 Hello everyone welcome back to another session in dentistry and more We have a small session regarding the hypothesis testing So we'll be looking into t-test and our test and chi-square test This is just a preliminary level of explanation. So I already uploaded many videos regarding biostatistics regarding the hypothesis testing and when how On what occasions the tests of significant changes and what are the parametric testing non parametric testing So I'm not going into details of all those things So this session is preliminary steps how to perform a hypothesis testing So if a question comes of t-test and our test or chi-square test you can write these six steps because this is the Basic steps for any testing. So if t-test is been asked you can write but only thing is The formula in the step four will be changed Similarly for ANOVA test and also for chi-square test So the first step of a testing before that we need to understand what and when to do a testing. Suppose we are having two groups where we apply a particular trick this A and B are having received two different drugs and we are checking the efficacy of these tricks and this group has around 30 people and this is also having around 30 people So we are trying to find out the efficacy of these two drugs, which is better and which is Not good. So first we need to create the null hypothesis for any study. We are Telling that there is no Significant difference or there is no particular effect for these trick that is the null hypothesis null means nil We are stating that there is no Difference between these two tricks. It says that both the tricks are equal So that is null hypothesis. Okay, so this is how every research starts Now we have to find out the degree of freedom. So let's not go into detail about the degree of freedom It's a little complicated Statistical terminology Let's learn degree of freedom and degree of freedom we can find out by n minus 1 if it is a one group and n1 plus n2 minus 2 n is nothing but the number Okay, so we are not going into the calculation of all this So first thing is to create the null hypothesis and find out the degree of freedom Then setting the p value p means the probability value The probability value of rejecting type one error type one error So just by heart all this i'm doing this for my undergraduate students So probability You usually kept 5 percentage that is 0.05 that is the error rate type one error rate is 5 by 100 That is 0.05 or 5 percentage So that is setting the p value or probability value or the probability value of having a type one error Okay, just by heart it notice type one error. We are not going to explain Next is a crucial step that is applying the test. So we have data that is we are checking Blood pressure for say we are checking blood pressure reduction So we have blood pressure of these a group and b group. Okay So around 120 this is 110 and this is a mean systolic blood pressure Now we need to apply these value into the test if we are using Test within two groups, then we need to apply t-test. So t-test we are having two classification independent t-test and pair t-test So in this case, we are going to apply independent t-test because both the groups are different We have a group and we have a b group P pair t-test that is the test which is being applied when we have a before after experiment That is we are applying a drug on group a and we are checking after 30 days That is before after experiment in a single group Then we'll be using pair t-test if we have two different groups That is group a and group b We must use independent t-test. Okay. So when we have Two groups sometimes it will be paired sometimes it will be unpaired. So in that case We need to use independent or pair t-test So after that we need to find out the critical value So there will be a table a con a fixed table just like our logarithm table Every test t-test is having one table and over is having f-test value chi-square test is having another table So in that table with respect to the degree of freedom, there is a cutoff value So that cutoff value we need to find out Depending upon the degree of freedom for this group. So this group degree of freedom will be n1 plus n2 that is 30 plus 30 minus 2 That will be 58 So critical value with respect to the degree of freedom 58 we need to find out. Okay So once we get this value by Putting these values mean standard deviation the sample size Suppose we get 50 Okay, and the critical value What we got is 58 So here the t value Is less than the critical value So we will not reject the null hypothesis we should accept the null hypothesis and We should state that there is no difference between the two groups. That is both the dregs We used are almost equal. We cannot say one is better than the other one But if this is greater than the critical value if we get This as 60 So this 60 Will be greater than 58. So we reject the null hypothesis And we says that there is a difference between these two groups. One drug is more Effective than the other one. So this is about the t test. Okay. So in ANOVA also We have different tests. So the formula is very complicated But in ANOVA There will be a third group when we have three groups Okay, so t test is always used when we have two groups And t test is three groups So the rest everything is same only thing is the formula will be different But ANOVA will be applied only when there is presence of more than two groups Okay, so that is the difference between t test and ANOVA. You can write all these tips in ANOVA also, but we need to have three groups Whereas the chi-square test chi-square test is a different format Where here in t test and ANOVA What we are commonly using is mean And standard deviation But in chi-square test, we will not be using mean and standard deviation. Instead we will be using Proportion, proportion means percentage Okay So the rest everything is same only the formula is different. This is observed frequency Expected frequency or is for observed frequency expected frequency. And this is expected frequency And finally we get a value. We Calculate it with our critical value with respect to the degree of freedom. Then we reject or Accept the null hypothesis in t test and ANOVA mean and standard deviation will be using chi-square Mostly the percentage will be using the number of people with diabetes number of people without diabetes, but in Cases where we are checking the mean RBS that is a random blood sugar level in those cases We will be using t test and ANOVA depending upon the number of groups two groups mean t test Three groups means ANOVA two or any excess number three four five six any number we'll be using ANOVA Chi-square test means number of people with one disease or number of people without disease that proportion percentage of people with one particular thing is Coming under chi-square test I know it is very Complicated one and it is beyond an undergraduate level, but still You can write these six step means definitely would get good marks Only thing is you need to write the formula for chi-square test and t test Mostly t test will be asking chi-square test. Sometimes they'll be asking But you need to write all these six steps. This is common for all steps. That is testing Okay, so I'll come up with a new topic in dentistry mo. Thank you