 Hello and welcome to the session in which we would look at the topic of data visualization. This topic is covered on the CPA exam and it's going to be covered more and more whether you are taken BEC or auditing. Data visualization is important and I'm pretty sure down the road it will be integrated with drag not yet but you will bet I'll bet with you it will be integrated with drag as well because most accounting courses as well as FAR because most accounting courses these days they always add they're starting to add maybe an appendix or a full chapter for data visualization this means if you're a future accountant future CPA this is the future also if you are a business students this is the future but from my perspective what I care about if you are a CPA candidate this is what I need to help you pass the exam farhatlectures.com no I don't replace your CPA review course you keep this I'm a useful addition to your CPA review course I'm going to explain this topic differently approach it from approach it from a different perspective by doing so add 10 to 15 points to your CPA exam score by helping you understand your review better you will do better on the CPA exam your risk is one month of subscription give me a try I might be able to help you you will find out I have helped hundreds if not thousands of students pass the exam that's your risk your potential gain is passing the exam is that risk worth it one month of subscription and if not for anything take a look at my website to find out how well or not well your university doing on the CPA exam I do have resources for other accounting finance audit courses as well if you haven't connected with me on LinkedIn please do so and take a look at my LinkedIn recommendation like this recording share it with other connect with me on Instagram Facebook Twitter and connect with me follow me on Reddit especially Reddit so let's go ahead and talk about data visualization why data visualization is important now it was always important but because of technology we are being as human we are becoming more creative and as human we are visually biased we enjoy looking at things we have a saying that says a picture is worth a thousand words so it's easier for us to look at pictures and absorb information rather than rather than read tax and this is why for example marketing people would use infographics to present their product to present what they have to offer it's easier for the consumer to look at this product but what is data visualization it's basically presenting data in a pictorial or a graphical format and what's the purpose why do we do that again because we human are visually biased not like dogs dogs like the smell so if we are dealing with dogs we give them something that smells and it will appeal to them but we are we like we like to look at really nice especially if it's nice okay but the purpose is to help decision makers business maker use analytics presented visually so what so the users that business decision can grasp and understand difficult concept or identify new pattern in the data so it's easier when you graph it you can see the pattern in the data and although you might be good at data analytics but if you cannot communicate this information in a way that's understandable it's usually not as effective so that's why even though data analytics is a powerful tool because it's you're using the computer power to compute data but you still have to communicate and this is where data visualization comes into place how do you translate this information to the users and by the way this is an art not a science so keep that in mind when we are dealing with data visualization we need to know what is the purpose of the data visualization we have two purpose and we have to understand the type of data that you are dealing with because depending on the type we will determine what's the best way to present this information we have two purpose for the data visualization there's a declarative purpose which is explaining the results of previously done analysis and usually if you're an entry level if you are not a programmer if you're an entry level staff accountant and if you did not specialize in data analytics you'll be expected to do some explaining taken the data declaring something from the data so the audience you're presenting the finding the results to an audience versus exploratory data here you are exploring the data via visualization here you don't know what you have in declarative you already find found something and you're showing it here you're just visualizing the data to gain an insight to somehow pick up something that's not read that you did not see from the raw data okay so gaining insight while you're interacting with the data you don't know initially what you are looking for you're hoping to find something this is called exploratory you're exploring sometimes it's called data mining data discovery knowledge discovery whatever you want to call it this is why this is what the purpose of data visualization and we have basically two type of data qualitative data or data driven and qualitative which is also concept called conceptual now we're going to talk about each one of them separately quantitative quantitative and qualitative data starting with qualitative data those data are also called categorical data categories categorical right categories you can count them you can group them and in some cases you can rank them what are we talking about here categorical data also can be broken down into further they can be broken down into further categories so categorical data can be broken down into two categories one is nominal data what is nominal data this is the simplest type of data we're talking about hair color gender ethnic groups age similar things you cannot rank or average them forget about age age you can rank but usually those things you cannot rank or average them for example you cannot say you know a male better than female by gender or a hair color the black black hair color is better than the blonde so on and so forth you cannot you cannot rank them or one ethnic group is better than the other or you cannot find the average hair color you cannot mix all the hair color and find the average hair color for a group okay this is nominal data ordinal data the other type of qualitative data you can count and categorize them but you can also rank them here we're talking about for example when in the olympics they win the medal gold silver and bronze one two and three you are ranking them you can count them categorize them and also rank them or for example when you rank your professors at the end of the semester rating from one to five you scale your teacher letter grades is a form of ranking letter not not not a numerical you can count categorize and rank like a b c d you can rank them but you cannot find the average okay or the standard deviation or any other statistical method now bear in mind you can run something called proportion on these type of data for example you can find out the number of people with blonde hair as part of the total let's assume the total is 100 we would say that 20 of the people have blonde hair in this group so qualitative data whether it's nominal or ordinal can also be referred to as conceptual data because such data are tax-driven and represent concept instead of number so you have to keep that in mind this is what quality qualitative data it's not number we're not talking with quantity and quantity usually imply numbers this data is more complex it's just more than just counting things it usually it's made up of observations counted and ranked okay just like or ordinal qualitative data you can rank them you can count them but here you can compute averages standard deviation any statistical number you would like to also they can be categorized into two different groups and this is there's you know some discussion whether those two groups are are needed or not one is called ratio which have a zero a zero number that means zero an absence of something so when we say zero dollar it means you have no money zero dollar it means there is nothing the other group the ratios is intervals here zero means something for example when we said zero Fahrenheit or zero Celsius as the temperature it means something it doesn't mean there's no temperature it means it's freezing we're getting to the freezing temperature so that's those are the two subgroups under quantitative data okay it could also be it can be further categorized into whether it's discrete or continuous and what is discrete discrete when the numbers represent whole number when the figure represent whole number for example points in a basketball game it cannot be 143.5 versus 133 right it's 143 versus 133 the score or a soccer game you know one team one two to two to zero so it's just you know you cannot be one and a half right it's those are discrete continuous data continuous data as data that can take any any value within a range for example the height for 1.7679 centimeter okay so continuous and you're going to see that the continuous would would use a line chart now what are some good charts for qualitative data qualitative data we can start with a bar chart and I hope you are familiar with a bar chart bar chart is a graph or a chart that represent categorical data with rectangular bars with heights and or length proportional to the value they represent the bars can be plotted vertically or horizontally a vertical bar is called something like a vertical bar is something called column chart and I'm pretty sure you want to see one this is what a bar chart would look like asking the children what's their favorite color and it seems yellow is the most popular color my son like green but happens to be yellow here so this is what a bar chart looks like pretty simple this is one way to show to show qualitative data bar chart okay we could also represent information in a pie chart think about a pizza pie the pizza pie would look like a circle and you will cut it down usually into eight pieces eight equal pieces now what we do size of each slice is proportional to the data at represent for example this is a pie chart now the pie chart a lot of people don't like the pie chart for example here you cannot differentiate between a lot between the blue and the yellow are they equal to each other is the blue a little bit bigger than the yellow however if you if you if you if you put them in a in a bar chart it looks much better you would know what's going on much much more easily on a pie chart now the pie chart if you add the percentages to them then it's easier then you would know that this is 12% and this is 11% this is 26 and the orange is 28 28 versus 26 so it's much easier it's much easier to read okay you can also use what we called stag bar chart again depending on what you are looking at you could use the numbers you know or you could use poor percentages but those are charts that are usually good for qualitative data we can also use other charts to show the proportion and qualitative data such as tree maps and heat maps symbol maps word clouds which is sometimes it's called sentiment analysis what is tree map or tree mapping is a method for displaying hierarch hierarchical data using nested figure usually rectangles something that looks like this if you watch cmbc or follow the stock market this is the smp 500 and for example apple alone represent 0.23 of the smp 500 microsoft represent this much not 0.23 this is the change for that day but the rectangle tells you for example between microsoft and apple they represent a huge portion and google so notice and amazon those are the big names let's see where facebook is uh where's facebook maybe it wasn't as big when this uh when this chart was making but now if you look at facebook it's it's it was it's large okay so notice what it shows it shows you the proportion but in form of rectangular okay so you could easily identify the big pieces in any group another one is symbol maps it's basically using markers often a circle position on a map the marker is sized according to the quantitative value something that looks like this the location of walmart stores in the united states for example here that's the biggest circle it means 100 stores for example if we see here texas california florida they have 100 stores for example here we see less than you know small stores maybe five around five stores okay so it's easier to see the data like this clouds word clouds or sentiment analysis this is when you have text and it's a visual representation of words and what i did actually i did run a word clouds on on my youtube on my youtube on my youtube comments and basically great and explanation uh those were common words that they are used uh please use so on and so forth those are the repeat words smaller one so explanation is good it's basically great explanation i get this a lot so that's why it's showing great explanation i thought it it used to be thank you a long time ago i don't know what happened now people are just using you know signs images for thank you now let's talk about techniques that deal with quantitative data all the method that we mentioned for the qualitative data can also work to show quantitative data okay so they work for both of course but we have we can use more comp we can you we can we can have better charts for quantitative data one of them is line chart and i hope you are familiar with line chart show similar information to what a chart would show okay but line chart are good for showing data across time okay so it looks something like this okay and for example here it could be monthly okay or january february march and this could be sales okay we can see how sales is changing over time useful for continuous data bar chart is usually good for discrete data and line chart are not recommended for qualitative data if it's qualitative data you know you know if whether someone has uh blonde hair black hair it doesn't really matter okay which by nature are being categorical and never and never be continuous anyway so this is what we what we mean by line chart it shows you what's happening for example the change of sale over time box and whisker plot that's another data useful for when quartiles median and outliners are required for analysis and insight and i i i like box and whisker plots but you have to understand how it's work for example um let's take a look at this here here let's assume this is the uh the 9-1-1 weekly calls to a calling center week 1 10 calls week 2 17 week 3 5 week 4 32 week 5 16 18 and 20 so how do we build this box and whisker plot for this call center you know how many calls did they get 9-1-1 so we can staff it okay first we put them we would put them in order 5 10 16 17 18 20 and 22 okay and what we do first is we draw a line and first we're going to pick the extremes the upper extreme and the lower extreme we notice in one week we only had five calls so that's going to be the 5 the minimum this is the 5 here and in one week we had 32 calls those are the outliers or the extremes then we need to find out the median the median is the middle point well we count which one is the middle point one two there are seven of them and the middle point is 17 the middle so this is this is the 17 this is the median okay if we if we're drawing the line here this will be the median 17 this is the median 17 and 17 is the middle it cuts it cuts the data into the upper and lower quartile okay so notice what happened here 17 right here we have this is the upper and this is the lower okay now if we have even numbers because here we have only seven numbers what we do if we have even numbers like eight we'll take the two in the middle and we average them up to find the the median number now what we do is we find the median and the upper quartile which is 20 and this will be 20 this will be 20 and the median and the lower quartile which will be 10 and this will be the number 10 now we draw a box okay now what we say is this yes sometimes we do get five calls sometimes we do get 32 calls but those are extreme upper and lower extreme most of the stuff occur between 10 and yeah 10 and 20 10 20 so when we need the staff we need to kind of expect 10 to 20 so most of them 10 to 20 now this is 25 percent 25 percent 25 percent quarter okay quartile quarter so this is we break the data into four quartile to find out where the data falls around the middle another technique to show qualitative quantitative data is scattered plot it identified the correlation between two variables for identifying a trend line or or line of best fit the best way to kind of explain explain scattered plot let's assume you are studying for the CPA exam I'll have to say the more hours you put the higher is or the more likely you are going to pass the more likely you will get 75 or more so this is basically what we're doing is we're finding the relationship between a dependent and an independent variable so your score is dependent upon the time you study assuming you are study using valuable time the more time you study even for your even for your courses the more time you put into your studies usually the higher is your score but scattered plot is used in the real world and usually it's advertising and I'm pretty sure you saw this advertising versus sales and the more advertising you spend data shows the higher is your sales and usually what marketing people they will show you this on a graph this is a scattered plot and they usually they would run a regression and you will have r to find out what's the to measure the the fitness of the line how close are these thoughts the line the closer they are the fittest is the line so this is 0.96 this is pretty fit filled geographic maps is another way to show quantitative data and this is a little bit different than the symbol maps filled the geographical map it's filled the colors are filled is used to illustrate data ranges for quantitative data and if you use tableau you can easily do this across different geographical areas such as state and countries and I'm going to actually show you my actual website data analytics for example here the blue are countries where I where my viewers are notice the dark blue in the US it means most of the viewers are in the US and I'm not sure if you can see this this is India and my next country is India like in terms of viewers it's a little bit lighter darker darker than the rest darker well not sure if you could see it on your screen or not and the rest Australia South Africa the Philippines some countries in Europe I have no one practically no one in Russia this is Russia this is why it's like blank okay um so the darker the blue the more users I have and obviously I have the numbers for example how many viewers I have in the US versus other countries and it shows me it showed me in a filled geographical area the quantitative aspect of it now what you need to know about what else you need to know about the data you need to be aware of a few things how to scale the data and what increments to use we don't we're not going to talk about the increment because the software you're using usually they would recommend a good increment but how much data do you need to share in visual uh individual to avoid misleading so you have to be very careful not to show too much or you're misleading or or not to omit stuff because you want to hide something so this is one of the questions that when you're presenting data how much should I include should I include two years versus four years so I only show the four quarters the key is not to hide any data that doesn't work for your expectation so for example if the last four quarters are good and the prior three years were not you would only show the first the last four quarters and what you do is you'll hide the other data that's not good that's not how you show data for example if you show it yearly it might it might look better than quarterly or you know a period of two years is better than four years because you know the first two years we were not doing good so how much data to include this is this is basically an ethical questions a professional you have to make a professional judgment how to deal with outliers when you have for example in one year not a lot of sales or a lot of sales or good one point that's really outlier out there should you include it should should it be displayed because if the purpose is to draw attention to it if it's important then you want to include it or if it's just noise like it's not really relevant because we can explain it so don't include it in the data because the data would look bad it's not that doesn't look bad it does it doesn't represent what we need to represent you keep it out it's considered noise so you have to be make a decision about the outlier and usually when you emit the outlier somehow you have to let them let the users know there was an outlier it was emitted for such and such reason also the scale you know the chart should begin usually with the baseline of zero now if zero is meaningless for the data you could find a different baseline that makes sense but the point is don't find a baseline that either overemphasize something or under emphasize something over exaggerate or or try to minimize something so the baseline is important the scaling where do you start also you have to provide context or reference point to make any scale meaningful for example if you're showing the stock price of a hundred dollar it's not it's meaningless is this high price low price okay we need to know the context how is that stock price doing over time it was in the 500 and a drop to 100 that's not good it was in the in the 20 dollars and now it's up to 100 then you know it's going up so we have to show it in in the context also we have to show the company's industry and its competitor stock prices of that relevant how are we doing versus others okay if we if we have a this is you know if we have a 10 growth but the competitors have 20 growth well if we show it then it makes sense that we're not really doing a good job but if the competitors they only have a 5 growth then 10 is good so we have to show the growth and the stock price versus either the industry overall or a specific competitor or some other piece of context because data does not make any sense in void it has to be giving within a context and this is when you present also ratios they have to be giving within a context at the end of this recording i would like to remind you to visit farhat lectures dot com again i don't replace your cpr review course all what i'm asking you is to give me a try for one month i can help you pass the exam how i explain the material differently you would keep your course you would use me as an alternative explanation good luck study hard and most importantly stay safe