Statistics Detective
Goal: Investigate real-world data and interpret key features.
Goal: Utilize technology to adapt or build and use mathematical formulas that include cell referencing to answer real-world questions and interpret results

Instructions: I selected a person in the class by random number generator and built clues from there. Use these clues in order to identify the "odd person out" in our class data set. To mark off eliminated people on the computer, be sure that View/Ribbon is checked and that you can see the paint can under Home. You can then use the fill-in feature to eliminate suspects by clicking on the row number and then the paint can. Once someone violates one clue, they are out of the running. In this way, your document will show suspects eliminated using the clues as painted rows, where each color represents a different clue.
The suspect...
1. supports the statement that says that the average person is better able to remember real words than nonsense words.
Eliminate people who contradict the clue and remember nonsense words at least as well as real words. Choose a color for this clue and highlight any eliminated rows in that color. You'll eliminate 5 people.

2. has a golden mean of (1+sqrt(5))/2 ~ 1.618, within a .200 margin of error, embedded in the forearm to hand ratio.
-What are lower and upper boundaries for the interval rounded to 3 decimal places (Hint: use 1.618 and the margin of error)?

lower boundary: _______________                                         upper boundary: _______________

-Which column letter represents forearm? _______________                        hand? _______________
-If we wanted to look at the ratio of columns J and K in Excel for person 2, we would use the Excel cell referencing command =J2/K2. Using Excel cell referencing, what is the Excel formula for forearm/hand for the first person in row 2 (don't forget the equal sign and use the columns you listed above)?

_______________ Type this into K2 and hit return. Go back to that box, go to the bottom right until you see the symbol change, click and fill down the column for the forearm to hand ratios.
-Eliminate those remaining who contradict the statement and have a ratio outside the interval since the suspect is within. Choose a clue 2 color and highlight any new eliminations in Excel. Eliminate anyone (who hasn't already been eliminated) less than the lower boundary. Eliminate anyone greater than the upper boundary.

3. falls within the middle 50% of the class on the boxplot of the distance from home.
-First, as a review, compute in Excel the 5-number summary in some empty boxes, using the following commands, and then roughly sketch the boxplot by-hand by creating a reasonable scale for the axis. Be sure that your commands match these:
Excel work:                                                                                By-hand sketch:
Q4 or maximum: =quartile(c2:c32,4) _______________
Q3: =quartile(c2:c32,3) _______________
Q2 or median: =quartile(c2:c32,2) _______________
Q1: =quartile(c2:c32,1) _______________
Q0 or minimum: =quartile(c2:c32,0) _______________

-Now eliminate rows who contradict the statement and who fall strictly outside the box borders - i.e.
eliminate those who are strictly below Q1 or strictly above Q3. Choose a different color and highlight any new eliminations in Excel.
4. hits the 2nd quartile (the median) "on the head" for the number of competitors of the family bathroom.

-Which column does the clue apply to? _______________

-Use Excel cell referencing. What is the Excel command for Q2 or the median? (don't forget the equal sign) _______________

-After you execute the command in Excel, what is Q2 of that column: ______
-Next, eliminate those who are different than Q2
-You should have 3 people remaining at this point. Write down the number of each row that is left.

_______         _______         _______

5. has a y-value within 1cm of the best fit line for "does armspan predict height."
-What is the equation of the best fit line from #6 on the correlation lab (include all the (mis)measurements)? If you don't have your lab, you'll need to reconstruct this in Excel and you can review the instructions for random number and armspan and apply them to armspan and height.

y=_____________________________
-Compute the predicted line value of height by plugging in the armspan (the x value) of ONLY your remaining suspects into the equation of the line to solve for the y value height. Show work. Also write down the actual height of each person from Excel.

Predicted height y=_____________________________ = ______________. Actual height=_______

Predicted height y=_____________________________ = ______________. Actual height=_______

Predicted height y=_____________________________ = ______________. Actual height=_______

-Compare the predicted line height value with the actual height and eliminate the people whose difference is greater than 1cm

-Which row is the "winner" _______
-Does the "winner" drag the line up or down? I.e. is their actual height value above or below the predicted line value?

Circle one:             up             down             exactly on the line