[music] welcome back to my channel. If today is the fourth time coming across my channel, please consider subscribing for the old returning subscribers? Thank you for your support! Today’s video is the continuation of the confirmatory factor analysis. So i’ve linked the video just down here so today, i will show you how to use the modification indices to improve your model of fit results there before we do that, let’s have a quick look at what a good model of fit results looks like if the summary of what a good model fit resort should look like. So this is what we should expect in our results for the choice by p-value. The target level is 0. 5, so this is wrong? Blue castro, when you have 0. 7 or above their 0. 5, its overfitting and left at 0. 5, is under fitting for the swimming swimming between 2 to 5, it’s a good fit, but when we have above 5 it’s not acceptable? Mf i have known fit index. My first, the same and above is a good fit, so we’ll have less than that. You have less than 90 percent.

If not acceptable! You can stick on the vatican 90, but never 90 percent is not a good fit, but for a perfect fit 95 percent and closed one is a perfect fit same, goes to 3, f, i, and also gf. I does the goodness of fit index for a pc. F!

I above 60% is a perfect fit for the model. Rm see the tidal level is practicing so should consider looking at different things!

So we’ll have our mc of modern 0? 08 is not acceptable, but less than their planes ever eats is acceptable. So this is it so this it means here that zero point zero five, two zero point: zero seven comes to be considered for our mc or above zero point! Zero eight is not acceptable, but for a perfect fit zero point a blow or less than 0. 05 is a perfect treat for our mc. So let’s compare this with the results.

We had our previous results to see if we are within b stretch hold or yes now. All we need to do that?

We need to use the most efficient indices.

To improve! There is au is the result from the previous video here!

I would like to say if you miss the last video on cfa, you need to see that first before washing this to be able to understand what is happening here, check my link, videos and click on the cs e to be able to understand [music] compared to all just discussed now. Underwater i’m, a good model fits so we have the same mean here 2? 25! So this is okay because it’s close to 2.

5? So this is acceptable for the g of i is zero point eight six? So this is below zero point. Five, zero point: nine five, so this is not quite good. Also, the nf i is a point. A -! It’s not acceptable, it’s great, but not the but not perfect? Okay, the cfi, cf i also 0. 89. It’s not so good what we need to use the modification indices on these results so plain to the air mc rmc 0! 06 is okay.

You know 0. 08? This is great, but the p clause is not significant! So we need to use the matrix on indices to improve these results? So, to do that, we go back today: model yeah.

Okay, we go to the outputs, the analyzer party, then the output? Okay, you click on demolition in this is here! So how do you save all right? it’s a while see ready so go to the view text chamber, okay, so go to be modification in this is after is in the matter of hc exterior did not change so go to the model fit in this. As you can see, we have quite a lot of numbers coming out here. So all this, if we need to go with these results, it means we are going to join all these together, which is so much this!

Is this so much okay? so we need to connect all these together. So what we need to do is we go to this much which i in this again and look at the look at the numbers that are more than 10? if they are more than five numbers here, that’s higher than it are 10 points, then we need to change the threshold to train so that i reduce this, because this is why i do not. You can not connect each and every every our result here before this is quite much so, let’s check for numbers above 10 here we have eleven point: five, nine! This is one of them these to the three these for these five so ready! We have up to five this six seven. So this is much where we already have more than five numbers above 10 point here. So what we need to is we go back to a results again to the model again go back today and property okay here demo future innocent we used for. That is why we have a lot of them common day!

So now, i have to change it to think save and run the resources, so go back to the v text? Much mission in this way- okay- not see. So this is what we have now have reduce dignam.

So this is these are all numbers above thing and they add numbers affecting the model! So now we’re going to join you to connect our designs together? Okay, to reduce our mean to improve the results, so we are going to drain artists, so this is pretty one we drain it, which one to eat to turn in each route, to unhear, 19, 20, 17 and 18! So we go back to our model. We have t920. So even this, this is the covariance drawing him using some connect. 19 and 20 is 19 and 20 nestings, 21 and 22, which will entreat you of 17 and 18 of 16 and 24 16. Here on 24, fifteen and twenty fifteen and twenty fifteen and twenty reason then i’m 27, then introducing this is 10 and 18 in and 18, six or seven six.

Therefore, an eating for an 18 year for an 18 is here five, a four and five four and five year idea. Two and three in the last of it is 1 & 2, so 1 & 2 is 1, and yes true? So now, after connecting the covariances now, what we’ll do now is receive. They will test the model again. Okay, so go back today, but but now who taught you want to turn into snow and see we’re having only one, only one, the whole new chick on it again, but before then, let’s check for the others it and see.

If it’s improved you’re, not okay, you can see now seminoles reduced one point. Six eight digit effect on yeah swimming blow for is very, very perfect for the model! So this is good. Okay, you can see this cf i now i mean sorry bgf i now i was increased to 0? 9 single this accepting the cf i is 0. 94 that very good, very perfect. Okay, the nf is thinning, 0. 89, okay, so let’s check for the rnc rnc?

Is there a point there for these good perfect? so he is very good fit.

We have a very good 3d, so this result is acceptable, so we can reports that we have a very good fit in now. So basically, this is the steps to take to do the modification indices, and that is all for today. Thank you for watching! If there’s any question on it, please drop it in the comment section below or email me on my facebook page at kristen’s analytics and please remember to subscribe. If you are yet to subscribe to my channel. Thank you and see you soon. [music]. .