The assignment requirments are in the last slide of the Week 8 lecture pdf.
Data SiteTotal Patients ScreenedMRI Eligible Patients (High risk)High Risk Patients who were called at least once by the NavigatorPatients who were scheduled for MRIPatients who received their MRI Week 1268737097 Week 2274453285 Week 32994200 Week 42956964129 Week 528868681914 Week 62389810 Week 729733291410 Week 8225635153 Week 92577500 Week 10291282586 Week 11253262432 Week 1227061561311 Week 13270332686 Week 1430043301311 Week 15292433087 Week 162245237119 Week 17286413876 Week 182904300 Week 19276171664 Week 202963730149 Week 212813200 Week 22262473396 Week 23295604297 Week 2422640281210 Week 2523868632316 Week 26293121021 Week 2722647391913 Week 2828171531914 Week 292784635129 Week 302911000 Week 312170000 Week 32254121054 Week 3329854442114 Week 342943432108 Week 3520835321510 Week 36207171321 Week 3726354471513 Week 3828769521814 Opt-in Rates (2) HunterdonOSF St. Francis Consent to Contact RateConsent to Contact RateEligible Patients ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!ERROR:#REF!ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 1ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 2ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 3ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 4ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 5ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 6ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 7ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 8ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 9ERROR:#REF! ERROR:#REF!ERROR:#REF!Week 10ERROR:#REF! Week 11ERROR:#REF! Week 12ERROR:#REF! Week 13ERROR:#REF! Week 14ERROR:#REF! Week 15ERROR:#REF! PowerPoint Presentation MGT599 – Capstone Week 8 8/3/2021 Agenda Week 7 Review Data Visualization Overview Design Principles Charts Storytelling • City Hospital MRI Case Week 7 Review I am running an AR(4) model looking at past yearly historical stock price on predicting future stock prices. What is the time lag in this model? What type of variable is the dependent variable in logistic regression? Why can’t we use a linear regression instead? The format of a linear regression is: y = a + b1x1 + b2x2 + b3x3 +…. What is the format of a logistic regression? Logistic Regression Data Visualization Overview Data Visualization Overview Data visualization is the visual representation and presentation of data to enhance understanding. This allows for: • better ability to grasp difficult concepts • identify new patterns • more efficiently take away key messages. Data visualization is an essential aspect of effectively communicating data and findings. Data Visualization Overview Key benefits include: • Comprehend information quickly • Identify relationships and patterns • Pinpoint outliers and potential data quality issues • We already looked at residual plots, but it’s great practice to just look at the data to see if there is anything weird • Communicate the story to others Design Principles Google’s Six Principles for Chart Design https://medium.com/google-design/redefining-data-visualization-at-google- 9bdcf2e447c6 https://medium.com/google-design/redefining-data-visualization-at-google-9bdcf2e447c6 Balance Useful when thinking of dashboards Symmetrical • Each side of the dashboard mimics the other. This type of balance feels elegant, formal, and conservative. Asymmetrical • Both sides are unique but carry a similar visual weight. “Heavier” elements jump out, while the “lighter” ones recede. This type of balance feels casual, free, and energetic. Radial • A central object anchors the dashboard, with additional objects placed around it. This type of balance draws your eyes inwards to the center focal point. Symmetric Dashboard Asymmetric Dashboard Radial Dashboard Emphasis Draw users attention to important information • Color, size, negative space, contrast • I.e. People read left to right, top to bottom, so top left is the first place Charts Simple Comparisons Bars and Columns are great options Make sure the axes make sense and are not misleading https://paldhous.github.io/ucb/2016/dataviz/week2.html https://paldhous.github.io/ucb/2016/dataviz/week2.html Comparisons over time Multiple Comparisons Composition: Parts of a Whole Composition: Parts of a Whole Composition Change Over Time Composition Change Over Time Network Graphs Heat Maps Heat Maps Be aware of scale and size 3D can be hard to read Storytelling Storytelling Storytelling Principles City Imaging City Imaging City Imaging is the premier imaging center in Citysville, USA. The center focuses on breast imaging and provides mammography, MRI, and Ultrasound services. John Doe is the director of City Imaging. Currently, they do 15,000 mammograms a year but only about 100 or so MRIs. The MRI machine is vastly underused and he is looking to boost the utilization rate of the MRIs. He wants to target 500 MRIs/year. To fix this, he bought a new piece of software called “MRI-Now”. This software consists of: • Short survey for the patient to take • Risk calculations to determine if a patient meets MRI Criteria • A patient CRM for nurse navigators to contact patients to schedule them for an MRI City Imaging - Workflow Here is the current workflow: • Patient checks in for their mammogram • They take the MRI-Now Screening Survey • They are flagged as High Risk if they meet criteria for MRI screening • They are given some information that tells them they are eligible for an MRI, what that means, and why getting an MRI is important • A couple of days later, a nurse navigator calls the high risk patients and schedules them for an MRI (usually done a month later) • The patient comes back in a month and gets the MRI City Imaging - Objections There have been a few objections that have come up from the staff and patients. They include: • The navigators occasionally get backed up with their patient caseload and are not able to call all patients – sometimes patients are not at home and they need to play phone tag and leave messages back and forth – which can be time consuming • Patients sometimes don’t understand why they are at high risk and why an MRI is necessary – as a result, they might not show up or even indicate that they want one • Other concerns include insurance coverage (which an MRI is covered if they are at high risk based on guidelines) and transportation issues Business Review It is now 9 months since the program is started and City Imaging is coming up for renewal. You have a business review scheduled with Mr. Doe to go over performance and convince him to renew. Attached is the data to date for the program. Please prepare a 5-10 slide presentation for Mr. Doe for this business review. Assignment Main Assignment: Step 1: Review and analyze the data Step 2: Prepare a 5-10 slide presentation for Mr. Doe for this business review. Bonus (1 point bonus): Record a video of you giving this presentation to Mr. Doe (me). Record yourself over zoom and send me the link! Submission: The slides in PDF format and the zoom video link (optional). Microsoft Word - DataVizChecklist_v04.docx Data Visualization Checklist This checklist is meant to be used as a guide for the development of high impact data visualizations. Rate each aspect of the data visualization by circling the most appropriate number, where 2 points means the guideline was fully met, 1 means it was partially met, and 0 means it was not met at all. n/a should not be used frequently, but reserved for when the guideline truly does not apply. For example, a pie chart has no axes lines or tick marks to rate. Refer to the Data Visualization Anatomy Chart on the last page for guidance on vocabulary. Guideline Rating Text Graphs don't contain much text, so existing text must encapsulate your message and pack a punch. 6-‐12 word descriptive title is left-‐justified in upper left corner 2 1 0 n/a Short titles enable readers to comprehend takeaway messages even while quickly skimming the graph. Rather than a generic phrase, use a descriptive sentence that encapsulates the graph’s finding or “so what?” Western cultures start reading in the upper left, so locate the title there. Subtitle and/or annotations provide additional information 2 1 0 n/a Subtitles and annotations (call-‐out text within the graph) can add explanatory and interpretive power to a graph. Use them to answer questions a viewer might have or to highlight one or two data points. Text size is hierarchical and readable 2 1 0 n/a Titles are in a larger size than subtitles or annotations, which are larger than labels, which are larger than axis labels, which are larger than source information. The smallest text -‐ axis labels -‐ are at least 9 point font size on paper, at least 20 on screen. Text is horizontal 2 1 0 n/a Titles, subtitles, annotations, and data labels are horizontal (not vertical or diagonal). Line labels and axis labels can deviate from this rule and still receive full points. Data are labeled directly 2 1 0 n/a Position data labels near the data rather than in a separate legend (e.g., on top of or next to bars or pie slices, and next to lines in line charts). Eliminate/embed legends when possible because eye movement back and forth between the legend and the data can interrupt the brain’s attempts to interpret the graph. Labels are used sparingly 2 1 0 n/a Focus attention by removing the redundancy. For example, in line charts, label every other year on an axis. by Stephanie Evergreen & Ann K. Emery May 2014 Arrangement Improper arrangement of graph elements can confuse readers at best and mislead viewer at worst. Thoughtful arrangement makes a data visualization easier for a viewer to interpret. Proportions are accurate 2 1 0 n/a A viewer should be able to take a ruler to measure the length or area of the graph and find that it matches the relationship in the underlying data. Data are intentionally ordered 2 1 0 n/a Data should be displayed in an order that makes logical sense to the viewer. Data may be ordered by frequency counts (e.g., from greatest to least for nominal categories), by groupings or bins (e.g., histograms), by time period (e.g., line charts), alphabetically, etc. Axis intervals are equidistant 2 1 0 n/a The spaces between axis intervals should be the same unit, even if every axis interval isn’t labeled. Graph is two-‐dimensional 2 1 0 n/a Avoid three-‐dimensional displays, bevels, and