Hello again,
If you look at the larger amplitude wave during the first half cycle it spans approximately 4 minor divisions (actually a tiny bit less than that but we'll call it that for now and you'll have to adjust yourself later). Since a half cycle is 180 degrees that means 2 minor divisions (half of that) is 90 degrees, and so 1 minor division is equivalent to 45 degrees.
Now looking at the last half cycle of the bigger wave, at the time it rises through zero (the leading edge) it is approximately 2/3 of one minor division past one major division, and since one minor is 45 degrees that means it (is 2/3)*45=30 degrees past the second to last major division. Looking at the smaller wave near the same major division, its rising edge going through zero is approximately one full minor division before the second to last major division, so that means is is 45 degrees before that major division. Since the smaller wave is 45 degrees before and the larger wave is 30 degrees after we add the two and get a total of 75 degrees phase shift.
The simplest way to find the phase shift from a graph is to look at the zero crossings where the waves are rising.
Keep in mind that the above calculation of 75 degrees was assuming the larger wave half cycle took up EXACTLY 4 minor divisions. If you look more closely you'll see that it is slightly less than that (note the second wave falling edge doesnt cross zero directly at a minor division and the third is even farther away than the second) so it's up to you to find a more accurate answer. For a hint, i would say the third wave falling edge goes through zero about half way before that last major division. I think it would be ok to approximate like this because the graph isnt that perfect, having some minor divisions that are a little wider than others. On the other hand approximating 180 degrees equals one half cycle may not be acceptable for the final result because it is clear from the larger wave that it looks perfectly synced with the minor divisions to start but toward the end it is not synced, indicating a difference that although not huge is still definitely noticeable.
It should also be noted that the leading edge of the larger wave is not 'exactly' 2/3 of the way past that second to last major division (at 6us), but a little less than that.
I might add that another way to do this is to convert everything to time and then measure the time between zero crossings and then convert back to degrees.
Given the above, lets see if you can come up with a good estimate for the phase shift now. A hint is the phase shift is between 60 and 90 degrees.
[As a final quick note, in real life we might be able to adjust the horizontal time scale to uncalibrated so we can adjust it until one full cycle spans exactly 8 minor divisions and then estimate the phase shift in a similar manner to the above]