3. How do we use statistics in agriculture?

Transcript

Welcome to the next lecture. In this lecture we are more focused towards how do we use statistics in agriculture. Till now we have been very hectically or very focusedly describing about only farm and agriculture. The same thing whatever we do you can use this as an analogy to your own area of interest like aqua or Horti. As I told you data is something which is very important for statistics. There are two types of data, one is called as qualitative data, the other one is called as quantitative data. In agriculture we many a times have qualitative data. The qualitative data deals with description. You should understand, it cannot be measured. few examples are color, smell, appearance, etc, etc. So now with this qualitative data if you want to do a statistical analysis it is going to be lot of subjectivity. In order to make our model more realistic, and more useful, we always try to convert qualitative data into quantitative data. For example, the color can be split into five different types. For example, yellowish green, then you can have royal green or royal yellow. something like that. for each one of it give two two marks. each one of the color variations give one mark or two mark. Establish a table, from that table you look at it and then you try to write down the marks. So, what did you do you converted a qualitative data into quantitative data. Quantitative data deals with numbers, why, because when you try to apply math you need to have numbers. So, you try to convert it into numbers, and all the quantitative data can be measured using an instrument. Let it be water content, let it be fertilizer content, let it be pollution content, let it be the ph of water, let it be some insects trying to hit at an area, all these things can be measured through a device or visually you try to convert your quality to data into quantitative data. Examples of quantitative data are length, temperature, weight, humidity, etc. etc. The data whatever we collect the data corresponding to grouping of the sample happens using a table or graph. This is how we try to collect the data. Second thing is the data are calculated for some data point or I would say information. Ok so you should understand this, qualitative data and quantitative data. The taste of a tea is a qualitative data, the ordour of a apple is a qualitative data, the diameter of the apple becomes quantitative data, the aspect ratio which is unit less parameter again becomes a quantitative data.

So, when we try to take the data and represent in a graphical form, it looks something like this. So, here I have used histograms. So, I have just put the data. So statistics is also describes the variation in the data. We use word location and dispersion to characterize distribution of data. When you have a data which is plotted you see there, there are a set of data which falls in the center of the plot. This is x falls in the x axis, so, now you see most data are falling in the center so, these things are called as center location. In terms of data so this is systematic variation of data we will try to do from the looking at the center or location data. This data is nothing but the spread of data for example when you try to find out the diameter of apple, you can try to say the diameter of the apple is maybe 50 millimeter. So, it is the center portion, if you try to say the diameter is 50 plus or minus 5 millimeter then you have data from 45 going up to 55. So, this is called the spread of the data. So, this is the spread of the data, okay. So, this is what we have plotted here, the center location is 50 and plus minus 5 is the spread. So here this are very common in naturally occurring phenomenas, like, a diameter of apple it can be a length of a leaf the, the length of a stem, the spread of a root, all these things can be. So,you are supposed to talk about two important things one is the center location, the other one is the spread of data. From the collector data we try to represent it in three different ways. So, the three different ways are going to be mean, median and mode. Mean tries to take an average of the available set of data after collecting the original data that is the primary data try to make a copy of it in another file then arrange it all either in ascending or descending fashion, then start playing around with these three parameters, mean, median and mode. So, mean tries to take the sum of all data divided by number of data. For example, you have 12 to 18, now when you try to take the mean you try to do 32 divided by 3 which is approximately 11. So, you had three data points and because of one data point which can be an accidental error, so you have one data point which has come up too low when we try to see this data it will try to dictate the overall output. So, trying to take the mean or the average and talking about the entire response is not correct or might lead to misleading information, but the most easiest and simplest form where everybody has accepted is the Mean. Next, one is the Median when I have three data points trying to choose the center value of the set of data is the median so if you see here there are three points the center fellow is chosen as the Median. When we talk about the Mode, Mode means he is the most popular fellow who has frequently occurred in the available set of data so then that is called as them Mode. Many a times when you try to see it, it is not as simple data whatever you have here, you will have 12, 18 and 2. You might have 12.2, 18.4, 2.9. So, trying to take the data and trying to arrange them in ascending order or in descending order, we do these three representations.

So, in this lecture what we have seen is data center point, data spread, and then we saw about three important parameters which is mean, median and mode. When we try to do agriculturist, or when we try to do agriculture, let it be aqua, Horti or in normal Agri culturing you will always try to look for all the three important things, because in nature you do not get exactly data points to the required one. Say for example, when we try to do engineered products we know the base dimension, but in naturally occurring where there is lot of probability involved getting these three variables are very very important.

Thank you.