1) You will be working exclusively with data from the dataset provided below (no need
for further data-gathering):
•
Excel version for download: F22 data for Lab 4 GRAPHAPALOOZA1.xlsx Download F22 data for Lab 4 GRAPHAPALOOZA-1.xlsx
Google sheet version for
view/download: https://docs.google.com/spreadsheets/d/17rTpcwNiXoshj862AzGRh38cVLIDHB/edit?usp=sharing&ouid=101248716521260025727&rtpof=true&sd
=true Links to an external site.
Please read the “Metadata” listed as the first tab in those sheets for notes about the
dataset and tips for how to work with them.
Use as much data as you can. Especially for drawing up scatterplots, it’s just
much more likely that we can discern a trend if there are more points in the
graph.
On a particular graph, if it’s cumbersome to work with so many points (like if
you had to re-categorize everybody’s response for a histogram or
something like that), you can consider just using the 12 people from the list
with names. But generally I think it’s best to try using every bit of data.
At least one of your graphs must use all the data from the 106-person
dataset.
You’re free to work with the parameters that interest you (provided that you satisfy the
requirements laid out below).
2) Here are the “rules” about what types graphs you can make. You’re going to make
four graphs in total:
1.
1. A graph of only categorical data.
? These often take the form of pie charts, infographics, or
histograms.
? Histograms sort of ride the line…. If you’re breaking up a
numerical parameter into categories to look at its
distribution, like graphing the distribution of human
height by binning people into “4-5 ft,” “5-6 ft,” etc. we
can call that a categorical.
2. A scatterplot of two quantitative parameters.
? Scatterplots (also called dot plots or xy scatterplots) are
often the way go when plotting two numerical
parameters (continuous or count data). Scatterplots allow
you to assess the shape of relationship between such
parameters.
? In this lab, I’d like you try this easy to use spreadsheetbased tool: Regression tool for HES.xls Download
Regression tool for HES.xls(at least, try it that way first).
The spreadsheet generates graphs two at a time… so
when you paste in data like latitude vs. longitude, you’ll
see it fit to a straight-line linear trend line or a quadratic
one. choose whichever graph you like better (or present
both), but recognize it will only count as one of your
graphs for this lab.
3. A graph of both categorical and quantitative data.
? These can be as simple or as complex as you want to
make them.
? One random example that I could imagine making from
data outside our dataset would be: a bar plot with 50
bars each representing average household income for
each state in the US (and maybe even some bars colored
in such a way that they highlight the different regions of
the US states!) you can often use color coding or
grouping to make interesting comparisons in bar graphs
or line graphs.
4. The fourth graph is whatever you want it to be! (as long as it is a
graph or infographic.)
? Creative graphing approaches welcome!
? no memes please!
5. Whenever possible, use the “whole” 106-person dataset. Using all
available data is cumbersome, but it is almost always the right move
in data exploration like this because we are more likely to pick up
on trends.
3) There will be more of a write-up for this one, which resembles a very very short
scientific paper (details below).
How to do the write-up:
The Introduction
For each of your four graphs, use bullet points or numbering to lay out:
•
·hy you thought the relationships between your chosen parameters were
potentially interesting (to you or in general)
the trend you expected before graphing them
your reason(s) for expecting those trends
Methods
Everyone’s methods will be pretty much the same for this, so you don’t have to describe
where you got the data and so forth. Instead, use this section to briefly note what you
did in terms of processing the data that you graphed. The following are some questions
you might consider addressing/answering in this methods section:
•
•
Did you find calculate averages for numeric data?
Did you clean up or simplify peoples’ responses for categorical or write-in
parameters?
Did you break a parameter down by “binning” the values so you could set up a
histogram? (if so, how did you choose the cutoffs for those “bins”?
Did you exclude certain datapoints because they seemed incorrect or
extreme? (or, did you try a graph with and without a datapoint that seemed
extreme?)
Results
Here you need to paste your 4 graphs into the write-up. Please verify that titles/axis
titles and other info come through correctly in your write up. One way to avoid shifts in
formatting is to screenshot graphs or paste them as images.
Underneath your graphs please write a simple name like “Figure 1,” “Figure 2,” etc. and a
brief caption. Something like this example works great:
Figure 1. Scatterplot of human height and weight collected from a 1996 Kansas
statewide survey. The quadratic trend line shown here had a fit of R2=0.89 and p
Purchase answer to see full
attachment