 We're going to end at 9. So we're going to cover chapter 1 and chapter 2, which is just a study unit. I must talk about study units because the chapters are from the book and your chapters from the books do not always going to align with your study unit number. So I'm going to use your study guide as a guide. So study unit 1, which covers basic statistics concepts. And also we're going to do study unit 2, which is data visualization. We will see how far we get in terms of that because it's too much work. And if we don't finish the data visualization, so on Saturday we will start with that and then we do a lot of other exercises. We will see how far we get. So I'm not going to do a lot of introduction because I hope you already went on to my UNISA. You saw who I am. I introduced myself also on the forum and I also introduced myself when we had the session I think last week. I introduced who I am. There is so much information that you can find out about me online everywhere else. So if you want to know who I am, I'm not also going to ask you to introduce yourselves. So I will also check on the chat every now and then to see who is responding, I might call out your name to get clarity on some of the things that you post there. Other than that, welcome and let's start with statistics. I hope you're going to enjoy the sessions and always come back week in, week out. So let's start with statistics and I think it doesn't want to move. OK, so by the end of the session, you should be able to learn what is statistics, what are the key concepts of statistics, understand the types of variables and know the different levels of measurement, sometimes called the scales of measurement. I must also say something here before we even start. Do not start with your assignment as yet, but if you have already, there's no harm with it because I'm here to help you complete your assignment but not do it for you, but help you in terms of making sure that you understand the way before you complete your assignment. I think when you do your assignment as well, you get three chances. Your assignment questions are different. I will elaborate this again on Saturday when we do the activities and exercises so that you understand how your assignments are structured. So they are different, they are shuffled around so every student will not receive the same question. So the question that you receive might be different to the next person. So when we do assignment questions, some of them they might be one of yours, but they look exactly the same. It's just there and there where there are changes. And not every student will receive the same type of question, but the structure of the questions are exactly the same. So it means if someone will get a question that gives them 100, the other one will get a question that gives them 300. So it's like that one will get a question that says Tom, the other one might get a question that says Pira, but all of them might be asking the same question, the same thing. OK, so let's get onto it. Assignment questions, you can ask them on Saturday on Wednesdays. This is the last thing that I'm going to do also now. On Wednesday, we discussed the content. So anything relating to content, what you didn't understand when you were studying, if you want me to elaborate more, you ask it on Wednesday. On Saturday, we do lots of activities, then you can weigh in your assignment questions as well. I think that introduction is meant for all the other sessions to come so that I don't have to repeat myself again. So tonight, by the end of the session, you should be able to learn what is statistics, the key concepts within statistics, understand the types of variables, and know the different levels of measurement. So what is statistic? I am a statistician. So with statistics, you get data from different sources. It can be from surveys. It can be from the systems, the CRM systems. It can be things that you wrote down. So different sources. It can be from Excel sheets that you have, or web document, or PDF. As long as it's got data or information on there, that is a data source. We take that data source, we transform it, make some calculations to it, calculate things like the mean, the average, and all that. And once we have enriched it with those calculations and pivoting it and doing a lot of summaries and putting it into context, then we present it. When we present it, we present it either as a graph or as a table or as pictures. These days, we use what we call infographics. And once we do that, then we give it to people who make decisions and they look at that information and they make decisions out of it. And that is basically what statistics is all about. Statistics is a method of transforming your data into meaningful, useful information for decision making. That's what statistics is. Why do we study statistics? Since there are so many things happening around us, we need to make sure that we are well informed. And these days, we talk about evidence-based decisions. We always want to see the evidence before we make the decision. For example, like we use statistics to develop an appreciation of variability, meaning the changes that are happening, how those changes affect how we develop our products or how we process things through the system and so forth and how we build the systems as well. Statistics also helps us in estimating the new values, the present so we can use the historical information to estimate what's going to happen tomorrow. There is someone who is not muted. I hear all the time. Clangy, your hand is up. When I do the presentation, sometimes I don't see the hands because I'm cooking. Clangy, is it a historical hand or something you have a question? Okay, no answer. We're going to mute. Okay, so we use historical data to estimate the present and predict what the future will look like. We use statistics as well to understand the basic things about the statistical reliability, whether the data that we are using is it going to make meaningful decisions or are we going to seek to make meaningful decisions as well? And also to check the stochastic processes, meaning to check if there are any of the fluctuations or any of the probabilities that we can calculate from there. We also use statistics in most of the sectors, even nowadays in the business, you use most of the businesses use statistics. For example, those who work in insurance companies, the people who manages the portfolios of insurances, they use the statistics measures, like measures of variability to check whether the portfolio that they're looking at is it going to be a risky one or not? Or can they save your clients or customers' money by looking at that? And if you are in the product development, they will use statistics to check if that product that they built, is it, do customers like it? Are they buying it? What is the average over time? They can also use it to predict how long will that product be on the market before people stop using it? Things like that. So many businesses are using statistics. Also the government is using statistics as well, like we also know that we have statistics in South Africa where they do surveys and they analyze the data and they present all those. And we also see that they also sometimes, the police, also the police minister, also uses the crime statistics, they report on that, the accident statistics, they report on those things. So statistics is everywhere. We use it on a daily basis as well. So we use it to solve problems in a general thing. So we take statistics and we use it to solve many of our problems that we have. So some of the interesting things that we use statistics in is with the COVID now. You saw that in the beginning, when we started with the COVID, there were a lot of models that people were building statistical models and even though it was in the area of medicine. So in medicine, there's what we call biostatistics. So it uses the medical information or medical data to make decisions out of it. So we call those the biostatistics. So they use that, or the epidemiologist, they use that data to make decisions based on diseases that are happening. They were tracking the COVID. We could see that it was exponentially increasing and so forth. So, and that is why statistics is very important in that regard. With the political campaigns, we always see on ENCA, they're always closer to the voting. You will see that they will be analyzing the statistics, like the political campaigns and saying this party will win, this party is gaining majority, this one is doing that. And those summaries that they make, it's part of what statistics is all about. Even though it is in the broadcasting advertising arena or area, but they are able to use the information to make decision to also inform the public in terms of what people are saying about things. And also statistics. So those who, yes, Tami? There is a hand from Etienne. Please check. Okay. Etienne? Sorry, you were on such a roll. I didn't want to stop you. Next time stop me if I'm moving too fast, because then I might, I'm not constantly looking at my phone, actually. The question is around the statement you made, Rob. Go ahead. Go ahead. Go ahead. Go ahead. Go ahead. Go ahead. Go ahead. The question is around the statement you made regarding statistics for like medical and political campaigns and things like that. How do they know which is the right model to use? I have the feeling that they have, even within these particular organizations, they have conflicting models that has the correct predictions. How do they know which is the right one to use? That's the thing. With statistics, you won't know which model, which is the right model. So we always check the efficacy of the model. We check the, there are measures that you use to test whether the model that you've built is a good model can fit the data well. So for example, let's say you wanted to predict, you want to predict whether there will be the third wave. So in, in biostats, they have their own type of models that they built, but all of them, they are based on the statistical capabilities, like the statistical models. So whether they build it as a regression model or they build it as a neural network or they build it as a decision tree or, so they are different models or they use Bayes' names, they use forecasting. So they are different. So it will also depend on the kind of data that you are using. And once you have built that model, in statistics, what we also prefer as well is to build as many models as possible using different techniques and different algorithm and then use the measures that you get from there. But it's a discussion for another day. So we use those measures like your MSE, which are your mean square errors. We use things like your SSE, which is your sum square errors. We use things like the misclassification to check whether every model that you built, which one has the LESA classification errors, which means it does not predict wrong things. It does not say one will fail, whereas the one will pass, whereas that person has passed. So there are different techniques that we use. So you will never know which one, unless if you put all the models together and use those measures. And different models produce different measures. So you need to put them like, evaluate each one separately and then look at them and see which one best fit the data. So it's fascinating. So you should love statistics and you should be looking forward to learning more about it. It's not a perfect science. Yes, no. That is why they say statistics can tell a lot of lies. That's why it works for government mostly. Yes. Thank you. You've answered my question. Thank you. All right. So with statistics, we do have two branches of statistics. So we have what we call the descriptive statistics and that is what the majority of the people are using. Like you use it on a daily basis. You talk about on average, I spend 20 rent. Those are just descriptive. You are describing how much you are spending on monthly, on petrol or something like that. And then descriptive statistics, sometimes like most of the time when they report on the crime statistics, they just tell you that from Easter holidays, gender-based violent cases have increased. In December, there were a lot of matters. So those are just descriptions because they just give you the percentages and this and that. And that is what we call descriptive statistics. So with descriptive statistics, you just collect the data and you summarize the data. So collecting the data, you can collect it via the surveys or you can use your CRM tables. Like when you go to a shop, let's say you go to shop right and you buy grocery and you buy things. The information gets taught on their database and they can take that and analyze it and do some manipulation of the data and analyze it and calculate them in the mode and all that. So collecting that information, it's part of a descriptive statistics. And then we summarize it and we visualize it in terms of tables or charts or graphs. And these days we use what we call infographics for nice pictures that tells one story. And we can also analyze it by manipulating the data and calculating things like your mean, your standard deviation and so forth and the mode and the median and so forth and so forth. So we also have on the other side. So with descriptive statistics, we are just describing the data. There is nothing other than describing what the data looks like. Inferential statistics on the other hand, yeah, we infer information. So like when Etienne was asking, there are a lot of models that people are building and which one can we trust? That is inferential statistics. That's where you take the data that you have, you sample it out and then you create a model. A statistical model or econometrical model or biostatistic model. You create that and based on the results, you can then, if depending also in terms of how you selected your sample, because then sometimes there is, you can select a sample, but your sample may not be a representative of your population. And I will discuss in more detail what I'm talking about right now. But for now bear with me. So you create a sample and you find that that sample is not the representative of your population. Then when you build your models, you cannot infer the results of your sample analysis to your population. You cannot say, because your results of the population say South African people are obese from that sample, you cannot say, yes, sorry, from the sample that you calculated, you cannot say South African people are obese if your sample was not a representative. So in terms of 1610, you do not need to know what I'm talking about by sampling the data. You only need to know the definition of what is a sample and what is a population. So if you continue with stats and you want to learn more, then we can talk about how we do the sampling from taking the sample from the population in order for us to make sure that that sample is a representative so that the results we get from using the sample, we can infer it back into the population and say the population looks like this. Okay, so with inferential statistics, we can do estimation where we estimate the population mean using the sample mean. Or we can also do a hypothesis testing for this estimation. We will do it in Chapter 8 when we do confidence in Tava and hypothesis testing. We will do it all throughout from when we do hypothesis testing and when we do the chi-square testing and all that. So all of them we will be applying hypothesis testing. So with hypothesis testing, we will be testing a claim that the research I make about the population with the sample data that we have. So, Miss Lizzie? Yes. Sorry, it's incorrect to say hypothesis testing is far richer than an estimation in statistics. Not necessarily. Remember the hypothesis, you're testing a claim, your estimation, you are estimating if the value falls within. So it's a different way. Yes. So you are estimating a new value and you can use hypothesis testing to test whether that estimation falls within that and make your decision based on that. Because if it falls within the lower boundary and the upper boundary, then you can say that the population mean is within. But if it's outside, then you can say that it's not within and you can make your decision to say that you can exclude or you do not have to trust the estimation and you cannot use the information. With hypothesis testing, you already have a claim. You know what you want to prove. So you say, South African people are obese. You need to go and prove that claim. And so that is the difference between the two. So with the estimation, you only want to test whether that sample mean is within the lower boundary or upper boundary and it fit falls within that. Then you can say the population mean is within as well. And you can rely on that data and use it to estimate any information and you can use that to estimate a new value. Hypothesis testing on the other hand, you have a claim. You want to prove it otherwise. So it's what we call the ease. You are guilty until proven innocent or something like that. You are innocent until proven guilty. That's the law thing. But there are two sides of a coin. There is a head or a tail and I want to prove that that coin always lands on a head. You have to prove it by testing that coin so many times and then collecting that information and summarizing it and say but that coin is not a fake coin. Things like that. Okay, so with inferential statistics, we draw conclusion about the population based on the sample that we created. Okay, so since I've already introduced the population and the sample, let me explain what those are. So a population is a set of all elements or all subjects that you are interested in studying. So South Africa is the population. The whole of South Africa. If I want to study the universe then the universe is my population. If I want to study African countries all African countries all African countries are my population. So it's elements that you are interested in studying. All of them. Sometimes when we do studies the population is too big that you cannot reach each and every person, even including with the census. You know that most of the time we say the census we count every person in South Africa. It's impossible to count everyone because you move around the time that the people come in you are not here so they won't count you. That is the census. That is why they call it the census. So to count the population of the countries. But population is everybody who's in South Africa. If statistic South Africa comes back after they did the census and they estimated and they said in South Africa we've got 59 million people therefore that is our population. Because the population is too big and we cannot use everything that we collect from the population or we cannot reach some of the population for a study. We then opt to create what we call a sample or before I move on to the sample. So when we create a population when you have defined your population and you start calculating you doing your descriptive statistics there and you calculate your median, your median, your mode, your standard deviation your population proportions and all that. Those measures that you create from the data that you collected from the population they are what we call parameters. So the minute you calculate the mean you are creating a parameter. So if you go and calculate so you go and later on when we do the descriptive statistics in study unit 3 in study unit 3 you will learn about the different parameters or measures but the parameters are like your mean your median your standard deviation your mode all of them they make parameters. So the minute you calculate your population measures then you are using or you're creating a parameter then because the population is too big we create what we call a sample. So for example like I said in this module you do not need to know how we do the sample but you just need to know that in your population we can select different people to make up a sample a new a base that we can use for doing the descriptive analysis or doing inferential statistics and that is what the sample is it's a subset of your population so we select different people so if we wanted our sample to have only five people so we can say that it becomes part of our sample and then we can calculate the measures from there from the sample and those measures from the sample when we calculate like the mean, the median, the standard deviation the mode and so forth that we calculate from the sample are called statistics and that is the reason why statistics is a terminology used because most of the things that we do in statistics we use the sample to infer what the population would look like but the measures we get from here from the sample we call them statistics or a statistic if it's the mean we just say it's a statistic not statistics and if there are so many we say there are statistics because there are measures like the mean, the median the standard deviation, the mode coefficient of variation and so forth okay now exercise I've been talking for almost 40 minutes in a hospital seven randomly selected patients have a blood type O, A, B, B, A, O, O and A from the information that I just read identify what is the population and identify what is the sample you can type in the chat if you are able to so you can use the chat to type and then I'll give you two minutes to think about your answer in a hospital seven randomly selected patients have the blood type O, A, B, B, A, O, O and A what is your population of study and what is your sample you can type in the chat let me see your chat is disabled we can't type there the chat is disabled okay I don't know why okay so you are not able to type in the chat with me okay think about it and then just now I will check why the chat is disabled I also cannot type in the chat okay so since I cannot type also in the chat you maybe want one of us to try and wing the answer over the conference over the yes I will give you a chance for someone to say okay so anyone you can unmute maybe if I raised it let me wing it yes you can wing it so I'm going to say the population is the blood types and the sample are the seven randomly selected patients I could be wrong okay anyone any take come guys back me up don't leave me alone okay the population is all the patients in the hospital and then the sample is the seven randomly selected patients yes so their population is all patients all patients in the hospital your sample is a subset of the patient which are only the seven randomly selected patients remember your population is your population is the element of interest blood type is not is not what you are studying you are studying patients from the hospital and from that study you wanted to look at their blood their blood type so it means your population of study will be all the patients and your sample will be your seven randomly selected so as long as you have this thing randomly selected it means they have sampled out from a bigger pool okay I think if they said in a hospital of 500 patients then they would have made because in our world if you get this in an exam question they will tell you in a hospital of 700 patients seven randomly then it makes more sense sorry so I understand now yeah okay so from this they also we want to study those seven patients but we we are interested in their blood type so blood type and the O's and the B's and the A's are what we call the variable and the O's and the B blood type is a variable and the O and the B is the data that comes from there okay why is a variable like I said blood type is your variable so a variable is the characteristics that define the population or a sample and that characteristics you can observe it or you can measure it like for example uh this days you don't want to say things wrong actually so I was going to say I'm going to go old tradition way here yeah so those who are um uh LGBTQ family please bear with me I'm going to apologize now while in advance I'm going to be a stereotype person here so I'm going to give you a an example that everybody can understand without insulting anybody so let's use gender I know that we no longer use gender we use sex orientations and all that but for now let's use gender so a variable let's say gender you can observe a person by looking at the person but these days we can no longer do that but previously in the old age you will observe a person and you will say that is a female and you will observe another person and say that is a male that is a variable so by just observing it or you look at the person and you cannot just say that person is 1.67 tall you need to take a measuring tape and measure the person's height and you will say that person is 1.67 and that is how you measure so the variable is something that you can observe or you can measure from the variable let's say I'm using gender as my example from my variable I can the minute I say when I observe that variable and I see there is a male or female then I'm talking about the data within that variable the data that described that variable are either a male or a female or 1.67 tall or 1.57 in terms of your height the data is just a set of individual values associated with your variable so a variable describes the population or the sample characteristics so like for example you get a blue pen a black pen and a red pen in terms of their color you get different type different models of a pen so those things are just the data a variable is the color of a pen the measure will be blue red and black which are the actual values associated with that variable color of a pen so we have two different types of variables we have the first one is what we call a categorical variable which is a qualitative variable it's variable that you can observe like your medical data your political party and these are variables that have a defined categories within them and you can observe them then we also have what we call numerical data or variables which we also call the quantitative data or variables because the data will be the values that comes from them so we have numerical or quantitative quantitative variables are variables that either you can measure or you can count and it's a measure that you can count it's what we call a discrete variable if it's a measure that you can measure it's what we call a continuous variable so if it has a decimal let's put it this way if it has a decimal it can be a continuous if it's a whole number if it takes a whole number then it can be a discrete because discrete we count them there are variables that are not going to be discrete even though we count them because of the manner in which we refer to them they in a normal environment for example age in a normal environment we would have said I am 21 years old I am 30 years old and in a normal environment we would have said age is discrete because it's a whole number we use it in a whole number age is continuous you must also know that age is continuous money is continuous even though we can count it but it's continuous because it's money it's in random sense that is why we have .00 at the end you need to always remember that so discrete variables are variables that you can count number of children that you have defects per hour like in a factory when you count how many defects the machine drops out or spills out you can count those because the items are individual items you can count how many defects are produced within by that machine or in that factory continuous things that we measure like weight, height, voltage temperature and so forth don't worry I am going to ask you to do an exercise just now so remember that categorical data data that you can observe and put into categories numerical data is data that you can either count when you count it it's called discrete variable quantitative discrete or numerical numerical discrete data that can be measured it's called continuous quantitative continuous variable your exercise from this earlier statement that we used A O A B A O A identify what is a variable identify what is a data identify whether the variable is numeric or categorical which means is a quantitative is it a quantitative or is it a quality tative you have one minute to think about the answer and and then I am going to call out anyone can answer the question are you done let's see anyone identify what is a variable in a hospital seven randomly selected patients have a following blood type O A B B A O O A what is a variable I have blood types blood types is your variable because blood type is a characteristic that defines those seven patients what is a data seven patients which one is the data these seven patients nope which one is the data remember data is the values associated with the blood types so which one are the values your data is the blood types are different blood types so the O the A B the B the A and so on these are your data okay identify whether the variable is a numeric or is categorical is the blood type numeric or categorical categorical can I measure them can I count them can I put them into categories categories can I measure can I count them or can I put them into categories we can put them into categories I can put them into categories so it means it's a categorical data and those are the types of variables that we have now since we understand the types of variables there are also what we call the levels of measurement or the scales of measurement not the units don't get me confused with the units like the kilometers and the meters and all that we're talking about the levels of these variables like your quantitative variables and qualitative variables they've got levels within them I'm not going to ask you one exercise we're just going to go straight into the levels of measurement so levels of measurement defines the highest order in terms of the variables that you have so the lowest the lowest levels of measurement are from your categorical because there is not much that you can do with them we just put them into categories and count how many they are within those categories then there is what we call the nominal order which I will explain later on what nominal nominal scale of measurement is so nominal scale of measurement categories have no order so like male and female will go into details in terms of that whereas with the ratio scale of measurement it is at the highest level it means it's got the strongest measure of is the strongest levels of measurement that you can get to because with that you are able to calculate things like the ratio, the difference between two distance and so forth so that is why it has the highest level with nominal scale it is the lowest because there's not much effort placed within it okay so let's dig deep into understanding each and every one of them so levels of measurement for nominal so for categorical data there are two so it's nominal and ordinal so with nominal there's no natural or logical order like meritorious data there is no order there's not when you think of nominal think about it as none of the the data within that variable none of them has a priority or superiority over the other there are just categories like yes and no yes is not superior to a no it's just that someone doesn't prefer this someone prefers this that's that you cannot also use it in a comparison manner where you do calculations but you can compare whether in terms of the values to say people who answered the yes questions were more than those who answered the no question but you can never say a is better than b it's not like that it's what the people prefer because of the levels or how they answered the question so you cannot do any comparison based on that actual value and I just gave examples in terms of the types of nominal variables that you can get like political affiliation it's not like because everybody is voting for the a and c and they win it means a and c superior than the other it means people just prefers to vote a and c than the other political parties so there is no one that is above the other race as well there is no one that is above the other things like that then we get what we call the nominal measures of scales of measurement or levels of measurement ordinal also for categorical data here you do have a natural or logical order for example when you rate a service let's say for example after the class I give you a survey and I say rate the level of service that you got from me today as a tutor you are going and I give you a scale and I say 0 means low and 5 means high so it means I did bet 5 would mean I am doing well so you are going to rate me according to that there is an order there is the highest and there is the lowest so there is an order in terms of that scale also with ordinal like with nominal you cannot use it in any calculation but you can use it to compare because then here you can compare how people have answered because here comparison can be made because then here you can make it you can deduce from the information you got that the service was poor the service was good and you can compare you can compare different measures or different levels with one another and because the data will be ordered as well so there is the highest value and the lowest value for example like the shoe size you can compare and say the majority of people are on average then you can do those you cannot do calculations but you can say most people prefer shoe size 5 or most people prefer shoe size 10 or something like that because they are levels within the shoe size sizes so you get from for the adult shoes you get from size 1, size 2, size 3, size 4 size 5, size 6, 7, 8, 9, 10 the bigger the size the bigger the number of the shoe size also with education level you start with primary secondary high school university or technical college things like that even within the university you start with a certificate a diploma a degree honors, masters doctorate there are levels so you move with the levels and that is ordinary then we also have what we call an interval interval and the ratio are from quantitative side of the variable so with interval the data can be in an audit scale and you are able to calculate the difference between two points and it will those difference will have meaningful information as well with interval there is no true zero point what I mean by that is with an interval it will not have a true zero point because let's say temperature let's say in South Africa we never go into the negative side unless if you stay in Sutherland I think in Sutherland they do get to the negative side of things and I'm going to say the truth probably since these days there is a snow that falls in that mountain area so they can go into the negatives I'm not sure if in South Africa have we ever been into the negative side of the temperature but temperature goes into the negative so it means zero has a meaning it's just sorry there's no meaning in terms of that it's just another temperature so it's zero degrees it's cold but there is another cold temperature which is minus some number minus four degrees minus eleven degrees it's cold it's way too colder than zero degrees so that is why it is called an interval it does not have a true zero point because zero for that is just another temperature but since it's got negative numbers you cannot do a ratio you cannot do a ratio of a negative number so a a ratio of two numbers cannot be well defined if there is a negative number you can do a ratio if all the values are positive and I just went into that example temperature is one of them bank balance bank balance can go into a minus so it can also become an interval so your bank balance can become an interval because if you are left with 25 franc in your in your bank account and then the bank charges you services and interest and all that your bank balance goes into a deficit into a minus and then that turns it into an interval scale of measurement for that category for that numerical data that you have and the last level which is at the highest order is your ratio a ratio similar to interval is an ordered scale and you are able to calculate the difference between the measurements and it has a true zero point if you have zero it means either you do not exist or that thing does not exist or has never been born or it's finished so if you have zero stock it means you don't have stock stock is finished if you are born and I don't know if it's zero born there is nobody with an age of zero so it means that person has never been born things like that you can think of more examples where there is a true meaning of zero meaning zero means something which zero means something that says that thing does not exist or is not in existence with the ratio you can do a comparison because like distance you can do your ratio of your distance from A to B you can calculate what is the ratio from moving from A to B and so forth I spoke about the fixed zero point meaning that thing does not exist like your weight if you weigh zero it means you are finished you don't you are like a feather a zero point but it's got a number like your height if the height of a building is zero it means that building was never been built distance if I move zero it means I have never even moved so zero means something so you must think about it as such that zero has a meaning now based on what I have just said you know me now for categorica there is no order ordinary categorica there is an order in which things happen highest to lowest interval there is no true meaning of zero ratio there is true meaning of zero it means nothing okay the weight of what amelons installed here you can just tell me you don't have to we don't have to to weight we can do one statement at a time there are about five statements what is the weight the ratio the weight is a ratio because weight means nothing it does not exist times of a day morning afternoon evening and night you must think very carefully you must think very carefully is it a categorica or is it a numeric that's that there if you were only looking at time it would have made it a numeric but because they say times of the day and they give you values within it which means it's a categorica it's categorica now you must think very hard and long about this one is this nominal is there an order it's ordinal ordinal there is an order of how things happen so there is morning then come afternoon then come the evening and then come the night so this will be ordinal distance from your place to the five nearest stores distance it's ratio because if you moved to your place it means you haven't moved to any of those five grocery stores airline companies serving at a given airport think about the airline companies at your South African airport AXA airport like OR Tumbles and Cape Town and Lanceria the ones in I can't even remember the name of the airport what airline companies are there and once you thought about it will that mean is it nominal or ordinal nominal it will be nominal because there is no natural order or logic last one the places in ranking of chess players first second third and fourth ordinal because it's ranking in first second third and fourth which are categories then it is ordinal before we move on I need to also mention this so sometimes ordinal and nominal values can be represented as numbers but they mean they will always refer to categorical information or variable that they are they will just represent the actual wins in numerical form but they will represent an ordinal or numerical for example they might say rank the levels of service with a scale of one to five so this scale of one to five it means nothing because they might say one meaning low and five meaning high so you can use one two three and four but they refer to certain categories within that so we can use numerical for nominal or ordinal but they will mean they will just be placeholders for the categories must always remember that so in terms of the levels of measurement what can you do on those when we do chapter three you will understand that we are going to mainly use numerical information to summarize the data in terms of calculating the mean, the median, the mode and the standard deviation we are going to use only numerical data but in terms of the levels of measurement and the variables we can count and we can create a mode based on nominal information also with ordinal information you can order that you can create counts with it you can the mode because the highest one will be the mode you can create the median because the middle one will be your the middle one will be your median you cannot calculate the mean because you cannot summarize all the values like one, two, three, four, five they are categories so you cannot add them together you cannot calculate do the ratio calculations on some of them the actual ratio calculation so you will not be able to calculate what we call the absolute zero with nominal or ordinal because they are just categories so in terms of intervals you can do everything except calculating the ratio because with interval remember it can go into the negative you cannot do a ratio of a negative number you cannot have an absolute zero because it can also go into the negative venues your venues will the absolute yes because it can go into the negative and the ratio you can apply all of them all the order of operations like count, mode, median mean difference adding and subtracting the ratio and multiplying and dividing and there is an absolute zero because the ratio has a meaning of zero which means zero means nothing okay so that is the levels of measurement so now any question before we move on to study unit two how are you doing so far are you lost are you seeing the light at the end of the tunnel how are you feeling have I clarified some of the things that you are not sure of so far with regards to study unit one or we still need more work to do that so far so far