Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Pr(Z<...) How much by hand.
Replies: 2   Last Post: Nov 5, 1996 1:44 PM

 Messages: [ Previous | Next ]
 Al Coons Posts: 898 Registered: 12/4/04
Pr(Z<...) How much by hand.
Posted: Nov 3, 1996 8:14 PM

The TI-83 allows us all to calculate (and if we want) see the shaded region
under various normal curves. Modern learning theory suggests that students
are more likely to get the correct answer by first using the TI to do
problems rather than having their attention divided between going to tables
and manipulating partial areas and, at the same time, trying to understand
conceptual what is going on. In other words, technology allows us to learn
the concepts first and then particulars later.
----------------------
Having said that I find that I am having my students do both the following
simulatenously:

a) complete write ups by hand using tables and probability notation Pr(z <
....) with no calculators,

and

b) checking their work using the full features of the TI.

While I am only a few days into this, my students say they find it very
satisfying to do it by hand and then check it by calculator...the calculator
confirmation makes them feel as if they really understand it (since they
generated an equivalent solution rather than just getting it out of an answer
key.)

------------------------

Questions:

1) If you are doing both tables and non-table (calculator only) work what
order do you do it in?

2) Do we really need tables anymore? Remember we certainly do not need or
use trig tables anymore yet when scientific calculators first came out many
argued that trig tables were essential. Are there some types of statistics
questions which can only be approached if you understand table look up????

3) Let's assume that we only use the calculator - no tables. Since the
calculator can calculate the area between two values for any normal curve

[e.g. normalcdf(40,82, 50,10) returns the area between 40 and 82
for N(52,10))]

should we be dealing with symmetry type questions? For example, should they
know that they could do the previous example as

normalcdf(-10000000,82, 50,10) - normalcdf(-10000000,40, 50,10)?

I feel that they should be I cannot explain why.

Albert Coons
Mathematics Department
Buckingham Browne & Nichols School
Cambridge, MA 02138
(617) 547-6100 x 264
AlCoons@aol.com

Date Subject Author
11/3/96 Al Coons
11/4/96 Pat Ballew
11/5/96 Bob Hayden