Isthereadifferenceportfolio-Eliserivera 1

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Is There A Difference Unit Portfolio

Part 1: Unit Summary


Learning Statement
Tell the story of how you made sense of the different concepts throughout this unit. Be sure to hit all
of the following talking points, and include all activities/worksheets that are applicable. Y
ou must use
artifacts (worksheet, turn-in assignments, notes, class discussions) from this unit as evidence.
1. Discuss the concept of sample, population, hypothesis, null hypothesis and sample fluctuation
(Assignments: Try This Case, Who Gets Measles, Who Gets As, Quality of the Investigation,
Two Different Differences, Changing The Difference, Questions Without Answers, The
Dunking

Principle).
2. Discuss how to use standard deviation and normal distribution to determine whether a
difference is significant. (Assignments: Mean and Standard Deviation Problem Set, Bacterial
Culture, Decisions With Deviation)
3. Discuss how to calculate 2 and how it can be used to measure weirdness. ( Assignments:
How Does 2 Work?, Measuring Weirdness With 2 , A Probability Table, Late In the Day)
4. Discuss how to use proportional reasoning to determine expected values in a two population
case. (Assignments: What Would You expect?, Whos Absent?, Big and Strong, and
Delivering Results).
5. Discuss how to apply 2 to a two population and theoretical model problem. ( Assignments:
Paper or Plastic, Is it Really Worth It, Two Different Differences-Revisited)
Part 1: Unit Summary Answers Here
Beautiful examples (Your assignment work here!
Evidence)
Throughout these assignments we had
to look at an 2 Probability Table.
Here is the link to this document.
https://drive.google.com/a/hightech
high.org/file/d/0B7XVsTKC__N6WllRek
VBWTRfbmM/view?usp=sharing

Amazing Narratives ( Answers to each of the


Discussion Questions)
1. Discuss the concept of sample,
population, hypothesis, null hypothesis
and sample fluctuation
Sample Population:
Is a group of units selected from a
larger group (the population). By
studying the sample it is hoped to
draw rapid conclusions about the
larger group.
Hypothesis:
Is a theory that has been put
forward, either because it is
believed to be true or because it
is used as a basis for argument,
but has not been proven.

Null Hypothesis:
Most of the time, the researcher
starts out with a neutral
assumption that precedes data
collection. The neutral assumption
is the Null Hypothesis.
Sample Fluctuation:
Is the notion that if you take
different (random) samples from a
large population, you would expect
to see some variation in the data
you collected, (do not expect all
samples to be exactly the same)
2.Discuss how to use standard deviation and
normal distribution to determine whether a
difference is significant.
Standard Deviation:
This is a set of data that measures
how spread out the data is. In
other words it says whether all the
data items bunch around close to
the mean or if they are all over
the place.
Normal Deviation:
When many measurements of the same
things are taken, the data fits a
normal distribution.
By using standard deviation and
normal distribution you can figure
out whether the difference is
significant or not. In Bacterial
Culture we had to take the average
number (1000) and the standard
deviation (50) to make a graph.
This graph was used to help
determine percentages that help
find out if the difference is
significant.
For normal distribution it helps
measure variation in science and
math.

3. Discuss how to calculate 2 and how it can be


used to measure weirdness.
So in this problem we randomized
two amounts for the observed amount
of heads and tails. To solve 2 you
take the sample size and divide it
by two. So we take 50/2=25 and plug
25 into both expected values.
Next we take the observed value for
heads/tails and subtract it by its
expected value. We square that and
divide it by the expected value.
You do this twice, one for heads,
one for tails. You dont have to
divide by the expected yet. Add the
numerators that are squared
together. So we end up with
98/25=3.92, which is our 2 .
To measure the weirdness you have
to use your total
2 and look at A Probability
Table to see the probability of
getting a 2 this size when the
null is true.
Then you take that number and turn
it into a percentage. The higher
the percentage the more likely you
should accept the null hypothesis,
and reject the hypothesis. The
lower the percentage the more
likely should accept the
hypothesis, and reject the null
hypothesis.

4. Discuss how to use proportional reasoning to


determine expected values in a two population
case.
Using proportions to determine
expected values in a two population
case requires you to make a table.
This table would contain the two
different populations, and the two
things that you are comparing with
the populations. With that, you
would have the total for each
population, the totals of what you
are comparing, and the overall
total. You then use the table to
determine the percentage of the two
comparing subjects, and times that
with the total number of the first
population, and then the second
population. This will determine the
expected values in the table. I did
this a lot in the packet Comparing
Populations, in all three Parts.

Link to handout:
https://docs.google.com/document/d/
11YSsKI-s7Ri0R9ehK4FRX0ZWGUmbF9eqlV
-cV6ho54k/edit
Below are images of the work I did
for this handout.

5. Discuss how to apply 2 to a two population


and theoretical model problem.
You use all of the
2s to get a total 2 . With this
total you can find the percentage,
on A Probability Table sheet. In
class we discussed that if the 2 is
lower than 5 we accept the
hypothesis, and reject the null.

Part 2: Putting it all together


How do you determine whether there is difference, statistically, between two samples from two
populations? Put another way how do you know that the difference isnt just sample fluctuation?
- Anonymous Visitor
How would you respond?
I would respond that, you could use the chi-squared to find the
percentage difference to figure out how different the two populations
really are. The difference is not sample fluctuation because if the
chi-squared percentage shows that the sample is weird, then it has to
be obvious that something weird is going on. If something weird were
going on,it would not be a coincidence.
THE END

You might also like