 starting back even before the first meeting in Paris of 2017 so today we're going to talk to you about I'm going to present some background about how the project was done and then Susan will present some of the results so this development of the survey was guided by research questions that we agreed to at the Paris meeting in 2017 so the goal of the survey and all the questions that were on the survey had to some way support these goals or help us to answer these questions was to understand the development of interest in science so that's the early years understand people's experiences in education and careers and whether there is a gender gap there look at the work-life balance for scientists how do people who work as scientists balance it with the rest of their lives family support like someone we were just discussing earlier and when I was talking here with my colleagues that many scientists have support from their families friends mentors things like that we also looked at the demographics of the scientific workforce and then one of the things we really wanted to look at was access to resources needed to conduct science and then opportunities to contribute to the scientific enterprise because our thinking was that these are areas where we might find inequality so everyone in the survey is already a scientist and what we wanted to know is for people who are scientists is there equality in these things because if you don't have access to the resources that you need then it's difficult to compete or to get your findings published and if you don't have those opportunities then your career may not advance as quickly so those were the research questions that we agreed to in Paris we also wanted to look at whether or not experiences are different in different geographical regions and Susan will be showing you a map of the regions that we looked at by discipline so the disciplines that were covered by the partner unions experiences that scientists have with different levels of human development so we use the UN human development index and Susan's going to present some results on that and then also experiences people have in different employment sectors to work on the questionnaire we started with an initial draft that was based on the 2009 IUPAP that's the physics global survey of physicists IUPAP has done three global surveys of physicists and we the first two were only of women and the third one included men so we'd have a comparison group to compare the women to the men to see if there was a gap and so we started with that survey because it was important to IUPAP to have some longitudinal results so we could see if things have changed we started with that draft of the questionnaire but then we had three regional meetings these were in different regions of the world that we wanted to make sure were included in the development of the questionnaire because we wanted to make sure that the questionnaire worked for all scientists in all types of countries and so this was a great opportunity for us to get input on the questionnaire from people who we ordinarily might not be able to talk to to make sure that the questions worked for them so there was an initial meeting there were meetings in Colombia where people gathered from Latin America Taiwan hosted by Mahoon and hosted by Mahoon in to first scientists from Asia and then hosted in South Africa to force scientists from Africa at the workshops we asked people to review the specific questions on the draft to make sure that they were going to work for all the regions represented that we didn't want it to be Western biased or Europe based or North American based everyone looked at the full survey instrument we call questionnaires instruments because it's what we use to measure things I'm a sociologist by the way and we wanted to again make sure that the questions worked for everyone and we also talked to the delegates at those regional workshops about just distributing the questionnaire so we get out to the people in less developed areas of the world and so some of you attended the regional workshops and I remember many of you from the two that I went to once the final questionnaire was approved by the executive committee of the project we used a professional translation service in the U.S. that's used to translating questionnaires into languages specifically questionnaires we wanted to ensure comparability across languages in the translation services use you know standard language things so that there's not idioms in there that are specific to different countries and they can be neutral about cultural differences and then to make sure that the translation and had the proper translation for science type things we had people on the project review the final translation to make sure that they were okay and their names are listed here Marie Francoise for French Sylvina for Spanish Seiko for Japanese I don't think she's here I looked through the program and I didn't see her name okay but she was very helpful with the Japanese translation and Mina and Shahara Zad who are both here helped us with Arabic and then we had staff members at AIP that helped with Russian and simplified Chinese the survey launched on May 1st of 2018 and was open until the end of the year and you can see it was available in seven languages English French I don't know which is Chinese and Japanese Russian Spanish and Arabic one of those is Chinese and one is Japanese so my apologies for not knowing that thank you Chinese first says Marie Francoise in Japanese second so the way that the questionnaire was distributed there's no list from which to draw a sample a representative sample of scientists we don't have a worldwide list of scientists we don't even have country-level lists of scientists so we sent it to people that we knew some people sent it to lists of people that they knew but still those lists weren't necessarily representative and then there were ways that people were asked to forward the questionnaire to their colleagues so there were instructions at the end of the survey in the cover letter that said you can forward it to your colleagues this is a way that social scientists used to get questioners out when there's no list of the group they're trying to study but that means that it's not representative of the entire population indeed we don't know who's in the entire population and we can't make generalizations to the population however this method does allow us to make important comparisons between men and women who answered and I think there are many things that we can learn from the results of this survey but I wanted you to know that this was how it was done and so when you see the results of the survey you can't conclude that you know scientists in North America are like this and scientists in Western Europe are like the other thing because it's not representative not necessarily representative of those scientists here are the number of respondents by discipline and these are the disciplines I didn't mention them at the beginning but they're represented by the partner unions so astronomy biology chemistry computer science math math applied and physics there were also many well not many there were some respondents in the history of science which is one of the partner unions and some people chose other disciplines because it was open to basically any scientists and some people didn't answer that question so in total we had about 30,000 respondents 21,000 or 22,000 from these disciplines about half for men and half for women so this is again not representative of the actual population of scientists because we know they're not half men and half women for most of the disciplines however women do tend to respond more frequently to surveys than men and we were also targeting women because we wanted to make sure we had enough women to do our analysis so we were pleased with the number of okay in the results that you'll see applied math is shown as a subset of math so I don't know if you have any slides about that but we'll see okay now there were I wanted to talk to you a little bit about our analysis plan in your report there's some bivariate or what we've called bivariate results because these are easy to see on a chart Susan here is holding up my colleague is holding up some bivariate charts the reason for this is it's easy to it's easy to visualize you can see like here's a bar for men and here's a bar for women and you can see what the difference is between the two genders you can also test in that simple way to see if there's a statistically significant difference between the women and the men and so it's a good way people like to look at these bivariate charts however this analysis can be confounded by other variables that are not accounted for in the chart for example we may have you know it may look like there's a difference between men and women but it really could be that the men in our group have more experience in their careers than the women so for this reason yeah it's different career stages let me go to the next slide so we also did and you'll see this at the beginning of our report what we call multivariate analysis these are statistical models where we can account for these confounding factors such as age discipline employment sector it's hard to visualize those results but Susan's going to give them to you and what you see in here is for example if we show you the results of a multivariate analysis that means and there's a gender difference that means that controlling for all the other things that could make a difference in that variable we still see a gender difference so it's a more robust way to look at at the gender differences so as I've already mentioned we did both bivariate and multivariate multivariate we think have the stronger results so again we're going to we started our report with the analysis of men's and women's differences overall and we did some bivariate analysis which you see in the report we looked at tables with gender within each discipline gender within geographic region and gender within a grouped level of development that was using the United Nations Human Development Index and now I'm going to turn this over to Susan who's new to the project but not new to the American Institute of Physics and she's going to talk about the results thank you Rachel promised I would show you a map and I did have a map in the earlier version of the slide so now I'm going to show you a map this is the second page two is the map of the the regions the geographic regions we used so and you don't really have to know it I just wanted to fulfill that promise that Rachel had made when two people do a presentation both of whom are jet lagged at the very end and trying to remember what was going to be in the slides but I'm a statistician and when you do statistics you have to worry about a type one error risk it's like a false positive it's when if you reject a true null and if you've ever done any hypothesis testing you know about that magic alpha and if the p-value you get is less than alpha you reject the null hypothesis in favor of the alternative when you're doing so many tests because if you look at that report there there are a lot of those bivariates in there and there we ran a lot of multivariate analysis if you don't adjust alpha you're gonna things are gonna show up as statistically significant even when they're really not because you've made a type one error of course you don't know you've made a type one error so the way to get around that we use Bonferroni we adjust alpha you make alpha lower you divide by the number of things in a family wise comparison and so instead of using 0.05 or 0.01 which are common we use 0.002 so that that's really all you need to know if you would like to have a discussion about type one analysis errors I'm here all week so to cut to the chase I'm now looking only at the multivariate analysis there was only one item we looked at which did not have statistically significant differences for men and women now again we're not testing for the population this test is if we had a different group of randomly selected respondents if the snowball had gone to a different group of people would we have still gotten this result that that's the statistical test it's not about the population so only one in every other and that was when did you choose your discipline there was no we couldn't find after accounting for the confounding factors we did not see a difference in when you choose your discipline everything else we looked at there was a difference now I'm gonna give you a test although the answer is already up there if you think women's experiences no no let me go the other way if you think men's experiences were more positive than women's raise your hand let's stretch a little bit and get awake I want to know your wake out there okay thank you yes you're correct in every instance every instance men's experiences are more positive than women's that's after accounting for all the confounding factors we use logistic regression and ordinal logistic regression because the dependent variable in some cases was yes no that's binary so that's logistic regression in other cases it was strongly agree agree neutral disagree strongly disagree so that's an ordinal variable or ordinal logistic regression so we use those two things with logistic regression it's easy to look at a coefficient and come up with something called an odds ratio how much more likely is one group to do something than another group and so in this I do report odds ratios with the ordinal logistic odds ratio is more complicated because it depends on where you started on strongly agree or strongly disagree and how many levels you want to move so I just report the direction of the movement as I just said now reminder and you may think Rachel and I are beating this into you but I'm stressing stressing stressing same age same discipline same employment sectors same geographic region same level of development that's what we mean by holding all other variables constant so when I look at this I'm looking at a man and a woman for whom everything else is the same they're the same age they're in the same discipline they work in the same employment sector they live in the same geographic region and they're in a country with same level of development by the way for the multivariate we use the the actual score the human development index score for the by variant we classified it into two groups we had highly developed and very highly developed in one group and medium and low development in another group so that was the by variant because to do the by variant you have to have discrete groups but in the multivariate we could just use the we'll call it continuous variable so it's analysis of the gender gap so here we go and let's smile timing of your when you chose your field and we the question we use was when did you choose your primary field of study so after accounting for the age of the respondent the discipline of the respondent the region the development index we found no statistically significant difference between men's and women's responses so that was that's the happy news doctoral program experiences how would you rate the quality of your doctoral program after accounting for age discipline geographic region development index men considered rated their program quality higher than women surprise in my doctoral experience I had support for my advisor or supervisor men were more likely to agree or strongly agree than women after accounting for what age discipline geographic region development index so I'm trying to stress here you can't people can't look at these results and say why else because they lived in different places are they studied different things are they're different ages no we've already accounted for that this is after accounting for that interruptions in your doctoral studies this was a yes no question so I can report an odds ratio women were 1.6 times more likely to report an interruption than men current workplace experiences and for these I also wanted to use employment sector because I can imagine they're working in academics is not the same as working in a primary or secondary school or working for an NGO at my current job my employer treats everyone fairly men were more likely even after accounting for age discipline geographic region employment sector and development index co-workers are respectful of everyone men were more likely to agree I feel like I'm we're bringing the mood of the room down career progress yeah we asked respondents how do you compare how do you compare your career progress with people who graduated at the same time as you men were more likely to report a faster response to the career progress than women now again these are perceptions we don't have any absolute measures of career progress we didn't ask for rank or salary or any of that so these are perceptions but perception can be reality do you think your salary is same higher lower men think they make more that's regardless of age discipline region development index and employment sector so here's one of the career advancing opportunities for example men were one point two times more likely to have been invited to give a talk at a conference we have we have several questions on discrimination one of them was I have never experienced discrimination men were five times more likely to say I have never experienced discrimination than women five times and that's not even the biggest number has your career influenced your decisions about your personal life women one point six times more likely to say yes than men becoming a parent so we had a statement my career a rate of promotion slowed significantly when I became a parent women were about three times more likely to agree my worker career did not change significantly men were three times more likely than women to agree I mean that that's what you call consistency in your results I did not make up those numbers that that's what they came out to be here's the number I have encountered sexual harassment at school or work speaking specifically now it's a sexual harassment after accounting for age after accounting for discipline after accounting for employment sector after accounting for geographic region and after accounting for the human development index women were more than 14 times more likely to say yes than men and again you can't say well that's because this person works in this industry and this person works you know in academics you can't say oh it's because this person is in astronomy and that person's in biology no we've already accounted for all of that dual careers in physics I've come from the American Institute of physics although I'm a statistician in physics this is called the two-body problem but it's dual careers so you know finding two two jobs together if you're in the same field it can be a lot harder to find two jobs in the same geographic area and women were three times more likely to say their spouse was in the same discipline than men so women may have a harder time dealing with the dual career issue so that was all the men and women that we were just looking at men and women guess what we found the same thing when we looked at geographic region we found the same thing when we looked at disciplines we found the same thing when we looked at levels of development we found the found the same thing when we looked at employment sector okay so that's the sad news now the happy news with the multivariate analysis you know I can't look at these other things and that gives us an opportunity to learn some lessons maybe there are places our disciplines our employment sectors that are doing you know that that women's experiences are more positive than men's experiences are at least the same as men's experiences and if so you know maybe we can learn lessons from those so doctoral studies I think the mathematicians are gonna like this one respondents who had studied in math programs were more likely to have a positive relationship with their advisors than respondents in all of the disciplines that's after accounting for age region development index so what is it about math Marie Francoise what is it about math doctoral students who had studied in northern America rated their program quality higher students who studied in northern America and in Oceana were more likely to have a positive relationship with their advisor maybe we could go learn lessons in those countries and regions so when we get it's a good question thank you this Marie Francoise saying this for men and women yes when we're looking at at the other things when we're looking at region when we're looking at discipline we're looking at men and women together we did not do the analysis to distinguish between men and women or we haven't done it yet that's a more involved analysis so this is just students in this region and students in that region the higher a nation's score on the human development index the lower the rating of program quality the lower the rating of advisor relationship perhaps what because I come from a very highly developed country I think well that can't really be true can it because that's where I went to school so for all of these we don't know if perceptions are driven by higher expectations by lower expectations but there's certainly reality people are perceiving so what what can we do about that so in some cases these may differ from the bivariate analysis because we're now taking into account confounding factors like discipline like age like employment sector so yes where they differ it's the bivariate may show men having more positive experience in some area than women and if it doesn't agree with a multivariate analysis that means that difference really isn't because of men and women that difference really is because of some other confounding factor did I mess up one of them's clearly wrong I will check and I will let you know yes thank you for catching me on that so we'll skip past that side all right right sorry this yeah right they're not wrong this is the development index which yes northern America is highly developed but it's not the only highly developed thank you thank you Rachel so this is this is holding geographic region constant and looking only at the human development index would that be like comparing only very highly developed countries to each other so it'd be like comparing North America to Western Europe yes that's exactly what that is thank you now is everyone thoroughly confused okay let me here we're looking at the map and we're holding the development index constant okay so there can be countries in a geographic region with different development indices okay so here that doesn't matter the development index is being held constant so essentially it's in the same country so northern America rated it higher than the other regions and there are also other regions that are very highly developed or largely very highly developed because you can go to that map on the second page to and see the regions this is now looking at the development index so there are other regions that are very highly developed in addition to northern America so overall those regions where the development index is higher actually it's not the regions it's not the country level because the development index was about a country tend to rate the quality of their doctoral program lower so you have to step back and remember what's being held constant everything else so here we're keeping the region constant changing the development index here we're keeping the development index constant and changing the region it's a subtlety I agree and one that I was gonna walk right past thank you Rachel respectful cohort was that another question yes holding a variable constant is a very good thing in statistical analysis it means that I can consider only here I can consider only changing the human development index but I'm not going to change a region so I might have to move from one country to another country in the same region but the development index changes yes no yeah let's hold that till after because I'll go through these and I promise I will come back to holding everything else constant because that's a it's a very long answer respectful co-workers look at all these employment sectors where the respondents reported feeling they were treated more respectfully than respondents in academics and government entities what can academics learn from industry it's a rhetorical question yes sorry respondents in northern Europe were more likely to agree their co-workers treated them with respect the higher that a human development index the more likely the respondent is to agree that their co-workers treated them respectfully and again those last two are rhetorical questions discrimination here's where the math and physics people will be you know happy to read this result respondents with the disciplinary background in math and in physics were more likely to have never experienced discrimination than those in other disciplines Rachel's theory is that's because they're all men but the gender was held constant so that doesn't matter the higher the human development index the more likely respondents were to have reported never experiencing discrimination northern America northern Europe Oceana however were less likely to report never experiencing discrimination so this is another one of those that's going to bother you with everything held held constant and I promise we'll come back to that because those seem counterintuitive I understand industry and primary secondary schools respondents working in those sectors were less likely to say that their career influenced their decisions about their personal lives respondents working in government and industry were more likely to say their rate of promotion had not changed after becoming a parent's respondents in northern America and South America were more likely to say their career influenced their decision about their personal decisions than respondents living in other geographic regions respondents working in Africa and the Caribbean and Central America East Southeast Asia northern America Western Asia and Western Europe were less likely to say their rate of promotion slowed after becoming a parent so this suggests that there are regional differences in the work-life balance so you can go to that map on on the second page to and all the regions that aren't listed it's what I'm comparing to however if we look only at the human development index there are no statistically significant differences about your career influencing your personal lives so based on respondents from about 30,000 there were 32,346 respondents to the first question but we had enough data from about 30,000 from 159 unique countries for our respondents women's experiences in science are less positive than men's experiences after accounting for age and we used age because we asked what year you were born and I'm using that as a proxy for your career stage the older you are I'm assuming the more advanced you are in your career the younger the less advanced discipline employment sector geographic region and level of development and that's the last slide so holding everything else constant yes 30,000 is about what we were using and for 8,000 so all of those were included in the analysis I've only pulled out the stuff for those sectors for those disciplines so the question is we had 30,000 and 20,000 so how many are actually included so I was trying to highlight in that one slide with the 21,000 22,000 the the disciplines that were listed there we used all available respondents in the analysis and so we also had history of science we also had other disciplines I just pulled those out to talk about so so there were also several there were also several questions on the questionnaire about your discipline there's the discipline you studied I don't know Rachel you know that questionnaire better than I do but it's possible that we're looking at different discipline questions it's possible that this slide referred to one discipline question and the other numbers refer to a different one and we can dig deeper and get you an answer on that yes that's true it can't so age is a proxy for career stage it can be different for men and women and that's definitely true but since we didn't have a question that ask about their career stage that was the best I had and could use yes so how much support did you get from your families and we frankly for the preliminary report ran into a time crunch and it had not run that variable yet in the multivariate model but I will so in the by variants it did show up that women were more likely to receive support and I have not run that in the multivariate yet and I will there was a question over here first and then we'll go back over there yes to present it on the map okay thank you that's a good yes okay so the question was when I listed out all those regions it might be easier if instead of listing out a bunch of regions if I presented that with a map you know with some of them brighter and some of them less bright that might be more helpful and what my friends was said about women reporting more support from families I think it's true because men actually get support from the wives or from the partners and also families probably so yes we will look at that in the multivariate yes I guess it's difficult to do the survey but if you could ask about how many women dropped their career because they did not have support exactly but that is should be another rhetorical question because you have how did career influence your marriage and family but what about the converse so how did the family relationships affect your career well I mean we have some of that like here's the dual career question becoming a parent did my career change after I became a parent so we have some of that if I understood well most of the survey was on perceptions on realities and that's important but humans is notorious to have mistaking perceptions so do you have some more objective data to compare these some at least some of these results to and do parrots if it's really a grand problem or if we need more propaganda of sorts so the results mirror we could not have objective measures of career progress because that's different in different countries but the results mirror studies that have been done in Europe and Northern America in the US that do look at different rates of promotion and different rates of tenure so those are shown to have family situations has shown to have effects on that in those regions we don't I don't think there's data on some of the other regions you but your point is well taken comparable proportions those studies didn't calculate the odds ratios like we did but I can give you references if you want during the break I think Susan and Rachel our eager Twitter audience is eager to know whether the results of the survey will be made public and I've answered that there will be a to a certain extent in the report that will be published but will there be also available in a kind of URL or some other interface so now the project is not finished as I said we are going to discuss the results and recommendations and you know so you get this report as part being a member of this group you don't distribute it because it's not final it's not the final report once we have a final report the final report will be published on the website of the project it will be also published as a book but for the moment this is not yet the final report the final report will look like that it will include updated reports from the three tasks and for example Suzanne just said that she will include something more and then probably also after the discussions of the two other tasks we are going to include something more and for the synthesis and recommendations we are going to include something else so it is not the final report the final report will look similar to that with of course more references also it will look more professional it will be a book and it will be available online for the moment if you distribute it please say it's nothing but a preliminary report we have already a long list of typos and things like that so it's not the final report but then the idea was that people could submit proposals to analyze part of the data are you going to tell how that's going to happen no oh the question was if the data was going to be made available so if people the results the results of the oh I understood it was about the data okay so there so that's a more complicated question and it's not yet a result for the moment the data are with AIP and there are conditions for people to access to the data which are to go through an application and so on so we are it's part of what we are discussing right now we don't know if the data are going to be available we hope they will be but we don't know but as in terms of the results of the project as analyzed by the various task groups they will be available and public the issue with the data is that the respondents were promised confidentiality and it doesn't take many cuts in the data until you get down to where in some of the smaller fields or even not in a small field you could potentially identify people so that's why we can't provide an answer to that question right now because we're still working on it and data suppression techniques and things like that hi at this stage I've got two comments one is I think it would be wonderful to compare these results with quantitative survey results perhaps through saga surveys or perhaps through the surveys that have been done from PISA and see if similar results are emerging and this is a comment for the question okay and the the second comment was I believe that results are emerging where there are major gender gaps and these can guide us in what we actually do what actions we undertake and I'd like to check that I'm right that some of the major ones in the gender gaps are the experience of discrimination and in other parts of the survey discouragement we know that women leave faster than men from other surveys the slower career after becoming a parent and the experience of harassment and I think these provide guidelines towards where the major impact could be made in terms of our action thank you so regarding comparing to other surveys it could only be done in a very broad or general sense because when when you ask a different if the questions are worded differently then the results are not directly comparable so this is a question for me Rachel my friends was in the first meeting in Paris we discussed about that availability and maybe by discipline maybe by region and we decided I thought that there would be a committee formed and that committee would look at a request for data and decide whether the data would be handed or not and my question is has been this committee has been already formed or not yet not to that not not my knowledge but so he answered the sorry the answer is yes we have a committee which is the coordination group of the project but the only request that was examined this year was coming from a group in Russia they wanted to get data from Russia and the Russian zone in order to do some gender gap analysis and it was we can say poorly formulated so there were some back and forth connection with the group saying please reformulate and say how you are going to protect the data and so on we get we may be received three successive answers and at the end it was not satisfactory and at the end they came back there is a mechanism but in but the only example we had didn't work so okay so let's try to do it for mathematics maybe it will work any other questions okay I'm gonna go back to holding other things constant and thank you so much for asking that because I now get to put up some slides that this is fine I now get to put up some slides that I originally didn't put up so let me just warn you you're about to see a lot of numbers I mean a lot of numbers please don't focus too much on the numbers just look at the technique I'm looking at Eagle she has seen these slides the this is the output from the multivariate regression for how would you rate the quality of your doctoral program well basically we're just saying your answer quality is a function of all these variables I'll post constant so that's that's what we're looking at and so what we mean by holding other things constant we mean not changing so so these are the coefficients and then you have the value which in some cases is a binary yes no one zero from in that region or I'm not in some cases is a continuous the age the development index so when I say holding everything else constant I mean I'm changing the coefficient for male female and holding everything else constant not changing any of those inputs does that answer that question now better and then the way you look at this so you also have to have a reference category so for male we made the reference category female reference categories are just arbitrarily picked the results don't really matter on which one you pick you just have to pick one yeah so here your input would be one if you're male zero if you're not and so there the age is just your age there is there's a history and philosophy of science which we did include in the regressions but not in the report which I can go back and correct so the first thing we have to do is I didn't finish reference category for the disciplines we arbitrarily picked astronomy again it's arbitrary the results don't change for region we arbitrarily picked Western Europe and we didn't have to have one for the development score because it's a it's a continuous variable it's from zero to one so and these are in different groups so you can only look at one group at a time so what do I mean by that well for example here I'm looking just at gender I'm looking just at gender so the first thing I have to do is I have to look at that p-value you always have to look at the p-value first and my p-value that I was looking for was point zero zero two if I'm less than point zero zero two I say it's significantly different yes it was within a series when we had a series of questions that were all basically the same question with a lot of subparts question was how do I calculate the denominator for the bond froning and it was by pieces of the questionnaire I think we used 14 questions by discipline by region by right then yeah so you have to multiply she's saying you have to multiply yeah so I'm multiplying the number of parts by two by two for the gender by however many regions by however many disciplines to get all that in there and then I'm using that as a denominator maybe that's clear as mud would be what I would say in Texas so oh well I'm not I'm not three yet then I can look at disciplines so now I'm just looking at the blue section I doesn't look blue back here I'm just looking at this section right here that's bigger and bolded well it's only bolded if it was significant and so here I have biology and math that were significant you also have to know the value whether high is is a more positive answer or more negative answer okay so I think we are going to finish here if you want to keep on discussing with Susan and Rachel you can do so during the coffee break so we'll be back in half an hour or what time do you think 11 so we will be back at 1115 thank you