Ask your own question, for FREE!
Mathematics 16 Online
OpenStudy (fibonaccichick666):

If you were to teach a statistics lesson on comparing means(two sample t-test, pooled t-test,), how would you structure it? What activities might you use?

OpenStudy (fibonaccichick666):

Any ideas are welcome. I'm trying not to just lecture/powerpoint

Miracrown (miracrown):

I probably would start with an example, involving two groups, to see if one is actually different from the other. Maybe it would be based on SAT scores based on gender, or some other way of categorizing the two groups. And if the sigmas are not known and the sample sizes are not too large, then t tests would be appropriate.

OpenStudy (fibonaccichick666):

SATs, nice. Ok, I can probably pull that data. The class doesn't really like doing jigsaw activities, but lectures are so boring. I don't just want to have them watch videos on it, but that is what I am considering at the moment. Ya know, a flipped classroom.

Miracrown (miracrown):

Other than using paired data, you could try almost any comparison. Maybe it could be something measured, comparing student athletes vs. non athletes. Using examples that the audience can relate to may pique their interest.

Miracrown (miracrown):

You would actually have them running analyses as you go through the examples. What technology are you using in the course?

OpenStudy (fibonaccichick666):

none, outside of a calculator. To be honest, I am only teaching this chapter because I have to. I am student teaching and my mentor wants me to get experience teaching something I don't know/understand. It's not that the procedure is difficult, but I'm having issues putting it into everyday words and explaining it.

OpenStudy (fibonaccichick666):

I thought maybe brainstorming activities would help.

Miracrown (miracrown):

The reason I asked is because if your class is using TI-84 calculators, they could do the number crunching on the spot.

OpenStudy (fibonaccichick666):

yea, they have inspires or 84s

Miracrown (miracrown):

Also, at the end, you could provide them a quiz in the form of a PowerPoint, in which one slide asks a question, and the next slide (or next few slides) would provide the answers.

Miracrown (miracrown):

You could ask things such as when do we pool the variances. And even bring in the ratio of variances to see if it exceeds 4 for the F statistic or when should a t test be used vs. a z test.

OpenStudy (fibonaccichick666):

What is an F statistic?

Miracrown (miracrown):

or even compare using a confidence interval in lieu of a hypothesis test as an alternate form of hyp test.

OpenStudy (fibonaccichick666):

we are doing both confidence intervals and hypothesis testing

Miracrown (miracrown):

Base your questions on the key points you want to emphasize.

Miracrown (miracrown):

\(\color{blue}{\text{Originally Posted by}}\) @FibonacciChick666 What is an F statistic? \(\color{blue}{\text{End of Quote}}\) The F statistic is used when comparing two variances or std devs, as in a two sample hyp test.

OpenStudy (fibonaccichick666):

We have done that(I think), but we have not used that term. Does it have another name?

Miracrown (miracrown):

But when determining whether the variances should be pooled, one way is to take the ratio of the larger variance over the smaller variance and if that ratio is less than 4, then you can pool the variances. The F statistic is the ratio of the two variances.

Miracrown (miracrown):

I am trying to recall, but cannot think of it at this time. F is for Fisher for whom the statistic and distribution were named.

OpenStudy (fibonaccichick666):

it's ok. Thanks. This is what happens when you walk into teaching the very end of a stats class.(I think they have 2 chapters left or something.) I have no memory of this. It's been 10 years since I've used it O.O

Miracrown (miracrown):

and F tables are available to provide critical values, based on significance level and respective degrees of freedom, but maybe you do not want to get into that if the class has not dealt with it before. But the rule of being less than 4 for the variance ratio is used often if the sigmas are not known for each population biing compared.

Miracrown (miracrown):

Good luck! :)

Miracrown (miracrown):

The key is to plan things out, and plan for contingencies and make sure that you have more than enough material, since running short can be very uncomfortable. I speak from my own experience.

OpenStudy (fibonaccichick666):

Thanks, and thanks for your input. I appreciate it. I think I'm going to use your example ideas and do a flipped lesson using edmentum That will let me grade them for doing their hw.

Miracrown (miracrown):

Sounds like a plan.

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!