Descriptive Analytics Chapter 6- PDF

Summary

This document is a chapter on descriptive analytics. It discusses various methods of data presentation including textual, tabular, and graphical representations. It also covers different chart types like line charts, bar graphs, pie charts, and pictographs.

Full Transcript

Chapter 6 Descriptive Analytics First Semester, A.Y. 2024-2025 Chapter Objectives Learn different methods of data presentation. Understand the basics of data visualization. Apply data storytelling techniques in business. Create reports and dashboards from data. Methods of...

Chapter 6 Descriptive Analytics First Semester, A.Y. 2024-2025 Chapter Objectives Learn different methods of data presentation. Understand the basics of data visualization. Apply data storytelling techniques in business. Create reports and dashboards from data. Methods of Data Presentation Textual Presentation Tabular Presentation Graphical Presentation Textual Presentation Textual presentation of data incorporates important figures in paragraph or text. Textual Presentation Textual presentation of data incorporates important figures in paragraph or text. Textual Presentation Simplest and most appropriate approach when there are only a few numbers to be presented Gives emphasis to significant figures and comparisons When a large mass of quantitative data is included in a text or paragraph, the presentation becomes almost incomprehensible. Written paragraphs can be tiresome to read especially if the same words are repeated so many times. Tabular Presentation Tabular presentation is the systematic organization of data in rows and columns. Tabular Presentation Tabular presentation is the systematic organization of data in rows and columns. Tabular Presentation More concise than textual presentation Easy to understand Presents data in a greater detail than a graph Facilitates comparisons and analysis of relationship among different categories Graphical Presentation A graph or chart is a device for showing numerical values or relationships in pictorial form. Graphical Presentation A graph or chart is a device for showing numerical values or relationships in pictorial form. Graphical Presentation The graph, as a presenting tool, is the most suitable when we need to show the results of the analysis to the general public. It is easier to see and understand in graphical form rather than in tabular form. It can also be more convincing in supporting our conclusions. We construct graphs not only for presentation purposes but also as an initial step in analysis. It can exhibit possible associations among the variables and can facilitate the comparison of different groups. Types of Graphs/Charts Line Chart Column or Vertical Bar Graph Horizontal Bar Graph Pie Chart Pictograph Statistical Map Line Chart Line charts are most commonly used in presenting historical data. Use it when you want to focus on the movement of the series over time. This chart can also be used to compare the trend of two or more time series data. Ratio of height to width is recommended to be around 2:3 or 3:4. Line Chart Column Chart Use column charts when presenting time series data that focuses on the magnitude of the series instead of the movement. This chart can also be used to show data changes over a period of time among several items. Column Chart Horizontal Bar Chart Use horizontal bar charts when presenting categorical data. Its main purpose is to present magnitude per category. The bars should be arranged according to the length of the bars (either ascending or descending). The “Others” section should be the first or last category in the graph. Horizontal Bar Chart Pie Chart Use pie chart to present categorical data but the focus is to show the components with respect to the total in terms of the percentage distribution. A rule of thumb of five or six categories should be used when constructing a pie chart. If the data has more than five categories, the excess categories should merge into an “Others” category. Arrange these categories according to magnitude. Plot the biggest slice at 12 o’clock, and the succeeding categories clockwise. Pie Chart Pictograph The pictograph is useful to get the attention of the reader while providing an overall picture of the data without presenting the exact figures. However, it still allows the comparison of different categories even if we present the approximate values only. The pictograph is also useful if the goal is for the readers of the chart to easily derive the values of different categories using the chart itself. Pictograph Statistical Map Statistical map is useful in showing data in geographical areas. We also call these types of charts as crosshatched maps or shaded maps. Geographic areas may be barangays, cities, districts, provinces, or countries. The figures in the map can be ratios, rates, percentages, or indices. Statistical Map Advanced Charts/Graphs Combo Chart The combo chart is a combination of more than one type of chart/graph. Use this chart type to highlight different types of information. Use it when you have mixed types of data and/or the range of values in the chart varies widely. Doughnut Chart A doughnut chart has a similar use with a pie chart. However, you can take advantage of the space in the middle of a doughnut chart to include figures. Treemap Use a treemap to compare values across hierarchy levels. Also, use this chart type to show proportions within hierarchical levels as rectangles. Use it when the data is organized hierarchically and has few categories. Sunburst Use a sunburst instead of a treemap to show proportions within hierarchical levels as rings instead of rectangles. Use it also instead of a treemap when your data is organized hierarchically and has many categories. Waterfall Use a waterfall to show cumulative effect of a series of positive and negative values. Use it when you have data representing inflows and outflows such as financial data. Funnel Chart Use a funnel to show progressively smaller stages in a process. Use it when data values show progressively decreasing proportions. What makes a GOOD graph? What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed What makes a GOOD graph? Accurate Clear Simple Professional Well-designed Critique the graph: Critique the graph: Critique the graph: Critique the graph: Critique the graph: Critique the graph: Basics of Data Visualization Why do we visualize data? After analyzing the data, it is important to think how to convey the message coming from the data to the audience. Data visualization transforms the data elements being readable by calculators and computers into instruments to call people to action. It is important to present the data, but it is more sensible to visualize the data in an effective way. Basics of Data Visualization Basics of Data Visualization Steps in Effective Data Visualization Go back to the problem. Choose the right chart/graph type. Create effective visualizations. Steps in Effective Data Visualization Go back to the problem. Go back to the problem to determine the questions that need to be answered. This is the first step to understand what the visualization should try to say. Choose the right chart/graph type. Key Questions to Consider: 1. Who is your audience? 2. What do they want to know? Create effective visualizations. 3. How will you communicate it? Steps in Effective Data Visualization Go back to the problem. Once the purpose for visualization has been determined, it is time to think about what types of chart can help achieve that purpose. Choose the right chart/graph type. The type of chart to be used depends on the purpose of the visualization. Create effective visualizations. Steps in Effective Data Visualization Example: Year Total Sales (in ) 2016 2,345,435 2017 5,643,643 2018 6,546,234 2019 8,268,901 2020 1,211,678 2021 1,567,234 2022 3,209,125 Steps in Effective Data Visualization Common Purposes: Composition: shows the make up of one variables usually in absolute manners or in Go back to the problem. normalized forms. Distribution: visualization methods that display frequency, how data spread out over an interval or is grouped. Choose the right chart/graph type. Relationship: visualization methods that show relationships and connections between the data or show associations between two Create effective visualizations. or more variables. Comparison: visualization methods that help show the differences or similarities between values. Steps in Effective Data Visualization Go back to the problem. Even if the right chart types have been chosen for the data, effective views do not always Choose the right chart/graph type. come naturally. Creating effective visualizations requires effort, intuition, attention to detail, and trial and error. Create effective visualizations. Steps in Effective Data Visualization Go back to the problem. Tips in Making Effective Visualizations: Emphasize the most important data. Orient the charts for legibility. Choose the right chart/graph type. Organize the charts. Avoid too many elements in visualizations. Limit the number of colors and shapes in a single view. Create effective visualizations. Chart Design Basics Emphasis: Pre-Attentive Attributes Chart Design Basics Example: 0.402118 0.447932 0.783153 0.491479 0.348332 0.620136 0.233339 0.563683 0.672747 0.293481 0.551161 0.054075 0.229786 0.621941 0.781805 0.51903 0.628101 0.696429 0.557072 0.580325 0.04495 0.134483 0.158648 0.801352 0.521584 Chart Design Basics Example: 0.402118 0.447932 0.783153 0.491479 0.348332 0.620136 0.233339 0.563683 0.672747 0.293481 0.551161 0.054075 0.229786 0.621941 0.781805 0.51903 0.628101 0.696429 0.557072 0.580325 0.04495 0.134483 0.158648 0.801352 0.521584 Chart Design Basics Example: 0.402118 0.447932 0.783153 0.491479 0.348332 0.620136 0.233339 0.563683 0.672747 0.293481 0.551161 0.054075 0.229786 0.621941 0.781805 0.51903 0.628101 0.696429 0.557072 0.580325 0.04495 0.134483 0.158648 0.801352 0.521584 Chart Design Basics Example: 0.402118 0.447932 0.783153 0.491479 0.348332 0.620136 0.233339 0.563683 0.672747 0.293481 0.551161 0.054075 0.229786 0.621941 0.781805 0.51903 0.628101 0.696429 0.557072 0.580325 0.04495 0.134483 0.158648 0.801352 0.521584 Chart Design Basics Clustering: Gestalt Laws of Visual Perception Chart Design Basics Comparison: Cleveland-McGill Scale Best used for comparison Avoid for comparison Data Storytelling for Business Data Visuals Narrative Data Storytelling for Business Data Visuals Narrative Data Storytelling for Business Data Visuals Narrative Data Storytelling for Business The narrative of a data story, together with the proper wrangling of the data and the informed choice of visualization, will allow you to effectively Narrative present your insights, leaving a lasting impression, and drive your audience to take the necessary business actions. Data Storytelling for Business Insight Generation Best Practices 1. Describe the outlier, trend, or feature. 2. Refer to the greater context which may not feature in the data. Narrative 3. Include a time reference if appropriate. 4. Add value beyond what is asked for. 5. Make your insights as hardworking and as layered as possible. Data Storytelling for Business Foundations of an Effective Data Story 1. Ensure vertical logic – All insights in a page or Narrative slide should be evidenced by the supporting visualization. Data Storytelling for Business Foundations of an Effective Data Story 2. Check for horizontal flow – In the case of staged or multi-page data stories, ensure that it flows from one stage to the next in a meaningful and logical fashion. a. Chronological – Telling a story over time in Narrative chronological order is a good way of ensuring flow. b. Top-Down Approach – Starting with the big picture or total market and progressively becoming more popular. c. Bottom-up Approach – Starting with specific details and progressively becoming more general. Data Storytelling for Business Avoid the Narrative Fallacy Are you certain that the trend you’re seeing is not just a one-off? Take a moment to think of other possible Narrative causes of why the data behaves that way. Remember that correlation does not guarantee causation (Chapter 5). Always remember that your story is just one of many possible interpretations. Data Storytelling for Business Data Visuals Narrative Data Analysis Report A data analysis report narrates the premise of the data analytics project conducted, the business problem and objectives involved, and the methodologies used in the project. This is also one way to present the results of a data analysis and provide recommendations based on data. Data Analysis Report Parts of a Data Analysis Report: 1. Introduction/Background 2. Business Problem and Objectives 3. Scope and Limitations 4. Methodology 5. Results and Discussion 6. Conclusion and Recommendations Dashboards Data dashboards are a summary of different, but related data sets, presented in a way that makes related information easier to understand. Dashboards are a type of data visualization, and often use common visualization tools such as graphs, cards, and tables. Ethics in Data Presentation Business Insights Blog Data ethics encompasses the moral obligations of gathering, protecting, and using personally identifiable information and how it affects individuals. Five principles of business data ethics include ownership, transparency, privacy, intention, and outcomes. Ethics in Data Presentation The Data Privacy Act of 2012 and its Implementing Rules and Regulations is the main basis of the Philippines in ensuring the data privacy of Filipinos. All entities, including business executives and entities, are expected to comply with the law in protecting the data privacy of the people during the execution of business processes and operations. Ethics in Data Presentation Ethics in Data Analysis In the case of an unfavorable analysis result, one should stick with the methodology and not to modify the result to fit the narrative, or the methodology itself to force the favorable outcome. Compromising the results and/or the methodology will never help. Instead, a post-hoc analysis can be done to further study the unfavorable outcome. A post-hoc analysis involves looking at the data after a study has been concluded, and trying to find patterns and conclusions that were not primary objectives of the said study. Ethics in Data Presentation Ethics in Data Reporting and Visualization When reporting figures and creating visualizations, ensure that the formats will highlight what the data are truthfully saying. There are instances where reports and visualizations are purposely distorted and/or modified to adjust to the narrative that the reporter wants to convey. Ethics in Data Presentation Ethics in Data Reporting and Visualization Total COVID-19 Cases in the Philippines Recovery Rate 98.30% 4,053,939 Fatality Rate 1.62% 66,755 Ethics in Data Presentation Ethics in Data Reporting and Visualization Ethics in Data Presentation Ethics in Data Reporting and Visualization End of Chapter 6 Descriptive Analytics First Semester, A.Y. 2024-2025

Use Quizgecko on...
Browser
Browser