Tableau Projects for Practice | Tableau Projects for Data Science | Tableau Training |Edureka Rewind

Key Takeaways

hi all I welcome you to this session on Tableau projects and in this module we are going to see how you can Leverage The Power of AI to explain data points with a single click on Tableau Tableau functionalities are based on Advanced statistical models which explain data and help you uncover insights you may not have found otherwise but before we begin let's look at our agenda for today so we are going to start out by introducing you to Tableau and then we are going to see how we can install Tableau desktop on our computer systems then we have three different case studies on three different data sets so first of all we have taken a foreign direct investment data set and we're going to analyze it from the perspective of your company and see that in which sector will it be most beneficial to start work so that your company can collaborate with the FDI and become successful next we have a case study on Boston real estate and here we shall look into a data set mentioning various factors and outcomes involved in real estate prices in Boston us and understand the correlation between parameters and the rates of the plots and finally we have a fun little case study on Imdb In which data set we are going to conduct exploratory data analysis and understand the relationship between various parameters in the data set using Tableau and finally we are going to conclude the session so with that I come to the end of my agenda so without Much Ado let's get started so first of all we are going to start by looking at what is Tableau what exactly is its significance so let's go right ahead so Tableau is basically a suite of services which Aid in business intelligence of a company or Tableau you can harness the power of your data and unleash the potential of numbers you can choose the analytics platform for your individual analyst for teams and organization for embedded analytics natural language interrogation of your data and see how your organization can create an impact using that data on the market from connection to collaboration Tableau is one of the most powerful if not the most powerful secure and flexible end-to-end analytics platform for your data it is designed for individuals as well as scale for the Enterprise and it is not only a business intelligence tool but it also turns your data into insights that drive action for your company now Tableau has a couple of different functionalities as I mentioned it is a complete end-to-end solution for business intelligence it has a tableau prep which is built of two products prep Builder and prep conductor now this is a personal data preparation tool that empowers the user with the ability to cleanse aggregate merge or otherwise prepare their data for analysis in Tableau then you have Tableau desktop which can basically extract massive data offline or from the web and bring it into the memory of your system for Limitless exploration in seconds it is a data visualization tool that helps you get actionable insights out of your data really quickly and then you can share your insights using the Tableau Servo or Tableau online which is essentially an online hosting platform to hold all your data workbooks data sources and more now moving on let's discuss a couple prerequisites so that you can follow through with this session you have to make sure that a version of Tableau desktop is downloaded and installed in your PC or laptop or your system whatever you're using right now so for me especially I am using the Tableau desktop public version which is the free version of Tableau desktop it basically has all the functionalities you have for Tableau desktop the paid version but it comes for free and you can easily create stunning interactive visualization on this platform without any coding you can explore and interact with the most extensive library of data visualizations in the world with over 1 million user generated possibilities as well now to download and install the Tableau desktop public your first log on to the Tableau public official website so first of all on your web browser you will have to login to public.ablo.com and you will be met with a web page like so all you need to do is enter your email address use your work email address and then click on download the app button which is this big orange button right here now this will start downloading the exe file for Windows by default because I'm using a Windows PC and you can see the downloading process in the bottom left corner of your website which is right here now this is going to take some time so kindly be patient now once it's done you can open the download file okay and you'll be met with a setup prompt like so here you can check the box which says I have read and accept the terms of the license agreement and then you can begin the installation by clicking on install that will start the installation process and once the installation is over it will automatically display the start screen of tableau now the start page of Tableau public looks something like this conveniently named as book one which basically means it's workbook one and you can connect to a bunch of different data sets by choosing one of the options on this panel on the left and you have a bunch of different tutorial videos how-to videos and visualizations of the day in the panel on your right so now that we have our Tableau public downloaded let's go ahead and look at our case studies and try solving them one by one I'm sure you must be tired of hearing this again and again but without much to do let's get started so our first case study is going to be based off of a data set on FDI or the foreign direct investment for those of you who aren't aware of this term a foreign direct investment or FDI is an investment in the form of a controlling ownership in a business in any country now this is by an entity based in another country it remains distinguished from a foreign portfolio investment by notion of Direct Control now we're going to be dealing with a data set which has various different sectors and the direct investment in them over the years now if you're a company or a startup and you want to venture into a certain economic sector you obviously will look at the sector which is most profitable to you now a case study is based off of that we shall look at various sectors and determine which of those will prove most beneficial to collaborative it as a company as a venture and we are going to answer a couple of questions along the way like the maximum and minimum direct funding for a given year we are going to also find out the trends for direct Investments for different sectors and how they can be grouped or clustered together we are also going to determine the sectors with the highest and least growth in the last five years and sectors which have a little shaky performance throughout the years we're also going to find out if there is any correlation between certain sectors from an FDI perspective and Are there specific clusters to present the data and finally we'll try to forecast the trend for the following year with that said let's move on to our Tableau start page and then connect to our data all right so this particular file is going to be a an Excel file so I'm just going to go to where all my files are and this might take some time so kindly be patient okay now this is our data as you can see the first row has sectors followed by all the other rows which are basically fiscal years and the amount of direct investment per year now as you can see apart from all of these data cells that we need we also have a few extras like showing 1 to 64 Power by visualize some rows that we do not need even the ones on top so what we can do is we can click on this option what says clean with data interpreter and now finally we have an automatically cleaned data set where we have all the columns and rows that we need you have all your sectors you have metallurgical you have mining power non-conventional coal production so on and so forth you have diamond gold ornaments tea and coffee miscellaneous and you have the grand total and then you have the direct Investments made in those sectors along the years now we have data starting from 2001 to 2017. now another thing is that we don't actually need the grand total because here we are trying to find out which sector will give us most profit if we don't take out the row grand total right now Tableau is going to start considering it as a sector of its own so for that we are going to add a filter you can always go back to your Excel sheet and clean this up but my purpose here is to show you how you can do it with Tableau so what I'm going to do now is I'm going to add a filter I'm going to click on sectors and then click on exclude I'm going to check the box exclude right here and I'm going to search for grand total here it is so I've excluded the grand total Row from my data set I'm going to click on OK now this is the simplest way in which you can clean your data in Tableau you can also go ahead and click on this link to review the results what this is going to do is it's going to open an Excel sheet where you have the key for the data interpreter the data that it has selected to keep you can look at all those things right here you can look at the data subtables okay when you go back to the key for the data interpreter you can see the different patterns and colors of the cells and what they indicate for example the red line indicates that the data is interpreted as column headers and if you go here you can see that the entire row number five is basically used by Tableau as header and that's all there is to it let's close this and go back to our Tableau Data set now let's look at what we can determine using the data that we have now before we begin I'd also like to point out something that the data we have is presented in a white format which is usually the most common type of representation of data government data sets and institutional data sets are usually presented this way now the problem with this type of format is that it is good for viewing purposes if you're just going to go ahead and look at numbers but from an analysis standpoint this quite doesn't hold so what I'm going to do is I'm going to convert it into a long format so I'm going to start out by clicking on the year 2000 to 2001 and then I'm going to go all the way to the end and press shift and click on this this is going to select all of the your columns then on any column I can just right click and select the option pivot and as a result of that we have three rows now what I'm going to do is the sector remains sector pivot field names I'm going to rename into year and the pivot field values I'm going to rename into FDA now what happens if we do this is now we are going to get all of our columns under Dimensions through which we are going to have a little more freedom to play around with our data which would not have been possible if we had chosen the wide format to work with yes okay so another issue I have is the representation of er sure the ear is given by a start value and an end value which I think I'm going to change now this is great for viewing purposes but I wouldn't prefer it for carrying out data analysis now this might differ from case to Case and you might feel differently there is nothing wrong with this it's perfectly fine it's just not in a way in which I would prefer it so what I'm going to do is I'm going to add a calculated field since we've already named it ear I'm just going to call it this clear or financial ill you can also call it f y and we should be using a function known as left so I'm going to put in your comma 4 and at the bottom it is going to show the calculation is valid so basically what I did was I'm excluding all characters from the left after the fourth space so when I apply it you can see my fiscal year only has the first four characters of every year which is 2000 in this case and 2001 and the next and now I have no use for this particular column which is the regular your column so I'm just going to hide the column now I have all three of my columns ready to go and I think I'm pretty satisfied with my data with that I'm gonna move on to my worksheet I can see all of my dimensions and measures here on the left in the data pane now for those of you who are new to this this is what a tableau worksheet looks like on the left you have all your data in the form of dimensions and measures we're going to play around with that and then you have a couple of shelves having rows and columns which is exactly how it is in Microsoft Excel except for here you can be pretty specific and drag your Dimensions or measures into rows and columns and your rows and columns are going to change from Context to context then this is our canvas this is where the visualizations are going to be viewed and on the left you have Pages filters and marks now these are the cards which are going to help us create more intuitive visualizations out of the ones we already have on the right top corner you have the show me option and right now you can see a couple of grayed out visualizations but once you have data appropriately input into the rows and column shelf it's automatically is going to highlight the graphs that you can make out of it so yeah with that we can get started right now I am going to go back into my presentation and here you can see that I have already figured out a couple of questions that we're going to answer through this exercise so first let's focus on the first two questions which is determining what sectors received the maximum direct funding or the minimum direct funding for a given year both of them are questions of the same nature so let's go ahead and find out so basically now what I'm gonna do is I am going to drag my sector Dimension into rows right here and you can see that all of this has been arranged in alphabetical order it's not in the order that was presented to us in our data set yeah then what we're going to do is we're going to get our FDI value and drag it to our table if you hover on the table you can get the various FDI values now all of these values are in crores because this is based off of India for people who are based outside of India one crore is basically 10 million so you can go ahead and do the math so the question said we needed to find out the maximum and minimum values for a given year so I'm going to drag this clear into the filters and here it's going to open up a fiscal year dialog box I'm going to go down click on 2016 because in cases like these we usually prefer the latest data and I'm going to click on apply and then okay you can just go ahead and click on OK and then if I go up and click on either of these sort buttons for example I'm going to click on the sort button for descending order you can see that my entire sheet has been arranged in a descending order and now I can see that the five sectors which get most Investments include the services sector telecommunications computer science and Hardware trading and electrical equipments and at the bottom we have photographic raw film and paper mathematical surveying and drawing instruments defense Industries coil and coal production now this is a great information if you suppose an analyst or you conduct market research this is actually great information that you could go ahead and give to your managers so that they know which industry is to invest in of course this is not the only Factor there are going to be many factors which we are going to discuss through the course of this project but this is simply one of them and the most fundamental if I might add so now if you want to take up this list to your manager and say Sir omam these are the industries in which we should invest you obviously would not take the entire list to them obviously you would take the top five or maybe top three top 10 of the sectors that deserve to be invested in right so for that what we need to do is go on filter go to edit filter here we're going to select all and then go to top and click on the option by field as you can see it says stop and then it says 10 I am going to change this to top five and change the sum to maximum and I'm going to apply that value and click on ok now you might have noticed that the order of the sectors has changed and that's only because we have selected all the years instead of just 2016 which was previously selected so you will see right now that the top five is not being filtered out and that is because we have not filtered out our sectors so we're going to drag the sectors into the filter and then we are going to this tab called top select by field again top five and then apply it okay and now in front of you you have all the top five performing sectors in which there have been most foreign director investment and I'm also going to rename the sheet let's call it Max FDI because you know it's always great to be organized so with that let's move on to our next worksheet for that I can just click on this new worksheet button or I could do something else and I could duplicate the max FDI sheet now what happens when you duplicate a sheet is that all the filters that you have applied on the previous sheet are the calculated Fields get copied along with the sheet which is why this is a greater option if you have to make my new changes and don't want to apply all those filters again so I'm going to rename this instead of Max FDI 2 I'm going to rename this into minimum FDI and all we have to do is we are going to edit the sector filter and instead of the top five we are going to select the bottom five and click on apply and here we have the bottom five values which are industrial instruments cold production so on and so forth so now that we have our maximum and minimum FDI sectors next what we have to do is we have to see if there's a trend for direct Investments for individual sectors and what I'm thinking is we could make a filter for individual sectors we should be able to filter out different sectors so now to determine Trends I am going to move on to sheet 2 and I'm going to again rename this as overall trends all right so what I'm going to do right now is I'm going to drag my fiscal here to the columns and the FDI value to my rows so almost intuitively Tableau has turned this into a bar graph but since we are showing Trends I think a line chart would be more apt but if I drag down here to the line chart either of the line charts it basically says it needs one date here at the bottom you will see requirements for a chart so if I hover on line chart you can see for lines continuous try one date Zero or more Dimensions one or more measures now we already have our date so why aren't we getting the line chart option now majorly because our fiscal here has been imported as a text format so basically we're gonna go on the drop down arrow change data type and convert this into date and the moment we convert it into a date it automatically turns into a line chart almost instantly and we get the trends from 2000 to 2016 17 with our line chart and it's a clear indication of what has happened this also might be the reason that sometimes you might have taken up the date as one of your columns but you still don't get the option for a line chart you can go ahead and check the data type in which your date column is imported and if it's not date you can turn it into date or date time and you will open up the option for a line chart now what you can see right here is a couple of high points 2008 2011 and a couple of down points which start at 2008 the downfall and go down 2009 and 2010 and I think it's pretty obvious because 2008 and 9 the fiscal year of 809 was actually the year of the global economic crisis and Global recession so it makes a lot more sense if you think about it that way and then it's short back up in the year 2011 and carried on to our current fiscal year which has most FDI value 2016 and 17. now if you wanted to make the chart a little more interesting what we could do is we could bring down the FDI values and maybe drop it on label this way we have labels on all of our data points and you can see how the FDI values have changed over time it started with some 2400 crores and now it's at 43 000 nearly 500 crores or you could call it 435 billion you can see all the trends changing overall the highs the lows the little bit of recovery after the 2008 debacle here in 2011. for that we have our overall Trends moving on so now if I go back and see what next I see that we have to find out the overall trend for individual sectors so for that we obviously have to use a filter let's go back and finish what we started in the overall trends so putting up a filter is pretty simple you can either drag your sector Dimension into filters and you will get a filter based off of sectors then you can select all apply click ok and then you can right click on the blue pill which says sector and click on show filter then it'll show the sector on the right or you can simply click on the dimension sector right click and use show filter then if I clear all the check marks I can just go on individual sectors and see how they have performed throughout the years let's take education for example as you can see there has been great investment in education in the year 2013 closely followed by 2015 and 2008 but since we have applied this filter this particular line chart type is kind of odd to be looking at so what you can do is you can click on pass in the marks card and select the kind of line you want in your line chart so one of the options is this jump line chart but what I really like is the step line chart here you can see individual Rises and Falls of every sector that you choose if I cut out on education and if I click on the search button here and select maybe Railway even Railway has seen massive growth between the Years 2013 and 14. I wonder what happened was it the year of the election was it the year after I am quite leaning towards a change in political parties in power if you guys know what happened in 2013 which affected this change do let me know in the comment section below maybe let's try food click on food processing industry again between the year 2013 and 14 you can see a massive growth in the investment in the food processing Industries and with that I go back to my general line graph I think I'm Gonna Keep it this way because I quite like the whole step line graph it will also look good on my dashboard I think okay so now let's go back to our presentation and see what do we have to do next so the next action of the item is to group certain sectors together based on similarities so I'm gonna go on to sheet four and I'm gonna get all sectors here if I right click on the sector pill in the dimensions you can see the create option where you get options like calculated field groups and sets I'm going to create a group and here I'm going to find similar looking groups and group them into a single group so I am going to group agricultural machinery and agricultural Services into one I'm just going to call it agriculture and similarly I'm gonna make a couple more groups just to show you guys and once I click on apply you can see that we have the groups that we had made earlier right up top here and if we bring our FDI values we have combined values from all the sectors that are present in one single group and that's as simple as it is so I'm going to rename the sheet as groups so moving on next let's see what else we have to do okay now we have to find out sectors which reported highest growth and highest decline in the past five years for that new sheet and then I'm going to drag the sector Dimensions into rows and I'm going to drag first clear to the columns just gonna drop the FDI value in there and as you can see remember the first few numbers so you can see the change happening what I'm gonna do is I am going to change the measure from some I'm going to go on quick table calculation and change it from sum to year over year growth and now as you can see all of the numbers we had changed into percentages except for one year which is 2000 and obviously you cannot see year over year growth for 2000 because we do not have the data for 1999. so now we're gonna drag our fiscal year into filters and now you can see we have our latest five years but 2012 is not going to show any values it has a completely blank column because of the same logic that I had mentioned earlier when all the years were in display because we filtered out all the years before 2012 by that comparison has no year-on-year growth so we are going to change the top five into top six so we actually get our top five and voila we have 2012 through 2016 and we have the latest years and now what I'm going to do is I'm going to click on this row 2016 and sort everything in a descending order so we can get our top five which is glue and gelatin cement and gypsum scientific instruments electrical instruments and telecommunication which are giving us the most year on your profit and then I select this filter and go on top five and this displays my top five now that we have our top five let's duplicate this so we can find out our highest decline in which case I am going to change this filter edit the filter and instead of top I'm going to go for bottom five and apply and as you see the bottom five sectors are defense photography mathematical and surveillance choir and coal production what happened defense Industries in 2016 almost a negative 100 growth it's kind of a shame people please go join the defense I could if I would but I can't but you guys can so go give the direct investment something to invest in in the defense so with that we have our highest growth sectors and our highest decline sectors so next as far as I can remember we had to show maximum variance for this I am going to rename the sheet I'm gonna call it variation and one of my favorite charts to do that to show variation is this box and viscose plot I can never remember the name so what I'm going to do is I'm going to drag sectors to columns and FDA values into rows and then I'm going to go into analysis and uncheck the Box okay so I basically switched them up I kind of like this view better then what I'm gonna do is I am going to arrange this in a descending order here if I hover over a certain box I can see the upper whisker upper hinge median lower hinge and the lower whisker which obviously clearly is a lot for the services sector but I'd like to point out something more interesting if you see the second and third boxes now the second box has an upper whisker of two six one three and the median is 872. the third one which is constructional development clearly has a higher upper whisker but the median is lesser than that of computer softwares which is actually why it comes on number three and not on number two and things like that are what make these statistical graphs more interesting these are also called The Five Point graphs basically because they give you five points from a separate sector or factor or whatever your data is so we have our maximum variation which is the services sector and the minimum variation which is this coil or qua or however it is pronounced I don't know this is a tableau video moving on so next we are going to find out what is the proportion of Investments between sectors from the fdis perspective for that very simple we are going to have a tree map I'm going to move on to the next sheet and rename it as portion of FDI and for that basically I am going to give sectors and FDI values and I'm going to select the tree map and it makes everything pretty visible so I'm pretty happy with that you can also put fiscal your in the filters here show filter on the right here you can see the filters now you can see your by your proportion of investment in a particular sector let's say computer software and Hardware was at top of its game in 2005 but in 2016 it was taken over by Services sector I think this is the clearer of the graphs so now that we have the proportions of Investments let's move on so there's a lot of visibility in here we can see telecommunications Trading so moving on let's see if we can find out specific clusters present in our data for that I'll bring in the sectors and obviously we are going to bring the FTI values in here now if we switch from the data to the analytics pane you can see in the model category only cluster is enabled so what I'm going to do is I'm going to drag the cluster and bring it right here if C on your right it automatically has made 11 clusters on its own but here you can add the number of clusters that you would like three four five seven any number of clusters but simply put this is basically going to get categorize all of your sectors based on similar properties so now we are going to close this and you can see from our data Tableau has picked out 11 clusters that means all of our I think 100 or something nearly 100 sectors can be categorized into 11 different clusters based on their similarities and how they attract their direct investment and the values are color coded according to the Clusters but this is going to take me a lot of hard work to sift through and I can't even imagine showing this to somebody else for example my manager or stakeholders of the company so so what I'm going to do is I'm going to click on the show me tab again and at the bottom I have packed bubble chart I'm going to select that and this makes it way more visually appealing we can see how many of our sectors are colored in the same color which represents the cluster they are in you can see that cluster 9 and cluster 11 have just one one sector each while cluster 5 includes computer hardware construction development and Telecommunications it's pretty self-explanatory I think we can also like filter out the Clusters let's see what is cluster nine cluster nine is the power sector pretty understandable isn't it and cluster 11 is the automobile industry yes again pretty understandable finally we have to forecast the trend for the following year and for that we are again going to open up a new sheet I'm going to name it Trend forecast so again for the last time I am going to get my fiscal year to the columns and FDI values to the rows and it automatically within a matter of seconds turns into a line chart so if I go into the analytics pane again and click on forecast and I drag it to my canvas it almost intuitively immediately gives me a forecast for the year 2017. the end of year 2016 it says the value is going to drop to 39 000 and in 2017 it's going to be 41 and 2018 it's going to be 44 000. now this is just an estimate with like the upper limit and lower limit grade out in a lighter blue shade for the next two years now what could be fun is maybe if we could find out say sector wise forecasts so I am just going to add a filter and let's see I work an ID so I am going to check the computer software and hardware and it seems like it will go from nearly 6 000 crores to almost half that's kind of sad now let's move on and see something else let's see what our leading sector has to offer let's select services and it comes down almost significantly as well so another way of doing this is the trend line so in the analytics pane again you're going to drag the trend line and add a linear trend line This basically is a linear equation it says to get the FDI value at any given certain point you have to use the equation given for FDI value which is a pretty big number the r squared value is 0.79 or 79 percent which basically means you can depend on it and the p-value is 0.0001 which let me tell you according to hypotheses testing if it is less than 0.05 or 5 percent you can basically depend on the value some basic statistics and that's exactly how you can forecast the direct investment or sector for the coming years based on the data that you already have now there are many other methods through which you can do this now this is just by using this a bi service and with that we've answered all of our questions now all there is left to do is create a dashboard for presentation purposes to your manager or your stakeholders or your customers whoever you might want to show it so for that I am going to create a new dashboard and all we have to do is drop sheets here and resize them as we go all right with that we have our dashboard this is our page one I've added a text box here which gives us the name of our overall workbook I've named it insights on FDI data you can name it however you like we have the proportion of FDI the Clusters the trend forecast and on page 2 we have our overall Trends with highest growth and decline we have individual filters on each of the pages and individual cluster filters on page one with indicators of the forecast here as well with that we come to the end of our FDI case study with that our project one let's move on to see which case study do we have next now for our second case study I have a data set of the Boston real estate now this was a data set which was taken maybe in the 1960s to see how different factors different parameters affect are involved in the real estate prices in Boston and understand the correlation between various parameters with each other through this exercise we will try to find out how crime affects the price of real estate how is the age of the building or the proportion of owner occupied and establishing building affect the price of the estate how our number of rooms related with the price of the flat you take and how these rates vary in time basically if you are going to create a couple of kpis which are going to help us make a decision regarding purchasing a plot in said Real Estate so without Much Ado let's move on to our Tableau desktop here I have already connected to the data it is a CSV file and here you can see all the different variables all the different columns which have different parameters we have crime rate we have the Knox rate we have age of the buildings we have tax we have the pupil teacher ratio we have the proportion of lower income class people living in a certain area and finally we have the median price offset plot so we are basically going to find out how do people in the industry determine the prices of a certain plot in Boston pretty simple so with that I'm going to move on to the first sheet you can see all your variables all your measures here on the left as usual so on sheet number one I'm not going to create any visualizations I'm going to start out with clarifying all the variables and what they mean and I'm going to put it up clearly here in place of its sheet name text and I'm also going to rename this sheet to variables so because I know all the names of columns are not very clear here I've put them up clearly okay let me just quickly put in some so here I have specified what each of the variables stand for so the first one is per capita crime ratio then z n is proportion of residential land zoned for Lots 25 000 square feet then next we have proportion of non-retail business ecos per capita then we have KNOX which basically weighs the concentration of nitric oxide in pass 4 million then we have average rooms per dwelling then we have proportion of preoccupied houses which were occupied before 1940 then we have distances to Boston employment centers and again distances to radial highways then we have the full value property tax for a ten thousand dollars then we have the pupil teacher ratio per capita then we have proportion of segregated minorities as well as lower income class in every area and finally the median value of homes in a particular estate so now that we know what all of our columns stand for let's go ahead and start visualizing so let's start out by opening a new sheet so first I'm going to bring in the price or the median value enter rows and then let's see its relationship with crime now as you can see right now we can only see one data point if this is happening with you all you have to do is go to analysis and uncheck the aggregate measures box so it gives you an unaggregated view or an unaverageed view on your screen you get to see all of the data points that exist so suppose I don't want to know about crime I will want to see where the median pricing moves if I correlate it with pupil teacher ratio in every school and here we can see there's no such rising or falling relationship let's try something else let's bring taxes all right as you can see here we are getting somewhere so once we have our average number of rooms per dwelling we can see that it shares a positive relationship with the median pricing so now basically I want to give this power to my user I want all these visualizations to be at his or her disposal so what I'm going to do is I'm going to go to the dimensions Pane and click on this drop down arrow and click on create parameter so I am going to call this measure of Interest and keep the data type as string click on list and I'm just going to add as I go but currently I am just going to put in some of the main measures so let's see age crime average number of rooms tax nitrous oxide picture ratio and like radial connectivity that's very important then I'm going to go right click on the parameter once your parameter is created you can see it here at the bottom under parameters and I'm going to right click and click on show parameter control so then you can check out your measure of interest and how it relates to the pricing of a particular area but you will notice that this is not working right now and that is because we have forgotten to establish a connection between said measure of Interest with the pricing so again we go to dimensions and we click on this arrow and create a calculated field we'll just call it so we're going to write a case statement a pretty simple key statement basically what we are trying to tell Tableau is that when we are selecting a certain option we need it to use that option as your column input let's see what all have we put in our measure of Interest there's age crime tax knocks speed Edition and finally we are going to end it with an end statement and apply it here on the left under measures you can see a measure of Interest which is our Dynamic measure and then we are going to click on OK then we're going to switch out this column from rooms to measure of Interest remove this and then we're going to see if our pricing is in average so for these kind of analysis usually your pricing needs to be an average everything else also needs to be an average and that's it now let's see how tax affects your pricing followed by crime affecting your pricing as you can see where there is zero crime the prices are really really high which is very obvious and here is the radial distance to your Highway pupil teacher ratio which is kind of more evenly distributed along the canvas and that is how you create a dynamic measure to determine the effect of various parameters on the pricing of a certain plot now that answers the first few questions in our case study next we are going to see how to reads fluctuate based on the crime rate in an area so here I'm going to introduce you to something known as bins where basically you are going to create bins of say three or four you're gonna Club two three numbers together treat categories basically so here I'm going to introduce you to something known as bins bins are nothing but ranges suppose your crime index is on a scale of 10. you can Club the first three together then the next three like zero to three together three to six together six to nine together that's how you're supposed to read bins so here is the green pill the crime pill I'm going to right click go to the create option and then bins as you can see it automatically suggests bin size as 3.3 I am going to keep it as just 3 then okay and right up in the dimensions you can see that the crime bin has been created so I'm going to bring this to the columns and then put in the price for rows and it automatically brings me this bar graph but here is one column with null value so we are just going to exclude it yeah but I'm still not happy with the representation so instead of this I'm going to change it into a box and whiskers plot I'm gonna make a few changes bring this right here but as you can see this does not open up so if you face a situation like this so you can basically go up to the median pill and select dimensions and that will give you the proper cats and whiskers or the Five Point plot and it happens so because Dimensions this option that we selected the dimension option basically brings out all the points whereas a measure Aggregates all the points to one yes now you see the median price of an area which lies in the crime index 0 to 3 is as high as twenty three thousand dollars and here are a couple outliers which go even higher and the highest price somebody is ready to pay is fifty thousand dollars and this just gets lower and lower as crime increases and you can see there's a direct relationship between crime and pricing of a certain plot and you can do this with all the other measures that there are but crime rate or crime index is a really important parameter for people to choose a locality in which they want to live so now let's go ahead and add a few metrics to this workbook so for the kpi I am going to drag out the measure name spill here and then maybe the measure values this immediately gives me a bar chart which is okay but I don't think it's an apt way of representing this data so I am going to convert it into a metric form a table form also for the sake of creating a dashboard later on I am going to swap out the rows and the columns and this gives me a swap version of the previous metric chart and I can choose to pick as many or as few of the parameters to display in my dashboard also it doesn't really make any sense that all the measure values are there in our sum form so I am going to convert them and or average you can see all your metrics change this makes more sense I'm not interested in all the measure values so here I'm going to keep only my top five measure values which I think is relevant if I had to pick a house so age of occupancy is really important Crest this I'm going to remove Prime distance the dynamic measure also I'm going to remove all right now how do you make use of it what I'm going to do I'm going to bring out the price into the filter and hence you can see all the values pricing suppose I'm interested in a house between 15 and 25 000 here I can see my average crime rate is 1.8 average distance to workplace is 4.1 I have average nitric oxide in the air pupil teacher ratio so on and so forth and that is how you turn these basic metrics into key performance indicators to make decisions on whether you want to buy a plot in a certain area or not and this can be used to create an application out of it for user benefits suppose you're working in real estate or your company has takes in it you're an analyst you have to conduct your market research this is how you can give all the power to the user and enhance user experience using tableau so how to enhance this metric further what I'm going to do is I'm going to create a calculated field and call it crime indicator and here I'm going to write a very simple if statement based on the crime bins that I created so let's see crime is less than equal to 3. then it's going to display low Prime is less than equal to ca 8. then moderate is equal to else hi all right so now we have our crime indicator this makes it easier for your viewer and enhances user interaction with this particular application it also makes it easier for people to read out rather than looking at lots and lots of numbers on your table so I'm just going to apply it and click on OK so for this to show I'm going to open a new sheet so I'm going to select this crime indicator and the price and this table is not over Justice I'm going to make it a pie chart and to make these colors more indicative of the crime in a certain area I am going to change for hi I'm going to keep it red Barlow I'm going to keep it green and for moderate let's keep it yellow let's turn them into a complete traffic signal I'm going to apply it and as you can see you can see it in the crime indicator box in the right also going to bring the price filter out and show filter okay so now if I set up a range say 15 to 25 you can hover over it and see the crime indicator for different medium values moving on let's see if we can find out an equation between various factors and the price going on new sheet and equation and here I'm also going to bring in my price and I'm going to pick crime again and this will give me a scattered chart what I'm going to do is I'm going to go to analytics and and select a trend line under the model and my scatter chart looks more like logarithmic chart so I'm going to use that and you can see a when you hover over the logarithmic line you see an equation for crime and you also find the r squared value and the p-value the r squared value is 0.28 and the p-value is again less than 0.05 which is 5 which means that you can trust this equation this is a usable equation now if we switch let's see the crime foreign you can see the equation has changed the R square value has come down to 26. so now basically we are trying to predict the nox value but if I go ahead and swap the two here you can see how the increase in NO2 affects the price of a certain real estate if I go up here on the line and show you how the logarithm and if for a moment I change this into a linear model you can see a formula for your price of a certain estate which is minus 33.9161 times Knox plus your 41.34 which is nothing but an intercept with your r squared value at 0.18 this basically is how much your how much your Knox value is going to affect your pricing and then you have your P value which as I had explained before is the value you get from hypotheses testing now this worksheet is a great way through which you can find out how certain factors affect the pricing of real estate I have taken Boston for example you can pick any data set that you like you have your equation for predicting price based on the Knox value your r squared value is 0.18 which just means you can barely use this variable as a complete indicator for price you have your crime indicators you have your metrics for the five most important things people look at while determining the pricing of a certain real estate you have your crime rate versus building pricing in boxes and whiskers plots here you have your Dynamic measures affecting the price and this is your personal help page for all the variables all in all this is a pretty useful method to make decisions by creating key performance indicators and economic measures let me go back and check if I have um covered the entire case study and I have so to conclude we basically studied a data set which had multiple factors affecting the prices and the outcomes of pricing on multiple parameters in a certain area and we understood the correlation between parameters and its rates with that let's move on to our next case study until the third and final case is based off of a data set on IMDb here we are going to conduct statistical data analysis or exploratory data analysis on the data set and understand the relationship between different parameters of factors in the data set we have a couple of movie budget and revenue related questions genre related questions we also have a few correlation questions between the revenue generated by the movie versus its impact on social media so all in all this is going to be not a step-by-step kind of a solving like we did for the two previous case studies this is going to be focused more on the implementation side it's going to be slightly more fast-paced since we have already done like a step-by-step in the previous two case studies I would assume it's safe to be presumptuous about you guys being able to catch up so let's get started so here's my data set we have color or no color director name duration Facebook like names of all the actors as you can see all the movies have at least two or more genres separated by pipe separators so we're gonna right click and click on split so now basically all the genre has been split into three different columns but if you want more columns you can just right click on either of these and select custom split so I'm going to hide the genre so you have a few numbers from Facebook the IMDb link country of production budget so on and so forth so let's look at our first question is the relationship between movie budget and revenue so we want to start out with sheet1 so normally we would assume that a high budget movies generally make higher revenues but we can be proof totally wrong through this analysis so let's go ahead and check it out so let's go ahead and bring our gross versus budget and here as you can see again it's showing us only one point which means we go to analysis and we uncheck the aggregate measures but I still can't understand anything on the scatter plot because all of these points are so closely gelled in together and we have a couple of outliers we have one point which has a budget of I don't know 12 billion yeah it's 12 billion which I'm pretty sure might be a mistake on the typist side but maybe if I click on this and I go on detail then I see that the movie's title is the host and it gives all the other factors and parameters for the movie so what I'm gonna do is I am going to put up a filter for budget I'm going to show filter so I can drill it down to at least see some points if I bring it down to 9 billion we have these many points and if I go further down I can se