 Hello, I welcome you all once again to my channel Explore Education and I am Dr. Rashmi Singh, Assistant Professor, Department of Education, S.S. Khanna, Girls to be College, University of Al-Ahabat and it is my email ID. And today I am going to discuss a new topic under the new category of statistical analysis. Okay, this is a very important subfield of education, psychology and many other disciplines in which we have to read statistics and have to do statistical calculations. Okay, so the topic of today's discussion is parametric and non-parametric tests and the lecture will be bilingual mode and will be very useful for various topics, examinations and courses. Okay, so do subscribe my channel as well. Generally, we have talked a lot about psychology, I have covered the science in almost every course generally. This was a portion of research and statistics, which I wanted to talk about for a long time. So right now I am starting with parametrics and non-parametric and there are a lot of things to be told about. Data, variable, scale, after that there are a lot of parametric and non-parametric parameters. First of all, the basic knowledge is that what is the meaning of the examinations and what is the meaning of the parametric, non-parametric, parametric, parametric, what is the meaning, what are the assumptions. Then there are a lot of things to learn. So we will learn slowly. First of all, the knowledge of the examinations. A statistical hypothesis test is a method of statistical inference used to decide whether data at hand sufficiently supports a particular hypothesis. Now we have started with the hypothesis, but we believe that you know what is the hypothesis. What is the hypothesis? Sorry. When we go to do some research, do some kind of work, then first of all we can start with the vacuums. We have to assume some assumption that this is the case. So the assumed concept, which we have assumed is the same hypothesis. Why do we need that statistical test? That we are able to tell whether the assumption that we have taken is right or not. So if we put the statistics on it, then we get a lot of power. That John Gairn is right. He is statistically right. He is technically right. Scientifically right. Just don't talk about it many times. So this is what we are saying that the statistical hypothesis test is the same type of statistical inference. We can also estimate the statistics at the level of the data we have that the particular hypothesis is not supporting it. Hypothesis testing allows us to make probabilistic statements about population parameter. Many concepts are not mentioned in this. What is the population? What is the parameter? What is the sample? We believe that you all know or we will talk about it later. Hypothesis testing allows us to say that we are in a probabilistic statement. This is the inference. Hypothesis testing allows us to say that we are in a probabilistic statement about population parameters. Hypothesis testing was popularized early in the 20th century. Early forms were used in the 1700s. This means that it is popular in the 20th century. But it is popular in the 1700s. The first use is credited to John Erbuth-Nott, followed by P. R. A. Simon Laplace. John Erbuth-Nott's name was generally not mentioned. But Laplace's name was mentioned. But Laplace was popular and Laplace was popular. In analyzing the human sex ratio and birth. This is taken from Wikipedia. This means that it is in the 17th and 17th centuries. It is in the 17th and 17th centuries. This means that it is in the 17th and 17th centuries. This means that it is in the 17th and 17th centuries. What is the statistical test? A statistical test is a formal technique based on probability distribution for arriving at a decision about the reasonableness of an assertion or hypothesis. The reasonableness of the parikalna that we have taken, means that you can bring power and support to it. The probability distribution of reaching there is based on it for my technique. The normality of the population distribution forms the basis for making statistical inferences about the sample drawn from the population. We will discuss the sample from the population. As we all know, the sample represents the population. We apply the technique to the sample. We will not go to it. We will go to it. The normality of the population distribution. We have studied it in NPC. You can see my old video on NPC. In general, NPC means the normal probability curve. If we take out any data, then it generally follows the normal distribution. As you can see, it does not have a diagram here. It has a height and weight. What does it mean? What is the tendency of the data? That it stays on the center. And what is extreme, it works there. Now we will not go to NPC. I will show the link of NPC in the description. I will share it with you. When NPC will happen, I will be equal to the median. Then what are they saying? These are statistical tests. Two kinds of assertions are involved. Whenever we take a sample test, then we take two types of assertions. An assertion directly related to the purpose of investigation. One assertion is that the purpose of our investigation is directly related to it. And other assertions to make a probability statement. The second assertion, which is our data, which is our path, which is probability statement. What is the meaning of our research? In general. This is a statistical test. It is a quantitative research. It is experimental research. What is its main goal? We will put a sample test and generalize it on the population. So, it is necessary for generalization to make a probability statement. You will be able to generalize it on a large population. The former is an assertion to be tested. The first one is the purpose of the investigation. We have to examine it. And it is technically called a hypothesis. And it is called a hypothesis. Whereas the set of all other assumptions is called the model. And the second assumption is the model. And these are the two things that the statistical test measures. The first is the hypothesis. This is the reasonableness. And the second model that we are following is so static. It is following the normal probability. It is not following the probability distribution. What are the things that we will see in the future? So, we have to apply the statistics of the data we have taken. According to the statistical test, we can draw its inference. That we are supporting that hypothesis. Whether the entire data or our hypothesis is accepting or not accepting. Sorry, are you rejecting or not rejecting? We will have to talk about the null hypothesis. To tell you a lot about the hypothesis, you should also know about the null hypothesis and alternative hypothesis. So, there are a lot of things that we are following. We know a lot about it. Then it will be the last one. So, the two statistical tests are mainly in two ways. Parametric or non-parametric. There is no need to know anything about India. I mean, it will be the old and the new. There is no need to know anything about India. Parametric or non-parametric. There is no need to know anything about India. So, what are the parametric statistics? Parametric statistics are based on assumptions about the distribution of the population from which the sample was taken. That means, everyone knows that we will do research on the entire population. We will draw a sample from the population. So, the parametric statistics are based on the distribution of the population. That means, we know the distribution of the sample from which the sample was taken. In parametric statistics, the information about the distribution of the population is known. We know the distribution of the population. We are a researcher. And it is based on a fixed set of parameters and some fixed parameters are there. What else is there? The parametric test is the hypothesis test. The parametric test is a comparison of a particle. That provides generalizations. What is this? The quantitative research has the purpose of generalizing it. The more you set the statistics, the better you can generalize the sample. So, parametric test is the hypothesis test that provides generalizations for making statements about the mean of the parent population. One more thing. Parametric tests are there because we know the distribution of the population. That is why there is a central tendency to take the mean. Why do we take the mean? You can see the classification in the future. Because the mean is not skewed in the data. The mean is the center tendency of the static. But if the data is skewed, then it is better to take the media. Because the media doesn't matter which data is at the extreme end. We take the ring of the middle and take the ring of the top and take the ring of the back. So, parametric test talks about the meaning of the parent population. All parametric tests assume that the outcome is approximately normal distributed in the population. All these parametric tests assume that the distribution of the population is normal. This does not mean the data in the observed sample but rather that the outcome follows a normal distribution in the full population that is non-observable. This means that the result will follow a normal distribution if we plot it on the graph. On that population which we have not observed. What is the assumption? What is the assumption of the distribution of the sample should be distributed normally distributed? Samples are in the form of interval ratios. Now let's look at the scale of the program. The scale of the program is the nominal-ordinal interval ratio. The interval ratio is higher than the normal-ordinal interval ratio. Where all the numbers of samples are made, the parameters are placed on them. The nominal-ordinal parameter is non-parametric. Observation must be independent. The population must have the same variance. Samples must have equal or nearly equal variance. Non-parametric states are based on assumptions that the data can be collected from a sample that does not follow a specific distribution. This means that we do not know the characteristics of the sample from that population. This normal probability does not follow. In non-parametric statistics, the information about the distribution of the population is unknown. And the parameters are not fixed. Which makes it necessary to test the hypothesis for the population. This test is mainly based on differences in media. This means that the skewed data is given to the media. Because we do not know how much data is collected. How much data is collected. How much data is differing. When the mean is followed, when we know everything, the normal probability is followed, then the mean will have to be reached. But the skewed data also has to be reached. Why? Because we only take the uncle. We have to understand it from the mind. Hence, it is alternately known as the distribution field. Hence, it is known as distribution free test. You should know that non-parametric test is also known as distribution free test. The test assumes that the variables are measured on nominal and ordinal. What was there before? Your interval ratios are of nominal and ordinal. And non-parametric statistics are classified into three types known as non-inferential statistical measure, inferential estimation techniques and hypothesis testing. This does not mean that these three types are followed by non-parametric statistics. And the common assumptions in non-parametric tests are randomness and independence. And the common assumptions in non-parametric tests are randomness and independence. And the key square test is the most important example. Difference. A lot of people ask what is the difference between parametric and non-parametric tests? The key difference between parametric and non-parametric test is that the parametric test relies on statistical distribution in data, whereas non-parametric do not depend on any distribution. Sahiba, if non-parametric does not make any assumptions and measure the central tendency with the median value, while parametric is a statistical test that assumes parameters and the distribution of the population is known. It uses a mean value, non-parametric uses median value. When the samples are small and the distribution of the outcome is not known and cannot be assumed to be approximately normally distributed, then alternative tests called non-parametric tests are appropriate. They are saying that when the sample size is small and you do not know its distribution, if the data is skewed, then use non-parametric. This is understood with a table. Properties are parametric tests and non-parametric tests. What is the assumption? We know that in parametric tests, we do not know in non-parametric tests. The value of the central tendency is in the mean or the median. In correlation, in parametric tests, Pearson runs and here Spearman runs. Because it is the random order of Spearman. Probabilistic distribution parameter test is normal. It is arbitrary. The population here is known. It is necessary that you know or it is not necessary. This is used in interval data. It is used in nominal data. The applicability is of variables. And this goes on attributes. And reasons to use. This is very repetitive. Please do not read it completely. You will understand that the underlying data do not meet the assumptions about the population sample. The population sample size is too small and the analyzed data is ordinal or nominal. So this is all about non-parametric and parametric tests. And thank you all. And do not forget to like and subscribe my channel. I hope that you understood this concept. Enjoy my telegram group.