Class+Summaries

We'll post summaries of each class session here.Here is a sample format I'd like you to refer to when writing the summary.
 * Class Summaries

The following mathematical ideas were the focus of today's class:• idea 1• idea 2• idea ......

The way we developed idea 1 (2, 3, etc) was ...An important thing to remember about idea 1(2, 3, etc) is ...

Idea 3 (idea developed late in the class session) is something that we'll revisit in later class periods but we got a start on this idea by discussing ...

Announcement: Here is a website: [] that provides many graphs from USA Today newspaper. Click through and ask yourself questions similar to those from p 37 in the text. Keep this at top of page so it is easy to find. Test out this website! Really, you should view these graphs and test yourself as to type of graph and variable! Share your responses in class!! Note to all: You have provided good detail in summaries but read the above directions. Rather than perhaps focusing on the order of the material in the class, focus on what are the big ideas we talked about and describe the discussion.

December 1st, 2010 Session 25

To do some reminders, on the 29th we gathered random data from our calculators on the ProbSim NumSim application. With this application we gathered data based on the following story on experimental Probability:

We were going to the store to purchase cereal. In each bag of cereal there contains one of the six Harry Potter action figures/pens. How many boxes of cereal would we have to buy to garner all six figures/pens?

So with that data, today we looked at the data once more on our calculators. Using that data we were able to construct a histogram, and a box and whisker plot of our data. We had a discussion about the data being categorical or numerical due to the histogram looking similar if not identical to a bar graph. We learned that the data is in fact numerical, because we did not group it into categories. We compared it to our survey data, and a final agreement was reached.

With the 1-Var stats application we were able to find the data's specifics statistics like median, mode, upper and lower quartiles, and many other things. With observing this investigation more, we were able to come to a conclusive equation to finding Experimental Probability:

Frequency of event or outcome divided by trials.

With all of this settled, what did we learn? What were the big ideas? We learned that the differences between experimental and theoretical probability differ greatly. That experimental probability deals with the trials of going through the experiment yourself, when on the other hand theoretical may have the same question, but can be figured out without trials. We learned the definition of random, that is has no specific pattern trend in the long short run however, in the long run we know the probabilities tend to stabilize; hence, we always want to run many trials to get an appropriate probability estimate. With random events, we cannot determine what will happen immediately, for example, on the next spin, or coin flip, or roll of the die, BUT we do know what all the possible outcomes are and with the "long run" probability estimates, we can predict the chances of a particular outcome occurring. We also learned that probability measures the likelihood of something happening. We learned how to map it out on calculators using the probsim application along with our knowledge of the plots and lists.

At the beginning of class we went over the homework that was assigned, including a discussion on the worksheet's contents. We found that everyone several different ways to use an area model. mapped their area models differently. We came upon disagreements on how it should be drawn, especially when dealing with the Princess question on the worksheet. But using the tree model and class discussion we all came upon a agreement on how to draw them, along with explanations on why the certain way is best for our class. Can you describe the agreement we reached??

We also had quickpolls about the homework. Since time was spent on the experimental data part of the lesson, our quiz on probability was moved to Monday December 6th. It will be taken at the beginning of class with 30minutes on the clock.

AHE: -Actively Read page174 -page 176 # 1-3 -Check out Dr. Browning's link to the Monty Carlo Simulation

Session #23 November 24th, 2010 P153 #5 This questions asked if Shekira should write one half as 50%, 0.5, or 1/2. We first asked how the class wrote it, and then asked if they number represented percentage.

We then took a look at the different ways of looking at one half, by putting it in a number line.

050%1 00.5--1 01/2--1

Students see 1/2 and 50% better than .5 because of how they see it. When they see .5, they see point 5 instead of five tenths, which would give the visual of 5/10. Ultimately, it's what Shekira says, make her think for herself and see what she writes down.

P155 #8

We first drew a contingency table involving the two variable of sex whether or not they've received a ticket.


 * || Tickets || No Tickets ||  ||
 * Male || 8 || 4 || 12 ||
 * Female || 3 || 5 || 8 ||
 * || 11 || 9 || 20 ||

B asks us to find the probability of males, who got tickets, and males who got tickets. P(A) = 12/20 P(B) = 11/20 P(A & B) = 8/20 P(A or B) = 15/20

C asks us to find the probability of females, how many didn't get tickets, and females who didn't get tickets P(A) = 8/20 P(B) = 9/20 P(A & B) = 5/20 P(A or B) = 12/20

D asks us if there is some sort of pattern that would help us predict The probability of A or B. When taking a look at B and C, we're able to see that if you take P(A) + P(B) - P(A & B) = P(A or B).

P187 #1 This question shows two coins. One has red on both sides, and the other has red on side and blue on the other. The questions asks if two reds means Player A wins and one red and one blue means Player B wins, is this problem fair?

After making a probability tree, we determined that it's fair, the outcomes are equally likely. You don't have to do both branches of the tree. The probability of flipping a red on the first flip is 2/2, and the probability of flipping a red or a blue on the second flip is 1/2. When you take a look at the math, 2/2 x 1/2 = 2/4 = 1/2, meaning that it's fair.

P187 #2

This questions has two spinners, one with 1,2 and 3 on it and the other has 2,3,4 and 5 on it. If the product of both spinner is even, then Player A wins. If it's odd, then Player B wins. We're trying to see if this is fair.

After drawing this tree we see that there are 8 even and 4 odd and that it's not odd. fair

We're asked if there's anything to make this fair and it's determined that finding the sum and not the product would make it fair.

A new problem was put up. There were three bags. Bag 1 had 1 red and 2 blue. Bag 2 had 2 red and 2 blue. Bag 3 had 1 red and 3 blue. 1 marble is pulled from a bag and kept out. Another marble is pulled. If both marbles match, then Player A wins. If they don't match, then Player B wins. The class split up and made three probability trees. After looking at each one, it was determined that Bag 1 would have Player A win 2/6 of the time and Player B win 4/6 of the time. Bag 2 would have Player A win 2/6 of the time and Player B win 4/6 of the time. Bag 3 would have Player A win 6/12 of the time and Player B win 6/12 of the time. We also set up an area model for examining this problem (and a few others, I believe).

So, what were the "big ideas" that were examined in this session??

AHE Actively read pp184-188, p187 #3-5 Actively read pp190-192, p193 #1 Finish the handout checked by: Julie Scott

Session # 22 Nov 22, 2010 Outcomes vs Variables ex. Problem 9 from AHE Variables- smoking status and ability to smell the smoke. All probability lies between 0-1, 0 meaning impossible and 1 meaning certain or 100%, 1/1 When giving percentages always give two decimal places such as .9298- 92.98% not 93% Types of probability- Experimental or Theoretical Experimental- collected data compared- you must do the experiment or survey and collect the actual data yourself Theoretical- not collecting data. No actual research, but able to find outcomes. i.e "Say I roll the die, what's the probability of getting a 6?" It's simple to We know that the probability is 1/6 because the numerator counts the number of ways the event can occur and the denominator counts... what? Someone else finish this. Dr B The better you want your estimate to be, the more  trials you need to conduct in the experiment. The percentage or decimal eventually stabilizes "in the long run." Somehow this connects to the idea of random. What does it mean when we say the outcomes are random? We covered pg. 158 Exploration Theoretical in class. <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">Prime- divisors are 1 and itself only- 1 is not a prime number <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">For <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> theoretical <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">probability fractions- the numerator is <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> the number of <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">specific outcomes <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> in the event <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">/ denominator being all possible outcomes <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> (see earlier statement about theoretical probability) <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">We were able to find tools and tables to help us compare and find the probability of things with different wholes. <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> *Branches <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;"> *Making a table with all outcomes <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">We looked at how to find the probability of events that have more than one stage. The first example: suppose you toss a coin and then roll a die (those are the two stages). We had to find P(H,5), the probability of getting a heads followed by rolling a 5. Tristan shared how she used a table to order the outcomes from two stages. Dr. B turned it into a rectangle to show how the area of the pieces of the rectangle show the probability of the outcomes. Cole described how to use the probabilities of the outcomes at each stage to find the final probability. The probability of getting a 6 is 1/6 but getting that WITH a heads will only happen 1/2 of the time. So 1/6 x 1/2 is 1/12; the P(H,5).

<span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">AHE- Finish the problems emailed last week, read through pg 144 to modify our definition of random, pg 152 1,7,9, pg 187 1,2 <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">Next project was handed out. <span style="color: #ff0000; font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">We need to find new partners for the second project. If we have a good reason to keep the same partner(s), then we need to submit our request in an email to Dr. B. <span style="font-family: Arial,Helvetica,sans-serif; font-size: 12px; line-height: 25px;">Checked by Amber Miller

<span style="color: #ff0085; font-family: Tahoma,Geneva,sans-serif; font-size: 110%;">Session #21 Nov 17, 2010
-Part- to- whole relationships with associations We cannot forget to remember that we cannot compare a cat/ cat association bar graph if the two categories are not of the same whole. I think there is a confusion between ideas here. The two different categorical variables can come from different wholes and we can compare the fractional pieces between them BUT we can not do arithmetic using the percents across the different outcomes of the variable if the wholes are different. Make sure you understand the difference. -What is a fraction A percent is not a number in itself, it is a symbol that represents a part-to-whole value. It must be converted to a numerical or fractional value. Remember 88% is 88 out of 100. Decimal Fractions (.06 is 6 one hundredths or 6/100) -Definitions of Random No specific pattern wasn't there more to our definition of random? -What is Probability The chances that something will happen we'll have to work more on this definition Theoretical Probability Predicting without experimenting (in a heads or tails situation, we predict 50% will be heads, 50% will be tails) Based on information about the event, we determine probabilities.

Submitted by Erika Swartz checked by Julie Scott

__ Session # 20 Nov 15, 2010 __
Main Ideas: · Difference in Variables and Data · How to make a scatter plot for a num/num association · How to do a cat/ cat association with a stacked bar graph

__Difference in Variable and Outcomes__ The class had trouble telling the difference between the two. So we discussed that the variables for the graph we made for homework were fat grams and calories. The outcomes were the numbers you got for each variable. Look back to Session Three 9/15 for a more detailed description.

__ How to make a scatter plot for a num/ num association __ · Used the data from page 115 to make a scatter plot. · Decided what out window was going to be. (Because the data is paired data you cannot sort it the way we usually do when trying to find the max and min.) · We then learned how to sort the data while maintaining the paired data. 1. Go to 2nd Stat. 2. over one to the right and click OPS 3. Select SortA( 4. 2nd stat and select fat (your screen should look like this SortA(LFat  5. Press the comma button (Above the STO button)  6. 2nd stat and select cal ( your screen should look like this SortA(Lfat,Lcal  7. Press enter  8. Now go into your lists and you will see that Fat is sorted and the cals are still paired with the appropriate fat.  (Remember the first list you put into the SortA( is the list getting sorted!) Once you know the max and min of the two lists you can easily determine your window. · Determined if there was an association and if so what would you say? · We said “The scatter plot shows a moderate to strong positive association because as your fat grams increase so does the total calories. It is a moderate association because the points are closely surrounding a possible trend line.” · Did manual fits · 2nd stat · Go left to get to calc · Select manual-fit · Once you selected your line you must write down your equation. Dr. Browing’s was y=12.76X+186.78. What if your equation disappears from the screen before you can write it down? · Hit “Y=” (Above the 2nd key…it’s a “soft key” as Dr. B would call it) · 2nd VAR (next to the clear button) · Select statistics (#3) · Go over to EQ (Equation) · Select RegEQ · If a problem was to say “find the equation and interpret it?” what would you say? · Discussed the different parts of the equation in order to help you answer this question. y=**__12.76X__**+186.78… 12.76 is the slope. Slope= rise/run.. which is ΔY/ ΔX. So you are saying that the change in y is compared to the change in X. For our example in class Calories/ Fat. Since the slope of our equation was 12.76 we would say 12.76/1…for every 1 gram of fat there are 12.76 calories. · y=12.76X+**__186.78__**... 186.78 is our y intercept. So when X=0, Y= 186.78. meaning if there are no grams of fat there are still 186.78 calories.

__How to do a cat/ cat association with a stacked bar graph__ · we used the data on page 67 · Page 68 we filled out the 1st stacked bar graph. · The dependent variable (X axis) is the variable placed outside of the stacked bar graph and the independent variable (Y axis) is the variable that makes up the bar graph. (Look at page 68 for example) · Make sure that after you make the bar graph to add on the side somewhere, where you got the percentage. · **__AHE: fill out the 2nd stacked bar graph!__** · Interpretation of thumb dominance: “Could you predict what eye dominance a person would be if given thumb dominance?” We said no because both were very close (Almost 50/50). · Example in class was page 71 with the smokers. When answering the question above you could say something like “If you are not a smoker you are more likely to detect a particular odor but if you were a smoker you could not tell an association.” __AHE:__ · Do page 67 fill out the 2nd stacked bar graph · Write your own definition of Random and then look it up and tell the definition that you found.

Checked by Cole Esio

Session # 19 -Nov 10, 2010
Main Ideas:

Displays for Cat/Num association:
 * dot plot
 * back to back stem and leaf
 * histograms

How can we tell when there is association?:
 * points will cluster around a trend line through the data.
 * a strong association is when points are closer to the trend line and more equally distributed.
 * a positive association is when points start lower on the y-axis and get larger as the number on the x-axis increases. The would cause the trend line to travel up to the right.
 * a negative association is when points start higher up on the y-axis and decrease as the values on the x-axia increase. This makes the trend line travel downward and to the right.

X and Y Variables: The variable on the x-axis is what we call the independent variable. When comparing, what happens on this variable is supposed to affect what happens on the y axis. The variable on the y-axis is the dependent variable because it would change depending on the independent variable.

Displays for Num/ Num association:
 * scatter plot
 * line graph ( trends over time)

How can we use those associations?
 * There is a difference between a comparison and an association YOu can compare any two random things, but to associate you are looking for connections.
 * Lurking variable s are variables that are not directly examined but may be what is causing an apparent association between two other variables that appear to be non-related, such as life expectancy and # of people per TV.

Cat/ Cat Displays:
 * contingency table

AHE Female Life expectancy /PPT association. pg 82 # 6 pg 85 #11 do problem plotting a line graph and also a trend line fit onto the line graph. Chose men's or women's. Page 115 # 18 a-c but do a manual fit line instead of a median-median fit line. If you have already completed p 85, you do not need to complete #18 as well unless you need more practice for proficiency. Complete your take home quiz.

<span style="color: #ee4490; font-family: Georgia,serif; font-size: 140%;">Session #18 November 8,2010
We started class today by putting our homework questions on the board. Since we were only assigned one problem aside from readings, the question on the board was p.132 2a The only problem people had with this was they were unsure about how to define L2=L1+10. We found that if we are simply in the header and enter second stat L1 +10 and then enter--- the list that we want will pop up! We looked at the problem and we concluded that if you add the same amount to the new list, to every value (10) the Standard Deviation will NOT change. However, if you multiply every value by the same number (2) the Standard Devaiation WILL change, and in this case, it was twice what it was originally; since we multiplied it by 2. We constructed a Box plot of the L1 and then a histogram in the same window. Turning plot 1 & 2 on at the same time. When constructing a box plot, it does not matter what the y axis is set to.

Next, we wanted to get some things straight so we came up with some rules that are always true: We said that a Standard Deviation measures a spread or distribution of data from the mean. We also said that the bigger the Standard Deviation is, the bigger the spread of the data from the mean.

Dr. B Then sent several quick polls to our calculator to make sure the whole class was on the same page. Of course we never got 100 percent, so we had to clarify some and say why they were yes, or no.

1. Is the mean average & median middle? They can in fact BOTH be averages, so a better way to say this would be mean is fair share, evening out, or an average && the median is the middle or an average. 2. Is the upper quartile one fourth of the data? A quartile is simply one point, not the range. 3. She made a dot plot where it had the same amount of data on each side but a huge gap in the middle, and then she asked us if the data was skewed? Just because the plot has values on either side does not mean the data is skewed. They have close to the same amount on each end, so it would actually be pretty symmetrical and maybe it would be bimodal.( Which means it is a data set with two modes) 4. Is there always the same amount of data above and below the mean? Because this is only true with the median. Careful, not ONLY true with median. The mean can have the same amount of data above and below (check out your writing assignment) it is just not ALWAYS true. The median is the point that has to be in the middle, there has to be an equal amount of values on either side. The mean however, does not have to be in the middle position but an equal fair sharing middle (or balanced middle). 5. Does this data set show how extremes impact the mean? ( The dot plot was a dot plot where all of the points were equally distributed across in one single line. Since the data is equally distributed there are no extremes. The data is equally spread.
 * Quick Poll Questions:**
 * NO**
 * NO**
 * NO**
 * NO**
 * NO**

__We now as a class have come up with 3 measures of spread:__ 1. Standard Deviation 2. Range 3. IQR ( Inner quartile range)

We then made a dot plot of our pulse, and if we exercised or not. Each of us went up and put our pulse rate on the number line, but if we exercised for at least 30 mins at least three times a week we marked it in black, if not we marked it in red. (This display is showing 2 variables, one numerical and one categorical) We then looked at the plot and wanted to see if we could find a connection between the 2 variables or come up with an
 * ASSOCIATION:** Which is being able to predict what the other variable will do by knowing the first variable.

With this data, we could not find an association, because our data was too spread. We changed the dot plot to have red on one side and black on the other and then it was easier to come up with some associations.Another example of an association that we looked at was between our arm span and height data from the PIQ. After doing a skater plot, we could easily see that there was an association because as one went up, so did the other.

There are 3 displays that visually show an association: 1.Dot plot 2. Histogram ( 2 sided) 3. Scatter plot

We then made a scatter plot of arm span and height data. This display showed us the spread and allowed us to come up with associations. Each dot represented the two data points for one person. Also, We could do the Manual Fit option to make a line, and then from that we could get an equation. __AHE__: p.48 # 1 Dr. B wants us to look at the Mc.Donalds and Burger King dot plot and see if we could find an association. p.135 # 2, 4, 5 and for #5 we are to use the data from p.g. 81

Submitted by Julie S. Checked by Amber M Dr B Checked by Sherrill H

Big ideas

 * 1) Box plots
 * 2) Outliers and interquartile range
 * 3) Measures of variability/dispersion/spread
 * 4) Administrative stuff

1.We discussed a few problems from the box plot homework. Cole lead a discussion on p 121 1e using ideas related to the balance and leveling off the block towers. The student's response in the text was incorrect, giving too much weight to the extra values of 93 and 81. Working as a class, we realized we had to consider the other 35 data values in finding a new mean. Using a block tower approach, we visualized 35 towers all the same height of 68.8 (the original mean) and placed two new towers of 93 and 81 in the picture. These two new towers added approximately 36 more blocks to redistribute to all 37 towers (key point) so each tower would only get a partial block, moving the new mean just a "smidge", to something a little less than 70. We used similar thinking with the balance idea. Make sure you understand how to explain where the "36 more blocks" came from.

2. Using the 4th grade TIMSS data, we explored the idea of outliers referring to a new measure of variability, that of interquartile range (or IQR). We can find the two quartiles on a box plot. REMEMBER: quartiles are points not groups (like fourths). This range is the difference between the two quartiles, showing us the spread of the middle 50% of a data set. An outlier is defined to be a data value that is more than 1.5 IQR from the upper or lower quartile. Thus outliers exist on the extreme ends of the data set. We uncovered two outliers in the 4th grade TIMSS data set. We used the 73 to view a box plot that shows up outliers on the plot itself. Kyla helped us with something but I forget what it was now! Help me remember.

3. We have range and IQR as measures of variability (MOV). The IQR is based on a distance using medians (median of upper and lower half). Dr. B led us through exploring another MOV based on the mean. We explored distances of data values from the mean of the same data set. Recalling the idea of the mean in general, we knew that if we added all of the deviations below the mean (negative values) and those above, the sum of those deviations would cancel each other out. So statisticians eliminated the negative value in looking at distances (see p 129 in text for example) and found the absolute value of the deviations (notice the difference between deviations and distances). We found the mean (absolute) deviation and used that as a measure of spread. Small mean deviations suggest the data set is clustered about the mean; large mean deviations suggest the data is spread out more from the mean. Only catch is, statisticians don't use mean deviations much; they use standard deviations. Standard deviations are computed very similarly to mean deviations only they square all of the deviations away from the mean (still eliminates the negative signs). The process has you finding the mean of all of the squared distances and then finally, taking the square root of that mean, producing the standard deviation (see p 130). Dr. B showed us where the SD is using 1 Var Stats on the 73. A lower case "sigma" symbol on the calc represents the SD we'll use in this course (see p 132). We need to be able to explain what the SD represents conceptually as well as compute it using the 73.

4. Survey project part 2 is due Monday. Need to have at least one graph using the TI-Connect software. One person in each group should have downloaded and installed the software from TI. You need a special cable to connect the calc to the computer. We need to see Dr B to get the cable but everyone has done that by now. We only need to construct graphs for 4 variables and we don't need measures of variability included in our discussions. We need to read the rubric carefully as all expectations are listed there. DO NOT put graphs in an appendix but put them next to the text that discusses the display; this expedites the evaluation process.

Calendar: Dr B wanted to have two more quizzes before the end of the semester do we voted to have them on Nov 15 (take home) and Dec 1st (in class). We will receive the take home quiz on next Wed and it will be due Nov 15. The last portion of the project is due 11/22 (notice the change in that date as well. Will she ever quit changing due dates???)

Our final exam is on Thursday Dec 16th at 8am. If we want a review session during finals week, we need to arrange the day and time. Dr. B can be there but she will not lead the review; we need to do that. She'll be there to assist our discussions.

Our exams were handed out at the end of class. Mean was 42/55 and the median was a 40.5/55. Upper extreme was a 54/55. Those having grades not up to their expectations should come to class early (7:30am) from now on to discuss homework, the exam, quizzes, etc to make sure there are no gaps in understanding. This is not required but HIGHLY recommended by Dr B. ALL students can earn Bs or better but effort and understanding is required. Good class. Math rocks!!

AHE: Actively read p127-130. Complete p 131 if you'd like to play with the lists on the 73. P 132 2a using the TIMSS 4th grade data. Finish any homework problems you have not fully completed. Make sure you understand the material and can clearly explain it.

On Monday we will return to p 66 and finish chapter 1 material. Written by an objective participant.

The following mathematical ideas were focused on today in class:

 * 1) measures of center
 * 2) box-and-whiskers plot by John Tukey

We started class by using our calculators to graph random numbers using the randint... function on our calculators. We then explored the medain, mean, and mode using this data. After doing this on our calculator we then went into the concept of box-and-whisker plot. My point for having us use this data set was to think through the process of describing a distribution of data. How do you begin? How do you decide what graph to use? What measures or statistics do you choose to use and what do they describe?

Something new we did on the calculator was put vertical lines in where our mean was. We did this by doing the following steps:
 * 1) press draw
 * 2) go down to vertical line
 * 3) you can move this line left or right

Something that was a key concept during class was that the range __is a distance__ ; it uses the min and max number to find that distance.

Another important factor we picked up today was that the median will not always sit in the middle of the range because it all depends on the distribution of the data.

__The Box-and-whiskers plot__ was invented by John Tukey. This type of display is clean and has only five points. The min and max, the median, and the upper and low quartiles. What can you read from the box plot?

=** HOMEWORK **= =** read actively page 117 **= =** complete exploration #1 **= =** p 121 #1-8 **= =** Also important dates to remember: **= =** Wednesday, November 3rd, 2010 How do measures of center compare? paper due **= =** Monday, November 8,2010 survey part two project due **=

Stuff added by Dr B = = = Session 14 October 25th, 2010 = Today we discussed homework problems to prepare for the midterm which is on Wed (aka tomorrow)

Pg. 101 #6 Conceptions of the mean: -Even-out -Fair-share -Broader topic- point/measure of center -Balance (distance away from center keeps the balance balanced, not the actual number its self) -Equalizing = evening out - We discussed that many of these above conceptions of mean start ideas with students about constructing/solving an algorithm -Most of these conceptions approximate the algorithm

Put #'s of #6 in numerical order 20, 21, 21, 21, 23, 23, 25, 25, 25, 26 -Make the visual "even-out" or equalize by: -tally marks -looking at the median- what does it take to get to the median? -bars/blocks

Pg. 101 #9 -Use the balancing method -Fulcrum is 147 -There are 7 points away from the fulcrum on one side, while the other side has 16 points away from the fulcrum -To even-out, the side with 7 points away from the fulcrum, 9 points need to be added

Pg. 101 #8 -3.45*6+3.12/7  -You can't add 2 numbers and divide (like normal) because one number represents 1 semester score and the ne number represents a score for 6 semesters (more weight at/on 3.45) -In order to find the cumulative GPA (mean) for the 7 semesters you need to add 3.12 to the other 6 semesters ((3.45 x 6) + 3.12) then divide the sum by 7.

Pg. 107 -Finding a "position" of the median -If the number is odd the median is in the data/or a data point (element in the data set) -If the number is even, you have to find the "in between" number of two points

checked by Cole checked by Ashley

Session 13 October 20th, 2010
Announcements:

-Midterm is Next WEDNESDAY! We will be tested on material from Day 1 till the 20th of Oct. Monday will be used as a "ask final hw questions day." -Survey Projects due Nov. 3rd. If you want a possible partner change, let Dr. Browning know about it soon. Reccomended to start this weekend! -Quizzes back today, Survey project part one not done yet (before Friday)(If you want to know your grade on Friday, email her your permission to send grades via email, and she will send you your survey step 1 grade. -Sit at new tables on Monday. Switching it up for the next 4-5 weeks.

__HW Questions asked in class__

pg 108 #7, pg 107 a,b,and 3a, pg 108 #6

__Big Ideas Covered in Today's Class__

//Numerical Graphical Displays:// - How many variables can each graphical display hold? We decided that the Dot plot can hold 1 to however many variables, granted there can be a limit (because of space, etc.) We also concluded that a stem and leaf plot is limited to only 2 variables. It can either be 1 stem and leaf, or a back to back stem and leaf. Finally with the graphical displays, we said that the Histogram is limited to 1 variable, and it is the only numerical graphical display we can do on our calculator.

//Our PIQ Data (Shoe Size)// As we gathered and plotted our data, we saw, as future teachers, that there can be some slight skewing in the data. Our graph made of Dr. Browning's labels and paper, had some slight off centered data. Some pieces of data were plotted, but they were off a little bit. As teachers we are going to see this, so we can adjust the number line spacings on dot plots and graphs to make plotting data easier. We also found the clumps, bumps, and holes. With the clumps, we saw heavy clumping in certain areas. As a class we decided that a bump is a word for "mode." The mode/bump was easily found, but we can learn that this isn't always the case with data. Holes in data can be various sizes. We saw small and large holes/spacings in our shoe size data.

//Symmetric/Skewed Data//

As a class we garnered a definition for Symmetric/Skewed data. When you are looking at a graph, for example a dot plot, and see the data pretty even and spaced evenly with one or two bumps and relatively the same amount of dots on either side and close to the same distances away from the "bump", then we can consider that data as symmetric. Skewed data is thought to be data, is data that has a bump with a much of the data in that area then has outliers that are very different from the rest of the data. An example of a skewed data set is the salary of the Longfellow teachers. When describing a skew you say that the data is skewed which ever way the outlier is. In our future lives as teachers we can use this information to teach children about these two types of data, when they gather their own data and making their own graphs. note:A graph can be neither symmetric or skewed.

//Mean/Median//

A question arose while arranging our PIQ shoe size data. If our data was symmetric, what piece of data could we add to make the mean change, but the median stay stagnant? We achieved a consensus that the value depends on the data, and the position depends on how many. We got to this consensus by adding data pieces to our data set, and seeing how the median and mean changed. With the data we had, we saw if we added numbers to either end of our data set, our median stayed the same, and the mean changed. We saw though, that the data added needed to be further away from the data set to severely change the mean, but //the median still stayed stagnant.//

//Categorical Displays://

Opening our Statistic's book, we looked at the graph about favorite yogurt flavors on page 105. We talked about graphing these in a stacked bar graph, or a regular bar graph, or even a circle graph. We decided for the data at hand, a regular bar graph would be best. After graphing this data, we started garnering information about it. The "mode" became the most popular flavor, but we realized that with categorical data, you cannot find a mean. Because as we all know FREQUENCY IS NOT THE DATA. In our future lives as teachers, we can use this example to test the children to see what they would do about the mean, would they use the frequency or figure out in the end that there is not a mean for categorical data.

Remember that our Midterm is on Wednesday. Come to class on Monday prepared with questions to ask!! checked by Cole and Julie S.

Session 12 October 18th 1. Pg. 97 #2 a) -11 bars above the mean @ 13 -11 bars below the mean @ 13 -this makes 13 the mean

b) -we need a balance above and below the mean -the line needs to be moved up -there are 60 bars above the line and 35 below the line that need to be accounted for -there are 15, 15, and 5 bars below the line -there are 35, and 25 bars above the line -you can distribute the numbers until all bars are equal @ 70 OR -you could also add the number of CD's up and divide it by 5 since you have 5 histogram bars. This is considered the "fair share" method. 35 divided by 5 is 70.

2. Pg. 97 #3 a) -we used balance beams to check our answers -Fulcrum: the balance point -put one weight at 10, one at 20, and one at 40. The other weight should be put at 70 since 10+20+40 = 70. -you could also count the difference between the weights. 50-40=10, 50-30=20, 50-10=40. (10+20+40=70) 80-50=30, 80-50=40. (40+30=70). They both equal 70.

3. mean=50 50*4=200 n=6 sum of numbers=200 Ex/n=x

4. QUIZ

5. Complete P 99 # 2, 3, 5-9. (should look familiar) Complete p 106 #2-7 (on Wed I'll add these problems plus maybe some others 8-14, 19) so the only new ones are p 106 #2-7. We have one more measure of center, mode, to examine more carefully then we are finished with those measures. Our exam is tentatively scheduled for Wed Oct 27th.

-**Project 1 Part 1 is due by Friday(10/15) at 12pm in Everett Tower 4th floor (make sure its dated, time stamped, and teaches name on it) if not turned in on class Wednesday (10/13)** __Reviewed homework problem__ pg. 62 #3c Male students are shaded in which equals about 11 and female students are un-shaded. Male students have more or longer cubit lengths. Female students are constant and have much variability. Two categories that are displayed are males and females. pg. 63 #4 The numbers (word lengths) that correspond on the y-axis correspond with each bar above it. We have determined that the research handbook is Article A because it has more variability in the word lengths. We also have determined that USA today is Article B. pg. 99 #1 dot plots: pg 100 # 2c pg 100 # 2d We now understand how to make a dot plot of 3 1/2 people we concluded.
 * Session #11 10/13/10**
 * data- the # in your family
 * distribution- how many times that # appeared
 * variable- # of people in your household
 * outcome- a numerical value
 * missing the raw data
 * no minimum and maxium
 * variation of the data set

-Main characteristics of a histogram is its a numerical type of graph and the horizontal axis-scale vs catergories and a bar graph is a categorical graph. -Frequency is NOT the data!!!! -A cubit is measured from the botton of your elbow to top of your wrist. -If you change the scale then the data would change as well. The data doesn't change but they will appear in different places on the axis if you change the scale (Dr B) Measures of center: typical/average -Sigma (looks like the letter E) x is equal to the sum of all the data values -N equals the original number you started with -Keep this in mind: The average cost of a homes is 150,000? What is that telling us by our measures of center. pg 103: -More sensitive to extreme values is the mean
 * mean- evened out;sharing fairly
 * median- middle (order)
 * mode- occurs most often
 * We posted dot plots on long white strips on the whiteboard to distingiush features of each one.
 * dot plot #1: range, bigger gaps, same # of data values in dot plot # 2, clusters
 * dot plot # 2: evenly distributed
 * QUIZ MONDAY** (10/18) on graphs
 * POP QUIZ IS STILL PENDING**
 * __AHE__**
 * complete pg. 99 #'s 1-3, 5-9
 * actively read pg 102 and through the explorations to clarify ideas developed in class today. pg 104 #2

<span style="display: block; font-family: Tahoma,Geneva,sans-serif; font-size: 130%;">Session #10 10/11/10

AHE: Intro to survey project + 5 surveys due wed 10/13/10 **can revise Wed and turn in Fri by noon but try to have it done Wed** survey info due 10/11/10(tonight) by midnight Actively read pg 96 complete pg 97 #2 pg 99 #1 & 2

Histograms

 * Numerical
 * use them if we have A LOT of data
 * start with numerical data, stays numerical data
 * grouped frequencies
 * if you collect by categories its categorical
 * *1-5, 6-10, 11-15<--- these are categories
 * if you collect by raw numerical data, but display in categories its still numerical
 * Histograms
 * 1,4,3,5,6,6,1,1,4,7,7,10...etc<---raw data
 * 1-3, 4-6, 7-9, 10-12...<---categories
 * intervals must all be the same range
 * check the intervals by subtracting the first number of every interval from each other
 * 1-3, 4-6, 7-9...
 * (4-1)=3; (7-4)=3

pg 58-59 #1.c.d.h
you had to examine Carlos' reasoning and his resulting table; there is an error in each
 * 1.c


 * he doesn't quite understand the interval lengths yet the first length is 41-46 which he thinks the length is 5 (46-41=5) but you have to take the first number of each interval and subtract them 41-46, 47-52... (47-41=6)
 * he also states there are 9 intervals because the range is 45 since he thinks the interval lengths are 5 still he did 45/5 which is 9. he got 45 because he subtracted 86-41=45 but you have to include all the numbers in the intervals so from 41 to 86 there are actually 46 scores
 * 1.d
 * the last question helps you figure out 'How many intervals the table should have if the interval lengths are five?' so our range is actually going to go from 41-90 now since the interval lengths go by 5.
 * = interval ||
 * = 41-45 ||
 * = 46-50 ||
 * = 51-55 ||
 * = 56-60 ||
 * = 61-65 ||
 * = 66-70 ||
 * = 71-75 ||
 * = 76-80 ||
 * = 81-85 ||
 * = 86-90 ||
 * there are ten intervals here**

make sure your window reads: -Xmin 53 -Xmax 104 -Xscal 5 -Ymin -3 (so we can see the bottom of the table) -Ymax 40 -Yscal 4 (children should know how to start counting with non-typical scales)
 * 1) 1.h we went thru this process together on the calculator

pg 61 #2.e
Q: approximately what percent of these students have the handedness ratio of 2 or greater?

we find % part/whole

we did dominate hand X's for 10 sec, then non-dominate hand X's for 10 seconds then you place Dom/Non-Dom (example 19 dominate, 10 non-dominate we get 1.9) the closer the numbers are to each other the closer the number gets to one meaning you are pretty balanced on both hands.

put the data from the graph into a table it will look like this the max data is between 3.6-3.7
 * intervals ||= frequency ||
 * 1.0-1.1 ||= 4 ||
 * 1.2-1.3 ||= 10 ||
 * 1.4-1.5 ||= 25 ||
 * 1.6.1.7 ||= 21 ||
 * 1.8-1.9 ||= 14 ||
 * 2.0-2.1 ||= 5 ||
 * 2.2-2.3 ||= 0 ||
 * 2.4-2.5 ||= 1 ||
 * 3.6-3.7 ||= 1 ||

there are 81 total frequencies and 7 of them are over 2.0 so 7/81=.086 so its approximately 9% of the students who have a handedness ratio of 2 or greater.

**go around room and vote for "what is statistics?" webs (vote up to three times)**
**__S__**tatistics**=** Process __**s**__tatistics= numerical values

use the cubes to make towers for 'How many people are in your household?' data: 2,2,3,3,6,6,6

Average--sharing/evening out
 * order them from small--->large
 * 6 was typical
 * order in a symmetrical shape (2366632)
 * 6 was typical
 * middle
 * more
 * bump
 * take cubes and share between towers
 * average 4
 * put all in one stack
 * 7 groups- share cubes between groups
 * keep taking one out put in each group till you run out and even it out
 * 4 each group average
 * Typical--most/middle/bump**

Measures of center
 * mode-most
 * median-middle
 * mean-average
 * younger kids will usually work with categorical data (pets? favorite color? breakfast?...etc)**


 * submitted by Kelsey Sheehan 10/11/10**

=<span style="display: block; font-family: Georgia,serif;">Session #9: 10/6/10 =

SURVEY QUESTIONS

====We started off class discussing and refining our current survey questions for our group projects. The new copy of "refined" survey questions was emailed to the class today. Issues such as word choice and overall question relevancy were discussed. A note pointed out during this discussion was that when we have categorical data, the intervals are not important in terms of length consistency. The first survey with this group of questions acted as our "pilot" data, which means in essence that it helped us to see what trends we are going to notice in our survey project so that we can be prepared for our results and audience selection.====

MYSTERY BALANCE WORKSHEET
====When looking at the homework Mystery Balance Worksheet we concluded that it is important to backup inferences with critical thinking, showing a good thought process as to why things turned out the way they do.==== Also noted was that when looking at dot plots and/or line plots it is good to notice the overall ranges of the information, we can draw inferences just by noticing slight differences in survey information. We not only needed to look at the frequency of information on the dot plot but also the distribution along the plot. - A side note: When we look at numerical data we tend to order things so that we can see the "big picture".

DATA SETS

Vocab word: Range- a spread from the smallest to largest (i.e. lowest to highest, minimum to maximum). The range helps us to find is the distance between the two; In order to find the "middle" of a data set we must count down the data until we reach the actual "middle" all pieces of data. Range is the distance between the highest and lowest data value. So if min of data set is 5 and max is 29 the range is 24 over the interval of 5 to 29. In data sets we can find clusters, open wholes/gaps, as well as a spread. This is used so that statisticians can notice distributions throughout the data set. Along with looking at distributions statisticians also look at high frequencies (bumps) in the data. This process of looking at the data is called "Describing a Distribution". Along side noticing these trends, we can look at intervals as well as the middle of our data. Generally we call these clumps, bumps and holes (clusters, data values that have high frequencies, and gaps respectively). HISTOGRAMS

Histograms help us group data closer together so that we can easily make interval groups. Histograms also help us deal with larger data sets in which thousands of points of data are grouped together. As goes typically in this process we tend to lose our raw data, however this does not matter to us as long as we can see the big picture.

From here we didn't quite make it to discussion homework questions, so we will be coving those next class.

AHE: WORK ON SURVEY PROJECT QUESTIONS AND GATHER DATA THROUGHOUT WEEK. First part of group project is due next Wed. Oct 13. Must submit 5 new surveys filled out by next Monday Oct. 11 by midnight.

Session #8: 10/4/10

USA TODAY GRAPH
We started class with a discussion about the graph in USA Today. The graph was titled, "What type of Accounts do your kids 13-22 have access to?" The graph displayed categorical data (the different types of accounts or savings). The class initially disagreed what type of graph it was. There were dollar bills so some said it was a picture graph, it was not a picture graph because the dollar bills did not represent a certain value. Others thought it could have been a circle or stacked bar graph because the "bars" were displaying a percentage. When we added the percentages they did not add up to 100 so circle and stacked bar graph were out because in those two graphs the data must display a part to whole and since we had no whole we could cancel those out. Finally we came to agreement on the graph being a bar graph even though it had no vertical axis, bars, or a whole, we said if it had those three things it would be more appropriate.
 * ===Categorical Data is answered by a category or group.===
 * ===Try not to say that something is categorical because it is NOT answered by a number.===
 * ===4 is a number but when you are putting something in a spot 0-5 you then have categorical data. Which cannot be put on a number line. Well, depending upon the context here. If you are displaying a histogram, then the interval 0-5 is on a number line. NOTE: It is possible to make numerical data categorical data, but you cannot make categorical data numerical.===

PG 50 #5 Beanie Baby Birthdays The stem and leaf plot on pg 5 was displaying the dates beanie babies were born. The students did not worry about the month or the year so, 2 I 3 meant a beanie baby was born on the 23rd of some month on some year. B) For the students that did not understand how to read the stem leaf some of us said we can ask the students, "look at the key, what does it mean to you?" Another question we could ask a class is "How would you display a birthday that was on the 26th?" both ways could get the class to really think about what the graph means and how to interpret it.

Notes:

 * ===When making the key use a number that is in your data.===
 * ===Do not use commas unless you have 2 numbers in your leaves, in which case you want to make all of them have 2 digits to keep things consistent.)===
 * ===Have data in numerical order going from center to edge (lowest to greatest)===

Spread
===When talking about spread, you are normally talking about the range of numbers in your //numerical data//. (it normally doesn't make sense to talk about the spread in categorical data.) In a stem and leaf plot if there is a small spread of data then the plot will look short and wide because most of the numbers will have the same stem. In a stem and leaf plot with a large spread would look long and thin because there would be many stems.=== Plus Rate Graphs

Dr. B. Handed out pulse rate graphs made by previous students and we critiqued them. - you can not put gaps between bars because it looks like you are putting gaps in the number line. -Normally it is more correct to have the vertical axis to show frequency.
 * 1) 1. Bar Graph showing numerical data that was put into categories.
 * 1) 2. This person didn't have spaces between bars but they used there vert. and horz. axis to display the data instead of data on horizontal axis and frequency on vertical axis. Then to show frequency they would make more bars with the same data.
 * 2) 3. This student made the vertical axis the data then put a bar for each student.
 * 1) 4 This student made a dot plot which is more appropriate to display numerical data but they made it confusing because they tried making the squares in the grid paper be the data instead of the lines on the paper.
 * 2) 6 .This student made a dot plot that was easy to read. This is the most correct way to display a class's pule rate

A.H.E.
===Dot Plots Mystery Balancers Worksheet; Histogram Read p. 55-56 actively; Complete exploration p.56-57; Complete p.58-65 # 1-4-- ALSO be sure to do your homework there is a possibility of a Homework Quiz in the near future!! :-)=== Also, last week's quizzes/tests were handed back. Reviewed by Amanda Locke && Julie Scott & Emily Spaulding

=Session #7 : 9/29/10= First idea talked about in class related to Categorical and Numerical data was looking at the question: Why does a bar graph work well to display the M & M data?

1- If we have different outcomes from the categorical variables we can see the differences between them with the bars.

2- We can see the frequency. The person looking at the graph does not need to count the data but can see the frequency.

3- Sizes of the frequency.


 * Second idea** talked about in class was how to take the data and make a stacked bar graph. Remember that in a bar graph each bar represents a part of the whole. As Dr. B demonstrated, if you cut out each bar and tap them end to end, this will represent the whole and each of the different colors of the bar graph from the M & M display represent the parts. Problem 3 on page 20 is what we used as an example.

Using the following data we constructed a stacked bar graph that was 12cm in height.

10 Blue 5 Red 14 Orange 7 yellow 14 Green 4 Blue Total = 54


 * Third Idea** talked about in class was thinking about how we could respond to Ahmed's understanding in question 2c on page 36. As a class we used the table on page 35 to determine the actual dollar amount contributed to the Democratic party 1993-1994 and 1995-1996. When finding these numbers we could for sure say that the Democrats did not receive more donations in 1993-1994 than 1995-1996.


 * Fourth Idea** talked about in class was the displays we made for pulse rate. Pulse rate is numerical data because when asking anyone what their pulse rate is, they will answer with a number. Bar graphs are not ideal for representing the data because we do not what the gaps in a bar graph to be confused for gaps in the data.

Fifth idea (idea developed late in the class session) is something that we'll revisit in later class periods but we got a start on this idea by discussing problem 7 on page 39. We will continue the next class talking about read, derive, and interpret questions.


 * Homework** - read pp 41-45 actively. Use the info from the PIQ on googledocs to complete p47 #2. Complete p 48 1-3,5,6,8 & 9. Read p 55 actively and look through the exploration.

checked by amber miller

=Session 6 9/27/10=

Categorical vs. Numerical Data
After collecting data you organize the data. Sometimes you organize the data into categories (color, size, etc.) Then you see the frequency of each category. This is what we did with the m&m data. The categorical data we collected we used to make picture and bar graphs.

We tried to make picture graphs and bar graphs on our calculators with the pulse rate data but were not able to because the data was numerical data. Line plots or dot plots are used to graph numerical data while picture graphs are used for categorical data only.

Reviewed Homework pg. 36 #3
At Home Extension: Someone did this calculation to figure out the number of math teachers in year 2. Why do you think this person set up the problem like this?

995000/100 =9950 Ans.x15.3 = 152235

Homework we still have questions on: -pg. 39 #7 -pg. 40 #8 -pg. 36 #2c

Survey Project
We reviewed the top 5 survey questions.

1)What is your current overall GPA?

2)Do you feel like you get enough sleep? Circle yes or no

3) Do you do better in classes when you get more sleep at night?

4) Do you miss class often due to lack of sleep the night before? Circle yes or no

5) How many hours of sleep do you typically get during the school week?

Survey Pilot due Thurs. @ midnight
-Make sure to hand the clipboard with the survey to each person

-Don't read the questions to the person and fill in the answers -Enter the data on Google Docs.

Quiz Wed.
-Types of data

-Types of graphs -Make sure you know how to make graphs on your calculator.

edited by Tristin Callahan 09/28/10 reveiwed by Ashley Hampton 9/28

Session 5: September 22, 2010: IDEA 1- Discussed Page 12, #5 A) 35 corner pairs at student cafeteria. 15 opposite pairs at student cafeteria. 29 corner pairs at hospital cafeteria. 12 opposite pairs at hospital cafeteria.

B) There are 6 possible seating arrangements. 4 of the 6 arrangements are corner to corner seating. 2 of the 6 are opposite seating.

C) People would be more likely to sit corner-to-corner. About 2 times more likely because there are 2 more options of corner seating (4 options) than opposite seating (2 options) Part/Whole= options of seating/ total # of ways to sit= corner to corner seating (4/6= 66%) opposite seating (2/6= 33%) Corner to Corner seating is about 2 times more likely than opposite seating.

D) The psychologist observed 91 pairs of people. Since there is a 2 to 1 ratio of corner seating to opposite seating the psychologist should expect 61 pairs to be corner seating, and 30 pairs to be opposite seating. 90/3= 30. 30*30+1=61 (corner seating), 30 (opposite seating) = 2 to 1 ratio. We add extra '1' to the corner seating because this is an estimate and corner seating is more likely.

E) The psychologist's data doesn't support the claim that people prefer corner to opposite seating while interacting because we do not know if people preferred to sit corner-to-corner of if they sat at random.

F) The data doesn't support the claim that people prefer corner seating to avoid eye contact- that is an assumption made by the psychologist.

We need to be very careful when we are reading and analyzing data. Many times when people are conducting experiments they are trying to convince us of an idea that isn't necessarily true, just like the problem discussed above.

IDEA 2- Discussed Circle Graphs: To find the central angle to make a circle graph accurate you have to find the percent of each group in the circle. EXAMPLE: (Page 21 Exploration)

9 Brown eyed people in class/ 20 students in the class= 45 % of class has brown eyes of class has brown eyes
 * Then we do** 360 **(# of degrees in cirlce) /** 45% **(% of eye color we're investigating)=** Angle **used on the circle to have accurate graph.**


 * __Eye Color % Angle__**


 * Blue 20 72 degrees**
 * Brown 45 162 degrees**
 * Green 25 90 degrees**
 * Other 10 36 degrees**


 * Note: A circle graph emphasizes a part to whole relationship and compares the part to the whole.**

IDEA 3- Discussed sorting on outcomes of variables:
 * Ways to sort on outcomes of variables:**
 * -"blobs"- as used with M&M data**
 * -size: small, medium, large**
 * -color**

IDEA 4- How to organize a display:
 * Ways to organizing displays:**
 * -Table**
 * -Put in colums/rows vs. Put in 'blobs'**
 * -Real Graphs: Using objects under investigation. (example: M&Ms)**
 * -Picture Graphs: 1 picture represents a count of 1. You must be able to "count & point" in a picture graph. (Example: page 15)**
 * -Bar graph: Read "up & over". (example: Stacked bar graph= page 18. Side-by-side (Multiple)= page 17. Back-to-back= page 17)**

HOMEWORK:
 * 1) Complete pp 35-42, #1-8**


 * 2) Make stat process visual (We turned these in on Monday 9/20. Change, fix, and add to it if you need to and bring back for Monday 9/27. If you think yours is a good representation, no need to make changes!)**


 * 3) Vote on 4 Survey questions: 2 Survey Questions from each set-** Due Friday at noon

NOTE: QUIZ 1- Wednesday 9/29 : Over covered material


 * BV**


 * Edited by Kristen Huff**

<span style="color: #800080; display: block; font-family: 'Comic Sans MS',cursive; font-size: 130%;">Session 4 September 20th
 * Types of data-**

-measurement- if you can find a partial unit (ex: shoe size, height, arm span)
 * Numerical data- data that is represented by a number

__-count- only whole numbers (ex: number of people in family)__

Quality of survey questions-


 * Is the question going to answer our first question?
 * Some survey questions give us different types of data.
 * When choosing your survey questions, you have to critique them to see if they give you the right type of data.

Types of Sampling- __(pg 10 #2)__

• Voluntary response __ asking for data from those who volunteer to give it __
 * Systematic sampling__- there is a structure to collecting data__
 * Random sampling__- no structure ; there is a structure to make sure the sampling is random __
 * Convenience sampling__- structure that is easiest for statistician__ __ ; sampling from those convenient to the data gatherer __

Session Three 9/15The following mathematical ideas were the focus of today's class:
 * What is a variable?
 * Class Definition: A characteristic that has numerical or categorical data/outcomes.
 * The outcomes determine whether an answer is categorical variable or numerical variable.
 * Examples of numerical variable:
 * Age (unless it is an age-band), Height, number of people in your family, etc.
 * Examples of categorical variable:
 * Where someone goes to school, Male/Female, Eye Color, etc.
 * Difference between frequency and the data.
 * Are you male or female?
 * M, M, F, F, F, F, M, F, F, F.
 * This is data because it is answering the first question whether they are male or female
 * How many males and how many females are in this class based on the data?
 * There are 3 Males and 7 Females.
 * The frequency is how often the outcome occurs in a poll.
 * Research Question:
 * In class we decided upon a question for our problem area.
 * "Does the amount of sleep a full-time WMU undergraduate gets on a regular school week have an affect on their GPA?"
 * We discussed ways to refine the research.
 * We decided we needed to look at who the target population would be.
 * We discussed a few questions in class but we will be revisiting this topic again soon becuase we will be making a questionaire to interview our target population with.
 * We had a few questions about homework to discuss:
 * Pg. 22
 * Pg. 12 #5
 * Pg. 24 #3
 * Pg. 25 #4
 * We only discussed Pg. 22. We will come back to the other three questions at a later date.
 * Homework for 9/15/10
 * If you have not been able to find the time to meet with Dr. Browning please email her the times that would work best for you so you can get something arranged.
 * Do Applications pg. 22-29 #1-4 using the data on pg. 30 when needed if you have not already done this for hw.
 * Read and complete pg 32-34
 * QUIZ session 6, September 27th. This will take place at the end of class and it will cover all material through 9/22/10.

AHE

· Refine survey problem and target population. Begin drafting 1 survey question, either a categorical or numerical question. Due by Friday 12pm on Wiki page · Complete applications pp 22-29, #1-4. Use data on p 30 when needed

Quiz session 6 Sept 27 at end of class on all material through next Wed class.

Session Two 9/13 The first idea that we looked at was "What is Statistics?" We looked at the list that we made as a class, and thought that maybe we could put everything under either A) Study of Comparison B) Collecting, organizing, sorting of data, or C) Telling the data story. The Second Idea that we looked at was the survey questions. We looked at the questions and discussed topics we liked. While we discussed this, we also looked at whether the topic was just that or if it was a question. We also discussed whether a topic was too narrow or not. The third idea started with a bag of M&M's. The first question was asked was "What type of information can we get from a bag of M&M's?" Then we broke up into groups and had to find how many were in each "color" group. Once we found that out, we had to display is how we thought a K-1, 2-3, or 4-5 grader might display the data.

AHE Make graph of M&M Data Vote by Tuesday @ Midnight for the topic you like most. Read Pages 13-20 and do the exploration on pg 21. We have 4 people with Blue Eyes, 9 with Brown, 5 with Green, and 2 people have Other color eyes. Do questions 1-4 on pages 22-29.)