Errors/fixes Try to have two decimal places for all sample proportions Leave jbs in tray until sample proportion is on number line - then clear - 1sec pause - next sample "what is the most common sample proportion" if you get it wrong you get a nice error voice but you can't go back to change it - change to try again when you go back from propn of .75 question, likely/unlikely shading disappears from histogram' The first finish off needs to come after a next button Went back from no. line with 5 propns on it and got table with .44999999999. Going back changes position of proportion dragging label When 5 samples propns going automatically onto line .35 etc doesn't show - box too small Takes too long to come up with next button after showing likely and unlikely regions In part 2 sim1 Next button appears but it doesn't wait for you to click on it In part 2 sim 1 fit curve button comes under but earlier it is above - can we be consistent? Also, ask for likely/unlikely -sep step When we show both sampling distributions, could we label them on the left with the sample size? In sim1 part 2 when proportion calc'd automatically for samples of size 50 it is displayed really close to number line - could we move it up to just nder tray as in earlier part of sim 1 and in sim2 Sim 2 - take another sample of 20 jelly beans from Bob's jar - can tray appear immediately after this voice instead of after clicking on the jar - as with Sue's jar? (for both 1st and 2nd samples) In sim2 replace recordprop.wav with recordprop_2.wav Change voice from speedup.wav to speedup_2.wav Again too long a delay between showing likely and unlikely regions and next button appearing.
As with earlier- if you've hit the back button before this stage the histogram is shown without the likely/unlikely shading.
Simulation 1 part 2 Samplesize50 SO far in this simulation we've been taking samples of size 20. You have built a sampling distribution that shows you what sort of samples you can expect if you take samples of 20 jelly beans from a population where 40% of the jelly beans are black. What would the sampling distribution look like if we took larger samples? Let's try samples of size 50. Takesample50 Click on the jar and we'll take a sample of size 50 and calculate the proportion of black jelly beans. Recordprop2 Let's record this proportion on the number line, as we did with the smaller samples Drag2 Drag the sample proportion onto the number line Takesample50_2 Click on the jar to take another sample of 50 jelly beans Speedup Let's speed things up again. Click on the button to take 200 samples of size 50 and we'll build a histogram Shapecorrect You're right, once again the sampling distribution is approximately symmetric and could be approximated with a normal curve. Click on