Microsoft Excel: Data Analysis & Graphing College of Engineering Engineering Education Innovation Center
Objectives Use relative, absolute, and mixed cell referencing Identify the types of graphs and their proper formatting Create and format charts in Excel Create and interpret trend lines Rev: 20141009, EPC Excel Data Analysis 2
Review - Cell Reference: Relative Addressing Relative Cell Address refer to the cell in the same relative position Ex: B12 relative reference to value in cell B12, use fill handle to drag to cell C17 Step 1 Step 2 Step 3 Rev: 20141009, EPC Excel Data Analysis 3
Review - Cell Reference: Absolute Addressing Absolute Cell Address always refer to the same cell Ex: $B$12 absolute reference to value in cell B12, use fill handle to drag to cell C17 Step 1 Step 2 Step 3 Rev: 20141009, EPC Excel Data Analysis 4
Graphing A graph is often the best way to present data so that it is easily understood. There are several types of graphs available to present data. These graphs serve different purposes. Rev: 20141009, EPC Excel Data Analysis 5
Graph Types: Categorical Data Column Chart Pie Chart Bar Range Chart Rev: 20141009, EPC Excel Data Analysis 6
Another Graph Type: Scatter Plot Rev: 20141009, EPC Excel Data Analysis 7
Categorical Charts vs. Scatter Plots Categorical charts have labels on the x- axis Scatter plots have values on the x-axis Rev: 20141009, EPC Excel Data Analysis 8
Example: Empirical Data What happened? Missing Data! BCM 6/20/12 Line Chart (Categorical) BCM 6/20/12 Scatter Plot Rev: 20141009, EPC Excel Data Analysis 9
Interpolating & Predicting Data Excel graphing allows you to graph data and then create an equation that best fits that data. You can choose the type of equation for Excel to use, such as linear, exponential, logarithmic, etc. With a best-fit equation you may be able to interpolate or predict data not included in the worksheet. This must be done with caution. Note: You must use X-Y Scatter Plots Rev: 20141009, EPC Excel Data Analysis 10
Example Predicting Values Year Births Deaths 2002 3,167,788 1,909,440 2003 3,612,258 1,989,841 2004 3,680,537 1,974,797 2005 3,669,141 2,039,369 2006 3,756,547 2,105,361 2007 3,909,510 2,167,999 2008 4,158,212 2,148,463 2009 4,110,907 2,169,518 2010 4,065,014 2,175,613 2011 4,000,240 2,268,553 2012 3,952,767 2,278,994 Question: How many births and deaths would we expect in 2016? Rev: 20141009, EPC Excel Data Analysis 11
On your own, plot the data Rev: 20141009, EPC Excel Data Analysis 12
Note Title Even Increments Different Markers Legend Rev: 20141009, EPC Excel Data Analysis 13
What is missing? Axis Labels (with units!) Author, Date Rev: 20141009, EPC Excel Data Analysis 14
Population (People) That s a lot of zeroes 4,500,000 US Population Statistics 4,000,000 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 Births Deaths 1,000,000 500,000 0 2000 2002 2004 2006 2008 2010 2012 2014 Year BCM 5/11/12 Rev: 20141009, EPC Excel Data Analysis 15
Display Ordinate in Thousands Rev: 20141009, EPC Excel Data Analysis 16
Population (Thousands of People) Properly Formatted Graph 4,500 US Population Statistics 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 Births Deaths 0 2000 2002 2004 2006 2008 2010 2012 2014 Year BCM 5/11/12 Rev: 20141009, EPC Excel Data Analysis 17
Adding a Linear Trendline 1. Right click on any data point and select the Add Trendline option. 2. Select Linear as the regression type (note that there are other options available). 3. Select this option to show the equation of the line on the chart Rev: 20141009, EPC Excel Data Analysis 18
Population (Thousands of People) Trendline for Births 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 US Population Statistics y = 71959x - 1E+08 0 2000 2002 2004 2006 2008 2010 2012 2014 Year Births Deaths Linear (Births) Note: You may need to re-format the trendline equation to improve resolution. You may also need to move the equation. Rev: 20141009, EPC Excel Data Analysis 19
Formatting Trendline Label 1. Select trendline label (left-click) 2. Right-click on edge of label box 3. Select Format Trendline Label Rev: 20141009, EPC Excel Data Analysis 20
Formatting Trendline Label Select Number with 0 decimal places Rev: 20141009, EPC Excel Data Analysis 21
Population (Thousands of People) Results Using these equations, we predict for 2016: Births = 4,474,063 Deaths = 2,428,399 Do you see any problem with these predictions? What can we do to improve our estimate? US Population Statistics 4,500 4,000 y = 71,959x - 140,595,281 3,500 3,000 2,500 y = 35,168x - 68,470,289 2,000 1,500 Births 1,000 Deaths Linear (Births) 500 Linear (Deaths) 0 2000 2002 2004 2006 2008 2010 2012 2014 Year BCM 5/11/12 Rev: 20141009, EPC Excel Data Analysis 22
NUMBER OF FATALITIES NUMBER OF FATALITIES Interpreting Charts Were traffic fatalities stable? Or were they cyclical? TRAFFIC FATALITIES IN OHIO 1992-2002 TRAFFIC FATALITIES IN OHIO 1992-2002 1600 1400 1200 1000 800 600 400 200 0 92 93 94 95 96 97 98 99 00 01 02 1500 1450 1400 1350 92 93 94 95 96 97 98 99 00 01 02 YEAR YEAR Rev: 20141009, EPC Excel Data Analysis 23
FATALITIES PER 100 MILLION MILES Interpreting Charts Maybe we should plot the number of fatalities per 100 million miles driven. Which of the three is correct? 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 TRAFFIC FATALITIES IN OHIO 1992-2002 92 93 94 95 96 97 98 99 00 01 02 YEAR Rev: 20141009, EPC Excel Data Analysis 24
Formatting a Plot: One Data Set Title Axis Labels Units! Even increments Author/Date BCM 6/20/12 Rev: 20141009, EPC Excel Data Analysis 25
Add Another Data Set Data markers different from each other Legend BCM 6/20/12 Rev: 20141009, EPC Excel Data Analysis 26
Questions? Rev: 20141009, EPC Excel Data Analysis 27