Video Transcript: Making Decisions Based on Statistical Data

Hello, welcome. We're going to discuss making decisions based on statistical  data. What is statistics? Statistics is the science dealing with the collection,  analysis, interpretation and presentation in numerical data. We're going to  collect metrics, and we are going to review these metrics, and we're going to  use them to interpret how well we are functioning as an organization, and what  our strengths and weaknesses and where our opportunities and where are our  threats. The word statistics is used in two ways. One, statistics can be described or a descriptive measure computed from a sample and used to make a  determination about a population. So we're going to collect this data out of a  sample pool, a population right inside this sample, we're going to test them.  We're going to see maybe how many pairs of shoes do 10 people own? And  then we'll make an inference about the types of shoes that this community likes,  so that we can then sell those type of shoes to that community and make sure  we target them. So statistics are very important in business to help you find  opportunities and take advantage of them. Statistics can be distributions used in the analysis of data. Example of statistics, suppose we have done a fruit survey  on which fruit is favorite, according to the collected data, we have done analysis. So you can see in the analysis that 25 people like oranges, 20 people like been  like apples, 15 like bananas, and 40 people like pineapples. So according to the  pie chart, we can say that pineapples are the most favorite fruit and banana is  the least favorite among them. And this is an example of statistics. You can tell  people's tastes and preferences through sampling the population. So to  statistics in marketing, business, statistics can be used to target a market, right? So just like, just like we spoke about a few minutes ago with the banana  illustration, we know that that segment of the population really, really likes  bananas. So we'll target more banana sales to that population. Example, online  another example online shopping for electronics. Let's suppose, after we review  previous sales data, we found that men do more online shopping of electronics  than women, we also find that the age range of male online shoppers is 25 to  40. According to these data figures, we can identify a target market and then  define our marketing strategy. So we define the target market. We know in this  example that men 25 to 40 do a lot more shopping online for electronics than  women do. So we can take this data and we can now market whatever it is that  we sell the electronics online towards shoppers that are male 25 to 40, hoping to increase our sales now, let's look at a statistics and finance. Let's look at an  example. Here. A study has been taken to rate Frankfurt, London and New York  for best climate for a financial center using several criteria. The criteria denotes  that it is one to five. Scale five denotes very good and one denotes very bad for  the personal and corporate taxes. Criteria, Frankfurt received the lowest score  of 2.44 while London and New York received 3.61 now for living and working  environment criteria. Out of these three cities, Frankfurt received a 2.62 rating  on the one to five skill, while London had 3.58 New York 3.62 so for a 

cumulative you can see that New York, based on these statistics, is the most  desired location for the best climate as they scored a 3.62 and a 3.61 while  London scored a 3.61 and a 3.58 New York had the better rate for the living and  working environment criteria, which from these statistical inferences shows that  New York is the better climate for the financial center statistics and  management. A survey conducted by company ABC Small Business Network  asked small businesses, small business owners, how they would characterize  themselves. Approximately 33% they describe themselves as seeing the big  picture. Who focuses on efficiency. 27% label themselves as problem solvers,  who concentrate on solving difficult problems. 16% were rainmakers, focusing  mainly on developing new business. And 11% were artists who involved, who  were more involved in creating new products than running the business. So this  is these were their thoughts about themselves. You can see how they  categorized who they thought they were, and you can make a statistical  inference that whoever was taken the sample size was taken out of this  population and whoever was questioned, then you can see that most people  that are in this population see themselves focusing on efficiency, sample,  population, sample and census. Okay, these are very critical terms to  understand in the statistics world, right? A population is a collection of persons,  objects or items of interest. When researchers gather data from the entire  population for a given measurement of interest, they call it a census. So now we have our our population. Now we've gathered the data. The measurement of  that data is called a census, a sample population. A sample is a portion of the  entire population. So there's 100 people inside this population. We want a  sample of that, 100 people. So we'll say survey 10 people inside that 100 and  have a census, descriptive and inferential if a business analyst is gathering data  on a group to describe or reach a conclusion About the same group. Most sports statistics, such as batting average, rebounds and first downs, are descriptive  statistics, right? So they're gathering data on a group to describe or reach a  conclusion about the same group. So inferential if a researcher gathers data  from a sample and uses the statistics generated to reach conclusions about the  population from which the sample was taken, right? So if a researcher gathers  data from a sample and uses the statistics to generate and reach conclusions  about that sample, right, so the very the impact of advertising on various market  segments are inferential statistics. So let's understand parameter and statistics.  Right. A parameter is a descriptive measure of the population. Parameters are  usually denoted by Greek letters. Right, you'll see a sigma or summation  examples of parameters are population mean, population variance and  population standard deviation, and we'll get into these a little more in depth later  in another video. Descriptive measure of a sample is called statistics. Statistics  are usually denoted by Roman letters. Examples of statistics are sample mean,  sample variance and sample standard deviation. Levels of data management, 

there are four common levels of data management, nominal level, the ordinal  level, interval and ratio level. So the nominal level, the nominal is the lowest  level of data measurement. Let's look at an example. Student identification  numbers are nominal. The numbers are only to differentiate students, not to  make value statements about them, so it's just an ID number. Other examples  zip codes, social security numbers, statistical techniques for analyzing this data  is very limited. We're not trying to prove anything. There's nothing that you can  really infer out of these nominal level numbers, because they are just an  identifier. Ordinal level. Ordinal level data measurement can be used to rank or  order statistics. Ordinal level data measurement is higher than the nominal level. We'll take an example. The computer tutorial is one, not helpful. Two, somewhat  helpful. Three, moderately helpful. Four, very helpful. Five, extremely helpful.  Here we can see five is the highest rank and one is the lowest. So according to  the answer, provided we can rank the usefulness of the tutorial. Some other  examples are measurement of risk and a mutual fund top 50 most admired  companies and the Fortune magazine. So this is really just like an opinion,  right? So we're ranking it one to five, and we thought that it was extremely  helpful to not helpful. So we can get a sense of direction on how a sample or a  population measured would feel about a certain thing. But you know, it's not  going to give us too much insight. We're just going to get an overall, broad feel  for what's going on inside that population at the ordinal level, interval level,  interval level data measurement is next to the highest level and interval data, it  measures the distance between consecutive numbers have meaning, and the  data is always numerical. The distances represented by consecutive numbers  are equal means. Interval data have equal intervals. Let's look at the example,  Fahrenheit temperature numbers. Temperature can be ranked, and amounts of  heat between consecutive readings are the same, 20 degrees, 21, 22 see  they're ranked. They're different. So you can rank the temperatures from coldest to warmest, if you'd like. Some other examples are the percentage change in  employment. How does the unemployment rate change the percentage return of a stop? Notice, it's interval right intervals, time period. What was the percentage  change of unemployment from 2011 2012 that's an example of an interval level  statistical measurement ratio level ratio level data has some properties as  interval data, but ratio level data has an absolute value, and the ratio of two  numbers is meaningful. The ratio level data measurement is the highest level of  data measurement. The notion of absolute zero means that zero is fixed, and  the zero value in the data represents the absence of the characteristics being  studied. Let's look an example weight with the ratio data. We can state that 90  kilograms of weight is twice as much as 45 kilograms, or we can make it into a  ratio. Notice 90 to 45, or two to one. Some other examples are height, time  volume, Kelvin, temperature, the production cycle time, work measurement time, number of trucks, sole number of employees. This can all be measured in ratios.

You so let's look at a comparison between non metric and metric data. Right?  Non metric data can be called qualitative data. Nominal and ordinal level data  are non metric. So qualitative data, let's go back to that real quick so you can  

see quantitative data on the quantitative data on the right, but qualitative means  that it's not tangible, right? It's a non metric data. It's almost like a feeling or an  observation of an attitude or an idea or something like that. Where a metric data  is quantitative, we can record it, we can see it, we can feel it, right? So non  nominal and ordinal level data are non metric. It is derived from imprecise  measurements, such as demographic data. It's imprecise, right? It's qualitative,  right? We can't really measure it precisely, whereas quantitatively, it's an interval and ratio level data, and they are metrics. So I can record it, I can physically go  out there and see the differentiation. So it's quantifiable, usually gathered by  precise instruments used in the engineering process.