ANALISIS BUTIR SOAL (DAYA PEMBEDA)

Bismillahirrohmanirrohim Assalamualaikum warohmatullohi wabarokatuh How are you all? I hope you are still excited

In this video, I will continue the previous material which is still related to learning evaluation, namely the second part of the analysis of the question mark. If the first part or the first part is in the previous video, you can check it on my channel.

Before I continue, don't forget to click subscribe, like, share and comment, okay? For the sake of improving my next videos. Okay, in the previous video, it was a question mark analysis that I have discussed, namely question mark analysis with classical theory test with

level of difficulty, classical theory level of difficulty in this video I will discuss about the differentiating force then later I will talk about the effectivity of the distractor what is the differentiating force?

The differentiating power is the ability of the subject or item or particle to distinguish between good and bad students. So, for example, in a group in that class, when the subject has a good differentiating power, the subject will be able to distinguish which student is good and bad who can answer this question and who cannot answer this question.

The difference can be analyzed in two ways, the first is the index of discrimination and the second is the point of Biserial. The index of discrimination is D = PA-PB

Then for the point of Biserial, the formula is Rpbis, this is the coefficient of the correlation of point of Biserial, R, the symbol is Rpbis, equals to xi squared minus x squared divided by Sx multiplied by the root of pi per 1 minus pi. Now this

What is the discrimination index? Why is there D, PA, PB? What is the point of the serial? Well, this is the explanation for the discrimination index DA = PA - PB That is D is the discrimination index PA is the proportion of participants who answer correctly in the upper group if PB the proportion of participants who answer correctly in the lower group this formula is the same as

P is the same as the level of difficulty so if you don't understand what the level of difficulty is please understand first how to calculate P or the level of difficulty can be watched in my previous video then for the differential force using the point-B serial correlation earlier the formula

rpbis = xi bar - x bar per Sx multiplied by the root of pi per 1 - pi. What is rpbis? rpbis is the coefficient of point correlation in the serial. Then xi bar is the average score or average

score X for the participant who answers correctly on the item or the point to I so if we are for example calculating the difference, number 1 or point to 1 means I is 1 if the point to 2 means I is 2 like that then X bar is the average score X

Sx is the deviation standard or the deviation of the bar from the X score so later from the X score, it's the total score, the total score of each participant of the test then Pi is the proportion of the participant of the test who answers correctly the item to i, the i can be 1, 2, 3, the same, the i is the same as the x i bar

from both of those formula there is a difference in power criteria the criteria is if the value of D or RPBIS the value of the index of discrimination or the correlation of the point B serial is less than 0 or negative it means it's not good the difference in power is not good then if the value is between 0 to 0.2

is enough if from 0.2 to 0.4 it's average then 0.4 to 0.7 is good and more than 0.7 to 1 yes from 0.7 to 1 it's very good so

same thing with the level of difficulty, the difference is a maximum of 1, there is no difference in the value of 1 comma or 2 comma like that, so the difference is a maximum of 1, if the difference is the gap, if yesterday the value is from 1 to 0, yes, if the difference is from

the score of the difference I write the value of the gap is negative 1 to 1 so there is no difference in the value of negative 2 like that or 1 comma so from negative 1 to if I make

sorry, if I make a line of numbers from here -1 to 1 here is 0 here is 0 okay, for more details I will give an example immediately we practice counting it through excel okay, this is the excel

here I have prepared two sheets for the index of discrimination, the formula of index of discrimination and point of biserial let's go to the index of discrimination first, the easiest one for example here number then student ID then how much do we want for example we make

the train, yes, B1 to Kela 5, okay, okay, how many students are there, for example, there are 10, yes, okay, you can SD underscore 001, okay, okay, this is me, I'm going to move the

from here we can know the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the difference in the

just make or calculate the difference, we just have to group it into two groups, the top and the bottom group, okay, we look for the score so that we can determine the top class and the bottom class, we auto-zoom enter, then we double click, okay, this is

this is the total score from here we want what is it? this question is only 5 yes it seems a little too much, no problem, from this we immediately sort it by large to small so from the highest to the lowest

okay expand the selection okay so follow the highest is this student with ID 3 this is the order this order is in order I will tidy it up again from here we will mark this one

we make the upper group because there are 10 students so we divide the upper group into two from number 1 to 5, the lower group is 6 to 10 but because here there is the same if we divide this into two this is the same in the middle this can be deleted so we use the 40%

40% above and 40% below like that so if for example this is the same if for example this is not the same this is for example one yes here one it's okay we

the top group is 5, the bottom group is 5 but because this is the same, we can delete this please pay attention again, when it's right at the division of the top group students, I mean the division of the top group students and the bottom group students, it turns out that there is a score that is the same when in the middle, we can delete it

don't worry, "well, the student will be the judge" no, we are not judging, we are doing a question-and-answer analysis so if the student is not being followed for a question-and-answer analysis, it's okay so, to make it easier, we just delete it this is the top group, I give it yellow

and this is the bottom group I give the color green from here we will determine Pa, right? Pa is calculated with B we look for B first, what is B? remember to look at the level of the equation B is the correct answer for the number 1 molecule so here

Hi, let's see one, then we can copy paste here, here, here, don't be here, this is just a score for us to calculate from the highest student to the lowest, then we look for P, the price is the same as B per

how many "per" ? 4 because the students are 4 in the upper group if the upper group has 15, it means that the correct answer is divided by 15 like that we copy to the side this is the value then we look for B for the lower group this is PB to zoom this is enter

Okay, let's copy the pb, 1/4 too, because the lower group is also 4 people. Okay, this is the result, so we just have to look for the D, D, okay, I give it a blue color, okay,

d, the formula is this, pa minus pb, which means it's the same as this, minus pa minus pb, okay, this is the value, the criteria, yes, the criteria, if from 0 to

less than zero is not good, right? Well, we can use the formula IF if there are many, because this is a criterion, because this is just a little, so I'll just say if this is not good, if the 0.5 is average

SD, if this is very good, this is not good and this is average, I wrote this manually, I don't use the IF formula because it's just a little bit, but later if there are many, you can use the IF formula, okay, that's for the difference in the force by using the index discrimination formula, now for the point in the serial, this point in the serial

it's a bit complicated because we have to find the per- what is it called per-butir with a pretty complex formula okay, I'll move this first I put it here okay, let's make it number in the student the butir is the same, just make 5 okay, it's too small okay

here we are the same, for example, the student is 10 if you practice, don't have to be a lot, 10 is enough, the important thing is that all friends understand the concept like that for example, SD 001 then we make the data first, run between 0.1 we control A, control C

please fill in if here we only need the total score we don't need to calculate so we still look for the total score, it just doesn't work for the calculation from the highest to the lowest, okay, this is the total score, from this formula

we have to find XI XI bar, the average score X for test participants who answer the correct item means like this, I write XI, which answers correctly for number 1, right? I mark it, yes, here it is

this is right, it means the score from these people that I give blue, we look for the average, it means the same as the average, yes, the average is from this and this, how many are there, 1234567, there are seven people who answer correctly for

for item number 1, okay, we enter 3, the average, okay, then we look for X bar, if this X bar is okay, I just write X bar because it's X bar, it's X bar, because it's harder to find the symbol X bar, I just write the bar like this,

Hi, this is the average of all the scores, so here it's the same as this bridge, close the curve, enter, okay, the average is this, then what else is SX, now SX

Sx is the slope of the slope is above the standard deviation if in Excel there is already a formula std there is P there is S we use the S based on a sample estimate standard deviation based on a sample the point is the standard deviation of the sample so later you will see the explanation here

we have to estimate standard deviation based on a sample from this score then we look for PI the concept is the same as the level of difficulty test participants who answer correctly

item to i, so the PI is this, the number of this divided by 10, right? This means that this is the level of the score for question number 1, yes, it's the same as the sum, this is how many people are right, divided by 10, okay, 0.7, yes, the PI, now

from here we look for 1 - pi 1 - pi = 1 - this this is already finished means we just need to substitute to this formula RPBIS = XI XI bar

this is the average, minus x bar, we divide the closed-end by the standard deviation. Now we multiply this by the square root, the root is the square root or sqrt of pi divided by this, 1-pi, closed-end. Okay, this is it.

Okay, we enter. Well, this is the result 0.4G. Well, this formula, this is the difference, it's good, right? Oh yes, I was wrong, if 0.5 is good, from 0.2 to 4, 4 to 6, 4 to 7, right?

this is B, B, B, this means the difference is good, if the difference is this formula, we can't just copy it to the side because what we can copy to the side is this X bar as long as this is a key so that it doesn't move, yes, this X bar we lock it, we give the dollar symbol, so

the X bar is this then we also block the SX like this then press F4 if on the laptop, okay, then the P is the same, we can copy this

the one that can't be copied is this one X bar, XI bar why? because we see one by one oh this is the right one, it means the score is this one plus this score plus this score well, find the average, divide by how many people it is so the average of this score this one with this one for number 2 so for this part

we have to check manually using the correlation formula actually there are applications that we can use to analyze the question but I share this so that you can know how the manual works later I will share how

analysis of question marks with application application analysis of question marks like that okay this can be continued friends can practice at home each yes hopefully if there are those who do not understand can be asked in the comments column yes I say thank you

that's all from me hopefully useful see you in the next videos I end Wassalamualaikum warohmatullohi wabarokatuh