letting us know that and by the way uh i have to convince you i might also be able to do it at craft this year if it's going to be live all right um all right we are looking forward to that i i hope that a lot of things are going to be live this coming year joseph thank you so much for joining us at stretch2021 i am showing you our gift of appreciation thank you for being one of our speakers we will be sending you this after the event has taken place um and um as you said i hope uh some of our audience will be able to get in touch with you via linkedin thank you so much for coming okay thanks a lot after i get out of zoom i'm going to try to log into hopkin if anybody wants to chat there thank you okay thank you have a nice day bye you